Translation as a human skill

From predisposition to expertise

 Bogusława Whyatt

Translation as a human skill

From predisposition to expertise

Wydział Anglistyki
ABSTRACT: Bogusława Whyatt, Translation as a human skill. From predisposition to
expertise. Faculty of English. Adam Mickiewicz University. Poznań 2012. Pp. 461.
ISBN???????. Text in English

The book explores translation as a human skill in its evolutionary perspective from the
predisposition to translate to translation expertise. By assuming that the human mind is
intrinsically a translating mind all people who know two languages are able to translate
but only some develop their natural ability into a more refined skill, fewer choose to
acquire translation competence and few attain the level of expertise. Starting with a
thorough analysis of the bilingual foundations on which translation as a human skill is
built the natural ability is analyzed and followed by an up-to-date account of translation
as a trained skill with the underlying translation competence. To account for the
developmental nature of translation as a skill a suggestion is made that the acquisition of
translation expertise can be seen as a process of learning to integrate knowledge for the
purpose of translating. While natural translators integrate only their bilingual knowledge
professional translators build a Knowledge Integration Network (KIN) in which all the
knowledge relevant for a task at hand is integrated and ready for use. The theoretical
assumptions are put to an empirical test with research tools including a battery of
questionnaires and Translog, a computer software program which allows to analyze the
translation process without compromising its ecological validity. The subjects include
translators at various stages on the developmental continuum. It is hoped that the
conclusions and implications will raise awareness of the developmental nature of
translation as a human skill and thus challenge the common misconceptions.

KEY WORDS: translation, ability, skill, competence, expertise, knowledge integration,

developmental continuum

Bogusława Whyatt, Faculty of English, Adam Mickiewicz University. Al. Niepodległości

4, 61-678 Poznań, Poland; email: bwhyatt@ifa.amu.edu.pl

© Copyright by Bogusława Whyatt, Poznań 2012

Cover design/Projekt okładki

ISBN XXX
Contents
Acknowledgements ................................................................................. 9
Introduction .......................................................................................... 11

Chapter 1: Establishing common grounds .................................... 17

1.1. Human translating mind .......................................................... 17
1.2. Translation as a widespread phenomenon ............................... 20
1.3. Translation proper – essential distinctions .............................. 22
1.4. Translation as a human skill – common misconceptions ........ 24
1.5. Translation as a predisposition, ability, skill, competence
and expertise ............................................................................ 25
1.6. The translator .......................................................................... 30
1.6.1. Need for a developmental perspective .......................... 31
1.6.2. The role of translators ................................................... 34
1.7. The scope of systematic research in translation expertise
development ............................................................................ 37
1.7.1. Linguistics ..................................................................... 37
1.7.2. Translation Studies ........................................................ 40
1.7.3. Psycholinguistics and cognitive psychology ................. 43
1.8. Conclusions ............................................................................. 47

Chapter 2: Bilingual foundations of translation ability ............... 49

2.1. Bilingual foundations and Translation Studies ....................... 49
2.2. The nature of human language and the knowledge
of two (or more) languages ..................................................... 52
2.3. A bilingual person ................................................................... 58
2.3.1. Natural bilinguals and translation ability ...................... 60
2.3.2. Bilingual competence of L2 learners as potential
translators ...................................................................... 63
2.4. The interaction of two languages in one mind – an L2
learner’s perspective ................................................................ 67
2.4.1. Cross-linguistic interference ......................................... 68
6 Contents

2.4.2. Three perspectives on bilingual coexistence in a

single mind ................................................................... 69
2.5. The notion of language control ............................................... 74
2.6. Bilingual memory .................................................................... 82
2.6.1. The hierarchical structure of the bilingual lexicon ........ 83
2.6.2. The revised hierarchical model (RHM) ........................ 87
2.6.3. Deficiencies and limitations of the RHM ...................... 88
2.6.4. Bilingual memory and translation ................................. 93
2.7. Cognitive effects of bilingualism ............................................ 98
2.8. Conclusions ........................................................................... 102

Chapter 3: Translation as an untrained ability .......................... 107

3.1. Dispute over bilingual knowledge and translation capacity .. 107
3.2. Translation as a natural ability of bilinguals ......................... 109
3.2.1. Studies in natural translation ....................................... 110
3.2.2. Natural translators are translators ................................ 116
3.3. L2 learners as natural translators ........................................... 119
3.3.1. Translation as the fifth skill of L2 learners ................. 120
3.3.2. Studies in L2 translation .............................................. 126
3.4. Developmental shift in translation as a bilingual ability
of L2 learners – TAP studies ................................................. 130
3.4.1. Intercultural competence ............................................. 133
3.4.2. Metalinguistic awareness ............................................ 139
3.4.3. Re-organization of the bilingual memory ................... 144
3.5. Conclusions ........................................................................... 148

Chapter 4: Developing translation competence .......................... 153

4.1. Translation competence ......................................................... 153
4.1.1. The additive perspective on TC .................................. 154
4.1.2. The componential perspective on TC ......................... 157
4.1.3. The holistic perspective on TC .................................... 163
4.2. Acquiring translation competence .......................................... 167
4.2.1. Some suggestions on how TC can be acquired ........... 167
4.2.2. The correlation between TC and translation
experience ................................................................... 172
4.3. TC acquisition and the “pedagogical gap” ............................. 177
4.3.1. Focus on the translation process ................................. 177
4.3.2. Learner-centred approaches to translation training ..... 181
4.4. Different routes to translation competence ............................ 185
 Contents 7

4.5. The professional self and translation as a profession ............. 188

4.6. Conclusions ........................................................................... 195

Chapter 5: Becoming a professional translator:

A proposal of a Knowledge Integration
Network in the development of translation
as a professional skill .................................................. 199
5.1. Creativity in translation ......................................................... 200
5.2. Theoretical foundations of KIN ............................................ 204
5.2.1. The human mind as a synchronized system ................. 208
5.2.2. Limited capacity of the human mind .......................... 210
5.2.3. Meta-cognitive supervision and goal orientedness ..... 213
5.3. Developmental nature of the ability to integrate
knowledge ............................................................................. 217
5.3.1. Approaching the SL text ............................................. 218
5.3.2. Approaching meaning ................................................. 222
5.3.3. Approaching external resources to aid translation ...... 223
5.3.4. Approaching the translation process: self-monitoring
and meta-cognitive supervision .................................. 228
5.3.5. Approaching revision .................................................. 232
5.4. The translator as an expert learner ........................................ 237
5.4.1. The translator as a Systems Intelligent person ............ 238
5.4.2. The translator as a self-confident expert ..................... 243
5.5. Undervalued affective factors in TC acquisition ................... 248
5.6. Conclusions ........................................................................... 251

Chapter 6: Empirical investigation into the development of a

Knowledge Integration Network in translators –
part one ....................................................................... 257
6.1. The purpose of the questionnaire study ................................. 257
6.2. Discussion of the results ........................................................ 262
6.2.1. Translation as a profession – assumptions and facts ... 262
6.2.2. Ideas about translation as an activity ........................... 272
6.2.3. Bilingual foundations of translation
as a human skill .......................................................... 278
6.2.4. A translating professional at work .............................. 294
6.2.5. The translator as an individual .................................... 316
6.3. Conclusions ........................................................................... 332
8 Contents

Chapter 7: Empirical investigation into the development of a

Knowledge Integration Network in translators –
part two ....................................................................... 339
7.1. Research method ................................................................... 339
7.2. Study 1 – objectives, participants, procedure ........................ 346
7.3. Study 1 – data analysis .......................................................... 349
7.3.1. Duration ....................................................................... 350
7.3.2. Duration and dictionary use ........................................ 354
7.3.3. Duration and typing speed ........................................... 359
7.3.4. Distribution of the remaining time in the translation
process ........................................................................ 365
7.4. Study 2 – objectives, participants, procedure ........................ 381
7.5. Study 2 – data analysis .......................................................... 382
7.5.1. KIN as a single variable .............................................. 383
7.5.2. Duration and other text production features ................ 384
7.6. Conclusions ........................................................................... 391

Conclusions and Implications ........................................................... 399

References ........................................................................................... 407
Appendix I ........................................................................................... 439
Appendix II ......................................................................................... 443
Appendix III ....................................................................................... 453

(Streszczenie) ....................................................................................... 459

Acknowledgements

Many people have contributed and helped me in various ways, both di-
rectly and indirectly, to plan, carry out and finalize my post-doctoral pro-
ject in the form of this book. I owe my research interest to my students at
Adam Mickiewicz University whom I have taught and who have inspired
me to investigate the human ability to translate. I would like to thank
Prof. Katarzyna Dziubalska-Kołaczyk, head of the School of English,
Adam Mickiewicz University for creating a friendly working environ-
ment in which teaching and doing research are both important and stimu-
lating for each other.
The ideas which I share in this book are an outcome of reading the
available literature and my own experience as a scholar, teacher and a
translator. Many authors and scholars whose ideas I found inspiring are
acknowledged by way of references. Some of them, for example Prof.
Gideon Toury and Prof. Maria Tymoczko, I was fortunate enough to have
met at the 5th EST congress in Ljubliana. Their comments and words of
encouragement following my paper, which was the starting point of the
project described in this book, had a guiding power.
The empirical part of the book was possible thanks to all the people
who agreed to participate in the project. I would like to express my grati-
tude to all the practicing translators who gave their valuable time and
shared their experience for the purpose of this research. I am also grateful
to all the students at Adam Mickiewicz University who acted as subjects
in the study described in this book.
Conducting the research would not have been possible without the gen-
erosity of Prof. Arnt Lykke Jakobsen from the Copenhagen Business
School who allowed me to use Translog for research purposes. I am also
grateful to Katarzyna Bejster for her help with statistics and to Chris
Whyatt for his language advice and careful proof-reading of the manu-
script.
My deepest thanks also go to Ewa Kowalkowska and Marta
Kowalkowska for their help in putting the manuscript together and their
10 Acknowledgements

patience in putting up with my long-lasting desire to introduce changes

and improve the book.
Finally and wholeheartedly I would like to thank my husband Chris
for his unfailing support and my sons Alex, Maks and Wiktor who in their
secret ways fueled my energies to write this book.
Introduction

Translation as a human skill is a broad complex issue. It can be studied

from various perspectives and every different approach can contribute
relevant points. Taking into account the multitude of languages in the
world and the fact that there are currently more people who speak at least
two languages, the number of potential translators must be impressive.
Yet, translation is still a socially misunderstood phenomenon. On the one
hand, the ability to translate from one language into another is socially
expected of anybody who can communicate in the two languages in ques-
tion. On the other hand, there are many voices which point out that there
are many publically available translations of poor quality and instead of
helping in cross-cultural communication, they confuse those who need
them to interact with others whose language they do not share. Examples
of such ambiguous and odd translations are encountered across the world
from restaurant menu translations to travel information available on the
Internet. There are translations of books which are a pleasure to read and
one rarely stops to think that what is an enjoyable read is in fact a transla-
tion but there are also books which have been so oddly translated that one
finds it difficult to follow the line of reasoning and might be tempted to
blame the author rather than the translator who made a mess of the au-
thor’s ideas by rendering them in a third rather than second language. Ob-
viously there must be a broad range of skills involved in producing a
translation and the unfortunate outcomes point to a lack of certain abili-
ties, competencies and expertise.
This book is a modest attempt to show the human ability to translate in
its developmental continuum from the predisposition to mediate meaning
across different systems of communication including human languages to
what is considered translation expertise. Taking this developmental per-
spective allows one to encompass a wide range of relevant factors which
affect the human ability to perform language translation. There are still
many unclear points on the developmental route a person who chooses a
12 Introduction

career in translation has to cover. The observations made in this book al-
lowed me to hypothesize that one of the essential aspects of translation ex-
pertise development is the human ability to integrate knowledge which is
needed to perform a translation task at hand. This ability still exceeds the
capacity of machine translators. Making an effort to grasp how translators
activate and integrate knowledge into what I termed a Knowledge Integra-
tion Network (KIN) is hoped to enhance our understanding of the human
translator before an attempt is made to apply it to machine translation. Be-
fore the futuristic dreams of computer translation are made true the need to
educate professional translators is constantly growing and calling for more
efficient translator training methodology.
The book consists of seven chapters and is divided into a theoretical
and empirical part. Chapter 1 aims to establish common grounds for a
comprehensive view of translation as a human skill presented in this
analysis. It starts with a suggestion that we, as humans, are in fact all
translators if we consider the nature of human language which encodes
our ideas and intentions and allows us to communicate with others and
deposit in language vital information for future generations. Yet, we rarely
think about language use as translation unless we experience problems
and suffer misunderstandings. This perception of a language user as a
translator might lie at the heart of the human predisposition to translate
across language barriers. It is this unique predisposition to transfer infor-
mation encoded in one language into another language which opens the
developmental continuum of translation as a human skill. Many forms
and facets of translation are an outcome of this predisposition which gives
people a natural ability to translate. For many intercultural encounters this
natural ability is sufficient to ensure social interaction in multilingual
communities. Translation is indeed a broad social phenomenon, bilingual
children translate on an everyday basis, L2 users translate, L2 learners
translate and of course practicing translators have their hands full to keep
communication going on a global and local market. The point which I
make in chapter 1 is that although they are all translators they occupy dif-
ferent stages on the developmental continuum of the human ability to
translate. Some attention is paid to the services provided to humanity by
practicing translators and the lack of thorough research into translation
expertise development is pointed out.
 Introduction 13

Chapter 2 discusses the bilingual foundations of the human ability to

translate. By providing a detailed look at what it means to know two lan-
guages the background for the development of translation competence is
analyzed. The body of knowledge frequently taken for granted by Trans-
lation Studies in tacit assumptions that all those who aspire to become
professional translators have mastered the two languages is shown as a
complex dynamic system of knowledge both declarative and procedural.
A point is made that bilingual knowledge is differently internalised by
natural bilinguals and L2 learners with different consequences for their L2
and L1 performance as well as for their translation performance. Some
light is shed on the issues relevant to translation as an activity which in-
clude bilingual memory, cross-linguistic influence and the need for lan-
guage control in the bilingual mind in which both languages interact with
one another and compete for dominance. Translation as a conscious op-
eration requires taking this dynamic interaction under sufficient control to
keep the two languages safely apart in translation performance.
Chapter 3 looks at translation as a natural ability of all bilinguals (i.e.,
everybody who can use two languages for communication). Two groups,
natural translators in multilingual communities and L2 learners are ana-
lyzed in detail with references providing an insight into research into
translation as an untrained ability, i.e., translation performed by those bi-
lingual language users who have not received any structured training.
Scant research on natural translation shows that this stage on the devel-
opmental continuum of translation as a human skill is worthy of further
investigation. The magnitude of the phenomenon of what has become re-
ferred to as language brokering provides a valuable insight into the addi-
tional capacities needed for efficient translation apart from the bilingual
foundations. They seem to include metalinguistic abilities which as
pointed out by Malakoff and Hakuta (1991) might initially play a more
important role than translation strategies. A closer look at L2 learners act-
ing as translators provides yet another perspective on the untrained ability
to translate pointing to the importance of one’s language acquisition his-
tory in the development of translation as a human skill. A point is made
that L2 translation is of interest not only to the study of translation but
also to SLA research and methodology. A discussion of the benefits L2
learners/users can draw from the structured exposure to translation tasks
including intercultural competence and metalinguistic awareness point to
14 Introduction

some transformations which have to take place for further progression on

the developmental continuum of translation as a human skill to reach the
level of translation competence and expertise.
Chapter 4 takes up the difficult issue of translation competence which has
to be acquired if one is interested in developing the human ability to translate
to the level of translation expertise. Three of the most representative ap-
proaches to translation competence are presented and lack of a unanimously
accepted definition of translation competence is noted. Relying on a shared
agreement among Translation Studies scholars that translation competence
underlying translation as a trained skill is experience-derived knowledge,
both declarative and procedural, an attempt is made to relate the experience
of translating to Translation Competence acquisition. The consequences of
the lack of clear understanding of how Translation Competence is acquired
bring the issue of a ‘pedagogical gap’ in institutionalized translator training.
The critical views of leading scholars are presented and new approaches to
structuring the learning environment so as to optimize the process of learning
translation as a trained skill are reviewed. Finally, different routes to transla-
tion expertise are acknowledged and a glimpse at translation as a profession
with its market requirements is presented. The development of the transla-
tor’s professional self is signaled to involve a further reorganization and re-
structuring of the existing knowledge structures but also an acquisition of a
new professional identity stemming from the awareness of translation as a
socially required service. How the reorganization and restructuring of knowl-
edge takes place and how it helps a developing translator in the actual per-
formance still remains unclear.
Chapter 5 includes my own attempt to understand what translators who
have chosen a career in language translation learn from experience. Relying
on the body of knowledge presented in the previous chapters a hypothesis
is put forward that professional translators in their course of translation
competence acquisition learn to build a Knowledge Integration Network
(KIN) for every translation task they perform. Drawing insights from psy-
cholinguistics, cognitive studies and expertise research the operating prin-
ciples behind the proposal of KIN are explained. The developmental nature
of the translator’s ability to integrate all the knowledge (declarative and
procedural) needed to translate a specific text is discussed in terms of the
differences in approaching the task of translating reported for novice and
experienced practicing translators. The differences exhibited between inex-
 Introduction 15

perienced and experienced translators from the stage of SL text reading to

the stage of revising the first draft of a translation seem to support the de-
velopmental nature of the translator’s ability to integrate knowledge for the
purpose of translation. Finally, the ability to build a Knowledge Integration
Network is seen from the perspective of a developing translator perceived
as an expert learner who through practice coupled with self-reflection be-
comes a Systems Intelligent person and in the process of acquiring his/her
translation competence develops professional self-confidence, a fundamen-
tal feature of translation expertise.
Chapter 6 opens the empirical part of the book. A questionnaire study
conducted among translators occupying various points on the developmental
continuum of translation skill development aimed at verifying the hypothe-
sized concept of the Knowledge Integration Network. Altogether 200 sub-
jects responded to a battery of questionnaires aimed at eliciting valid data re-
flecting their views on translation as a profession and translation as an activ-
ity. The 1BA (N80), 2BA (N40) and 2MA (N40) subjects were treated as
non-professional informants occupying early stages in the evolution of trans-
lation as a human skill and because of studying English at university level
(EFL) they were treated as potential practicing translators. The group of pro-
fessionally active practicing translators (N40) constituted a valuable source
of data on work procedures and the professional translator’s cognitive and
language-related profile. The data analysis was carried out to provide a de-
scriptive support for the hypothesized ability to build a Knowledge Integra-
tion Network for a specific translation task as a fundamental part of develop-
ing professional expertise.
Chapter 7 constitutes the second part of empirical analysis which veri-
fies the hypothesized ability to integrate knowledge looking at the actual
translation process data. Using a key logging computer software pro-
gramme, Translog (Jakobsen and Schou 1999) data are collected from
translators occupying different points on the developmental continuum.
Study 1 investigates translations of the same text performed by 2BA EFL
students (N8), Translation Trainees (N8) and practicing translators (N8).
Study 2 describes an experiment in which 2BA subjects (N48) were di-
vided into two groups which differed by one variable, that of an induced
Knowledge Integration Network. The results are discussed to verify re-
search hypotheses formulated to seek empirical validation of the Knowl-
edge Integration Network.
16 Introduction

Conclusions and possible implications both in terms of translation

pedagogy and future research are summarized and followed by a bibliog-
raphy listing all the depository of knowledge which is relevant to the
study of translation as a human skill, and without which this book could
not have been written.
Chapter 1

When we learn to speak, we are learning to translate

Paz (1971)

Establishing common grounds

The purpose of this chapter is to briefly discuss some essentially important

terms and notions which will frequently be used throughout the entire book.
Translation as a human skill and translation as an activity is frequently a
socially misunderstood phenomenon. As the voluminous literature on trans-
lation shows, it can mean different things to different people depending on
many factors (Snell-Hornby 2006). The need felt by the author to include
this chapter does not in any way mean that other approaches are insignifi-
cant but it aims to establish with the readers a mutually shared perspective
on the genesis and evolution of translation as a human skill.

1.1. Human translating mind

If we take a working definition of translation as a communication of

meaning involving a change of form in which it was originally expressed
it becomes plausible that we are all, in fact translators. In everyday life we
express our communicative intentions using various systems of signs. We
employ our body and senses to encode and decode information. A small
gesture, a wink of an eye, a stare, a smile or a blank face can frequently
say much more than words. In public spaces signs are used to impart in-
formation to large numbers of people. Sub-cultures and various social
groups invent their own codes of communication. Musicians, mathemati-
cians, farmers, grocers or computer scientists use different and unique
systems of communicating meaning within their own communities, and
will readily explain or translate that meaning when communicating with
others who do not share their code and their underlying knowledge. They
will put what they mean in simple words, which are within the reach of
comprehension for a layperson, establishing in this way a shared platform
for mutual understanding and successful communication. Following the
Cooperative Principle and the maxims of language use including the
maxim of quality, quantity, manner and relevance (Grice 1975, Cameron
18 Chapter 1

2002), communication and mutual understanding is possible unless the

participants choose otherwise. Generally, however as human hearers we
assume that people more or less adhere to Grice’s maxims or some kind
of discourse grammar (Paradis 2009) and we instinctively are cued to
make sense, to grasp the meaning intended by the speaker. In this way we
satisfy our social need for communicating with others even if we have to
read between the lines and ignore incidental ambiguities (Pinker 2007).
Language indeed is a peculiar system with its wonderful flexibility
which always remains a user sensitive device. The same message will be
structured differently when addressed to our boss, our colleague, our close
friend, our partner or a child. Indeed, the ability to structure the form in
which people express what they mean, or explain what somebody else
said to other people is so deeply ingrained that we are not aware that we
are in fact constantly translating for others and even for ourselves. As Pel-
latt (2009: 345) reminds “Ottavio Paz said that ‘when we learn to speak,
we are learning to translate’ (Paz 1971). Even a monolingual child learns
by explaining and paraphrasing” (Pellatt 2009: 345).
If one follows the Representational Theory of Meaning which, accord-
ing to Cattell (2006), underlies almost all current psychological research
on thinking, the view that we are all translators gets further support. Fo-
dor’s theory (1975, 2008) that the language of thought which has its own
combinatorial syntax and semantics just like any other natural language
implies that the act of putting our thoughts, intentions and ideas into
words entails in fact an act of translation which, however we remain un-
aware of most of the time, but not all the time. When we can see that our
communicative intentions are misunderstood or misinterpreted by our lis-
tener, we immediately say, ‘this is not what I wanted to say, ‘you don’t get
the point’, ‘what I meant was’, ‘let me put it this way’, etc. As Cattell
(2006: 75) explains in the classical Theory of Mind,

“mental representations are expressed in a language of thought, which

cannot be the same as any natural language, and which must be univer-
sal, in the sense that anyone, of whatever language background, must be
able to interpret it (unconsciously)” [emphasis mine].

In other terms it is not words that make us speak but our communicative
intentions which have to be verbalized as we get on with the business of
living in which numerous things have to be done through language use. In
Levelt’s (1989) speech production model below
 Establishing common grounds 19

conceptualization

formulation

articulation

Fig. 1. Levelt’s (1989) speech production model.

we deal with three levels: conceptualization (when we establish what we

want to say), formulation (when we decide how we want to say it) and ar-
ticulation when the what and the how is actually produced and becomes
available to others who on hearing it will have to, with supersonic speed,
reconstruct our route back to conceptualization so that they can see what
we mean. Since we practice these operations all our lives, that is we plan,
we execute and rehearse what and how we said something to self-assess
the communicative effect (Donald 2006) it is possible that we have never
thought of ourselves as translators.
Yet people differ in how they express themselves (Carroll 1993, Cam-
eron 2001). Some are very good confident communicators who always
seem to get what they want while others are often misunderstood or ig-
nored and unable to successfully communicate their needs, requests, opin-
ions and views. Although many people use language (or other systems of
communication when verbal language is denied due to disability), we all
differ in our ability to communicate and verbal communication is particu-
larly vulnerable to misinterpretation. When the understanding is not
reached and the meaning is not transparent but unclear or ambiguous we
might be forced to work harder to retrace the cognitive route via which
the message arrived, that is from its conceptualization via formulation to
articulation (see Green 1993). Occasionally our personal mental effort
might not be enough and we have to employ experts (i.e. lawyers, special-
ists) or refer to dictionaries (i.e. President Clinton during the Monika
Lewinsky scandal when he attempted to define his ‘relationship’ with the
aforementioned). As put by Nida (2002), the processes taking place in our
20 Chapter 1

mind when we speak are so incredibly rapid that some people might even
say something before they think and then have to bear the consequences
of what they said or more precisely, of what they were understood to have
said. All these problems with communication understood here as a trans-
lation of one’s intentions into words or interpreting words of others as
having a particular meaning occur on an everyday basis within one lan-
guage. Needless to say the risk gets higher when in communication one
has to surmount barriers of language and culture. More frequently than
not, however, people manage to get their message across and verbal
communication, providing that it is possible and granted by nature, is still
the most widespread and economical means of exchanging information
and sharing knowledge in human interaction. Indeed, the human desire to
communicate must be extremely strong to be able to cross barriers created
by a handicap, different modalities or foreign languages. To maximize our
chances we employ a wealth of means including much more than lan-
guage itself, gestures, facial expressions, prosodic features of tone in
speech and other devices referring to form and structure in writing.
The role of language as a communicative device allowing for ‘mind-
sharing’ within a given language community (Donald 1991, 2001) cannot
be underestimated. As pointed out by Donald (2006) knowing a language
allows us to plug in into the community’s cognitive cultural network in-
cluding a store of knowledge accumulated through history and recorded
in language (written records, books, etc.). David Crystal’s (1986) state-
ment that language is the key which opens the door is a valid metaphor
which expresses the role of language in providing access to knowledge
recorded in the language. To share knowledge with another language
community, to interact, do business and co-exist with speakers of other
languages we have to learn foreign languages or hire translators. It is in
this expanded international community that translation becomes more
tangible and widespread as an aspect of everyday life.

1.2. Translation as a widespread phenomenon

Trying to encompass all the forms and facets of translation in language

use Roman Jakobson (1959) in his essay On translation pointed to three
kinds of translation which involve:
 Establishing common grounds 21

a) intersemiotic translation or transmutation in which linguistic signs are

interpreted by means of signs belonging to nonlinguistic systems of
communication (i.e. road signs, sign language, notation in music, and
more recently computer language, etc.)
b) intralingual translation or rewording (paraphrase) in which a message
expressed in one language can be expressed in some other signs but
belonging to the same language
c) interlingual translation or translation proper

The three kinds of translation are what humans have at their everyday
disposal and they make use of them with different, individually based de-
grees of success. However, as Derrida (1992: 225) observes the fact that
Jakobson (1959) did not rephrase the term ‘interlingual translation’ by
giving a ‘definitional interpretation’ like in the case of intralingual trans-
lation being interpreted as rewording and intersemiotic translation being
defined as transmutation is symptomatic and implies a division between a
literal and figurative sense of the term ‘translation’. To quote,

For the two forms of translation which would not be translations

“proper”, Jakobson proposes a definitional equivalent and another
word: intralingual translation or rewording. The third likewise: in-
tersemiotic translation or transmutation. In these two cases, the transla-
tion of “translation” is a definitional interpretation. But in the case of
translation “proper”, translation in the ordinary sense, interlinguistic
and post-Babelian, Jakobson does not translate; he repeats the same
word: “interlingual translation or translation proper”. He supposes that
it is not necessary to translate; everyone understands what that means
because everyone has experienced it (…). In relation to this word, when
it is a question of translation “proper”, the other uses of the word
“translation” would be in a position of intralingual and inadequate
translation, like metaphors, in short, like twists or turns of translation in
the proper sense. There would thus be a translation in the proper sense
and a translation in the figurative sense (Derrida 1992: 225-226).

Following this observation we are all translators in the figurative sense in

our individual way of using one or different systems of communication.
We all experience problems with transferring meaning and master the
skill of communication throughout our lives. Yet, one might argue with
Derrida (1992) that the figurative literal distinction pointed out in the
above quote is not always clear cut, because people sometimes have to
translate what they said to ensure mutual understanding. It is not uncom-
22 Chapter 1

mon to hear somebody say ‘Can you translate it to me as I do not under-

stand what you have said?’ Although there has been very little research
into language processing in which interlingual translation and paraphrase
are cross-examined, scholars generally agree that the underlying proc-
esses share some similarities in both kinds of translation as well as dem-
onstrate differences due to the different codes involved.
Having acknowledged the problems with expressing and interpreting
meaning within one language in intralingual communication, being placed
in a situation when to communicate one has to cross language barriers
gives us a completely different perspective. If as it is assumed there are
about 6,900 languages in the world (figure quoted in 1992 the year set by
the United Nations as “The Year of Endangered Languages”) and we con-
sider ourselves proud users of two languages, it is theoretically possible
that in 6,898 cases we might have to use the services of interlingual trans-
lators. Generally, it is taken for granted that when communication can
only take place through translation (proper, as Jakobson (1959) called in-
terlingual translation), it then becomes the task, privilege and responsibil-
ity of those language users who have the knowledge of two languages and
adequate skills to mediate meaning across language barriers.

1.3. Translation proper – essential distinctions

To start with the term ‘translation’ has multiple meanings in the English
language. It refers to the product, i.e. a book which was translated, which is
the outcome of the translator’s work. This sense of the world is possibly the
most frequently used and common for people outside the translation profes-
sion or students of languages who are likely to view translation as an activ-
ity. For those who work as translators, the word ‘translation’ refers to both
the product and the process which leads to a translation of the source lan-
guage (SL) text which receives a new lease of life as its target language
(TL) version. The act of translating has become to be understood as a com-
plex process of problem solving and decision making with a dynamic inter-
play of cognitive, social and cultural factors (Snell-Hornby 1988, Tymoc-
zko 2005, Kiraly 2005a, Tymoczko and Genzler 2007). The aim of translat-
ing is to overcome communication barriers created by languages (whether
verbal or sign) and to make the meaning expressed in one language system
available to those who do not have the knowledge of that particular system.
Finally, the word translation refers also to the skills needed to perform the
 Establishing common grounds 23

act of translating. Translation is said to be the fifth skill of a bilingual per-

son apart from the basic four skills of reading, listening, speaking and writ-
ing. In this sense it is usually used as a modifier in noun phrases like in
‘translation course’, ‘translation competence’ or ‘translation expertise’.
Shreve (1997: 124) goes even further saying, “The widespread adoption of
the notion of translation competence indicates that there is general accep-
tance in the discipline that translation is a form of knowledge”.
De Groot (1997) defines translation in the following way,
Translation and interpretation involve the rephrasing of a communica-
tion expressed in one language, the source language (SL), in another
language, the target language (TL). The term translation is used both in
a broad and in a more narrow sense. In the broad sense, it refers to all
operations where an SL unit is turned into a TL unit, irrespective of the
modality of input and output (writing, speech or sign language). The
modalities of input and output may be the same or different. When the
term is used in its narrow sense, it refers only to the activity of reformu-
lating written SL text into written TL text (De Groot 1997: 25).

De Groot admits that the ambiguity in terminology can cause confusion,

especially if one uses the term translation to cover both oral and written
forms of translation. As De Groot (1997: 26) explains using “a single term
to refer to both may veil the – fundamental – differences between them”.
The differences, especially in terms of the processes involved, are in fact
so substantial that the two may require a different set of skills to be per-
formed optimally (Carroll 1978).
To avoid this kind of confusion and irrespective of the fact that both
oral and written translation share some general features (i.e. the sheer
communicative purpose, transfer of meaning into forms of a different lan-
guage, crossing linguistic and cultural barriers, etc), the term ‘interpreting’
(i.e. consecutive interpreting, simultaneous or conference interpreting, or
community interpreting) is used to refer to the specific processing demands
of oral translation as compared with written translation. Some authors like,
for example Gile (1995/2009) use the capitalized form ‘Translation’ to refer
to both written translation and interpreting when there is no need to distin-
guish between the two modalities. Since this work is about translation as a
human skill the term translation refers to those general shared features of
Translation in its oral and written modality. When the need arises the dis-
tinction will be drawn between interpreting and written translation, the ex-
perimental part of the work, however is based on written translation.
24 Chapter 1

1.4. Translation as a human skill – common misconceptions

The plausible possibility that we are all translators might have contributed
to some misconceptions about the human ability to translate between two
languages. As put by Holmes (1988: 103), “the translator is in this sim-
plistic common-sense view, a kind of cross-linguistic transcriber or copy-
ist, a slightly glorified typist”. This kind of commonly shared expectation
about the ease of translation comes from monolingual clients seeking
translation services and from bilinguals themselves until they sit down
with a text and try to perform the activity (Whyatt 2010).
For a lay person the skill of translating goes together with being bilin-
gual, for professionals as well as for researchers interested in the relation-
ship between bilingualism and translation skill, translating and especially
simultaneous interpreting is perceived as “perhaps an uncharacteristically
extreme version of bilingualism” (Paradis 2005: 411). These common
misconceptions about the human skill to translate texts/utterances ex-
pressed in one language into another language most probably result from
a simplified view of the translation process, which is, perceived by those
without any experience in the task as a process of linear transcoding of a
string of words in a source language (SL) text into a string of translation
equivalents of these words in the target language (TL) text. Needless to
say if this was the case computers would be easily able to replace the hu-
man translator and there would be no need to write this work. Yet, the
common belief that when translating a person who knows two languages
has to simply press a language switch button, or a lever somewhere in the
mind and while scanning the SL text can instantaneously produce its
translation, has led to unrealistic demands frequently directed to foreign
language students or to professional translators. As observed by Nida
(2002) it is not uncommon for people to commission translation work
from students in foreign language departments and it is not uncommon for
freelance translators to turn down a translation job which requires 100
pages of a financial report to be translated and ready on the chairman’s
desk before 10 a.m. the following morning. The tangible results of these
common expectations that translation as a human skill is automatically
granted to all users of a second language (hence L2 users) are usually
those translations which as pointed out by Korzeniowska and Kuhiwczak
(1994) should never see the light of day. Yet, it is possible that a lot of
misunderstanding and many misconceptions are, in fact, a result of termi-
 Establishing common grounds 25

nological confusion where terms like translation ability, translation skill,

translation competence or translation expertise are used almost inter-
changeably as meaning the same. Distinguishing between them in this
work is important.

1.5. Translation as a predisposition, ability, skill, competence and expertise

As observed by Carroll (1993: 3), “[a]lthough the term ability is in com-

mon usage both in everyday talk and in scientific discussions among psy-
chologists, educators, and other specialists, its precise definition is seldom
explicated or even considered. It is a word that seems to be accepted as a
sort of conceptual primitive, and in fact it is intimately related to such
commonly used words as able and the simple modal auxiliary can” (Carroll
1993: 3). What is more, as noted by Carroll dictionaries seem to be of little
help in providing exact definition of the term and frequently are circular in
the explanations they give. The American Heritage Dictionary, for example
defines ability, “as a quality of being able to do something; physical, men-
tal, financial, or legal power to perform” (…) but able is defined as “having
sufficient ability”. Furthermore, some dictionary definitions of “able”
though bring confusion especially when they give synonyms such as: skill,
faculty, talent, capacity, cleverness, efficiency, aptness or competence1. In
dictionaries of Psychology, especially the more modern ones, the term abil-
ity does not occur as an entry although it is frequently used in numerous
contexts (Encyclopedic Dictionary of Psychology by Harré and Lamb
1983). In older dictionaries, the term is usually defined as “actual power to
perform an act, physical or mental, whether or not attained by training and
education” (English and English 1958). Carroll suggests that the most logi-
cally and semantically justified view of ability is that of “potential” (Carroll
1993: 4) open to individual variation.
It seems that the terminological confusion between ability, skill and com-
petence is also present in the literature on translation, and although this prob-
lem will reoccur in further discussions, it is important to give it some atten-
tion before the major argument develops. Let us analyze the following quote:
Harris (1977) and Harris and Sherwood (1978) proposed the concept of
natural translation, which is said to be an ability of bilinguals, that is, it
is a derivative of bilingualism and appears as bilingualism develops.

–––––––––
1
http://en.wiktionary.org/wiki/ability
26 Chapter 1

Lörscher (1995) characterizes their position as follows: “Harris and

Sherwood emphasize that translation competence unfolds parallel to the
development of bilingualism, and that the degree of translation compe-
tence increases automatically to the extent to which a child’s ability to
use the two languages involved develops” (p. 113) (Shreve 1997: 121).

As can be observed in the above quote, it is taken for granted that the
meaning of “ability” is synonymous with the meaning of “competence”.
Looking further, translation competence is synonymous with translation
as a skill, like for example in the title of the article by Neubert, “Compe-
tence in translation: a complex skill, how to study and how to teach it”
(see Ivanova 1998: 92). Yet, in the same article Neubert defines compe-
tence as a combination of “complex knowledge and skill” (Neubert 1992:
412). Although the question of translation competence and the acquisition
of expert knowledge will be dealt with in detail in chapter four, I would
like to emphasize that equating ability with skill and competence may
lead to false assumptions that ability (potential) guarantees competence. It
seems much better to consider the three in terms of a developmental con-
tinuum which may take the following route: from predisposition to ability
to skill to competence and finally to expertise in translation. A similar ob-
servation was made by Shreve (1997: 125) who suggested that translation
ability should be viewed in “a kind of evolutionary space” where the
starting point is indeed the natural ability of bilinguals to translate. The
ultimate stage to which translation ability can evolve under favourable ex-
ternal circumstances (i.e. the need for translation services) and internal
conditions (the translator’s conscious effort to develop) into expertise,
still remains open to improvement through practice, or vulnerable to attri-
tion due to lack of practice.
If then we accept the definition of translation ability as a potential, a
‘mere predisposition to translate’ (Toury 1995), translation skill can be de-
fined as an actual demonstration of this potential. According to the MSN
Encarta On-line Dictionary2 skill can be defined as:

1. ability to do something well: the ability to do something well, usually

gained through training or experience
2. something requiring training to do well: something that requires tra-
ining and experience to do well, e.g. an art or trade
–––––––––
2
(http://encarta.msn.com/dictionary).
 Establishing common grounds 27

The term ‘competence’ is defined in the same dictionary as: “the ability to
do something well, measured against a standard, especially ability ac-
quired through experience or training”. The two examples which follow
illustrate the meaning, “People began to question her competence as a
teacher”, “I don’t doubt his scientific competence for a moment”. Conse-
quently, ‘expertise’ is defined as “the skill, knowledge or opinion pos-
sessed by an expert”. Expert in turn is defined as “somebody skilled or
knowledgeable: somebody with a great deal of knowledge about, or skill,
training, or experience in, a particular field of activity”.
This suggests that it is justified to see translation skill as evolving
from the human predisposition to mediate meaning serving as a basis for
a natural ability to translate (i.e. in bilingual children) and spreading over
a continuum of different developmental stages. Depending on how fre-
quently the ability is exercised in translation performance and whether or
not the experience of translating is, in a sense educational in terms of
leading to improved performance, the ability to translate will develop into
a more refined skill, which referring to point 1 above is then the ability
not only to translate but to translate well. Let us accept for the time being
that translating well as suggested by Shreve (1997: 125) means that the
results of the translation performance are good with reference to results
accepted as professional. The same developmental aim can be a product
of formal translation training where the natural ability to translate is ex-
plicitly developed into a professional skill which is hoped to further de-
velop into translation competence and later into translation expertise.
This ‘evolutionary space’ or a developmental continuum allows for a
wide spectrum of performance referring to how well translation is done
judged against professional standards from very poor, poor, adequate to
good, very good, outstanding, excellent. What is important though is the
premise that “[m]ovement within the space is not automatic or necessary
and the end point is not a single cognitive set shared by all translators
who arrive at professionalism” (Shreve 1997: 125). In other words, trans-
lation as a skill allowing for skilled performance is not something in its
entirety given, or as it was, and still is, frequently assumed granted by na-
ture to some privileged talented individuals (see Piotrowska 2007) but a
complex skill which undergoes developmental evolution under favourable
circumstances (Toury 1986). From this view of translation skill as a dy-
namic quality, the definition of translation competence as an underlying
knowledge of how to perform when translating is also an evolutionary
28 Chapter 1

quality with different levels of attainment. This view of translation com-

petence as ‘a cognitive set of knowledge’, however is still not kept clear
from other terms and is indeed used as synonymous with ability and skill.
The quote below demonstrates this circularity of the three terms dis-
cussed, translation ability, skill and competence:

We assume that translation competence is not to be understood only as

a repertoire, but rather as a role-specific competence. In this sense,
competence includes not only the means (repertoire) but also the pur-
ported result, that is, competence is defined as the appropriate use of
specific abilities according to surrounding demands (McClelland 1973),
i.e., as a goal-oriented behaviour. In the domain of Translation Compe-
tence it includes and specifies the notion of competence as ability
(Alves, et al. 2001: 47).

As indicated above, such terminological confusion allows for assumptions

that their might be a relation of equality between competence and being
able to translate, which everybody who knows two languages can do. In-
deed, everybody who has access to two language systems can translate but
the end result of the translation performance will differ depending whether
the translator is relying only on the natural ability or using his/her con-
sciously developed competence or expertise. This variability in translation
performance (Séguinot 1997) visible in the quality of translated texts is fre-
quently unknown to the general public which has a simplified view of
translation skill. To quote Korzeniowska and Kuhiwczak (1994: 11), “It
seems that translation is the skill most commonly expected of somebody
who speaks a foreign language. It is generally assumed that being able to
communicate in a foreign language corresponds to the ability to translate
from one language into another. In many respects, this conviction resem-
bles the widespread opinion according to which every Chinese person is an
excellent cook”. To safeguard this kind of misunderstanding, it is important
to view the three frequently used terms as an evolutionary continuum:

PREDISPOSITION → ABILITY → SKILL → COMPETENCE→EXPERTISE

This evolutionary perspective makes it possible to accept that all people

who know two languages are able to translate but their performance will dif-
fer depending on the stage they are at on the developmental continuum
demonstrated above. Consequently, the generic term ‘translator’ has to be
 Establishing common grounds 29

also viewed as a developmental continuum which reflects the movement

within the ‘evolutionary space’. The progression on the developmental con-
tinuum will have qualitative and quantitative consequences. First of all the
further on the developmental continuum a translator is the richer and more
refined set of abilities and skills he or she will have. In terms of quantity
however there will be a filtering effect. From all people who have access to
at least two languages and who by nature are predisposed to translate (just as
they are predisposed to communicate), potentially all can use this ability and
translate when the need arises. However, only some of those who translate
will make the effort to refine their ability and will become capable of a
skilled performance. Possibly some percentage of skilled translators will
choose to pursue a career in translation and will develop translation compe-
tence either with or without the support of structured education. Finally,
some of those who are competent practicing translators will develop to
reach the level of translation expertise (Hoffmann 1997). In a way just like
in any other complex skill there probably is a process of self-selection gov-
erned by a combination of individual factors (personal predispositions in-
cluding affective factors such as a love for languages, cognitive abilities,
personality features) and environmental impact (social need, personal cir-
cumstances) which decide that some, and frequently few individuals be-
come experts in their chosen areas of expertise (Ericsson and Smith 1991).
The pyramid-like figure 2 below illustrates the point.

expertise

competence

trained skill

untrained ability

natural predisposition

Fig. 2. Evolution of translation as a human skill.

30 Chapter 1

It is still very much unclear how the progression takes place although
some assumptions can be made relying on the research accumulated on
the cognitive psychology of expertise in diverse domains (Hoffman 1992,
1997, Chi, Glaser and Farr 1988, Séguinot 1989b, Ericsson and Smith
1991, Feltovich, Ford and Hoffman 1997, Green and Gilhooley 1992). It
can be expected that each stage is in itself a process of development or
self-development of an individual who deliberately seeks the experience
of translation. The progression from one stage to the next one is probably
a slow process leading to a developmental shift rather than a quick jump.
The one aspect which is seen as necessary for development that all schol-
ars agree upon is the practice of translation since translation competence
is commonly defined as experience-derived knowledge (PACTE 2003).
Leaving this complex issue aside as it will be perused in the latter chap-
ters of this book, let us first establish how the progression from translation
predisposition to ability, skill, competence and expertise bears on the per-
ception of the person involved, the translator.

1.6. The translator

Following the rules of morphology a person who drives is a driver, a per-

son who reads is a reader and a person who translates is a translator. Yet,
the word translator like writer is not commonly associated with anybody
who translates or writes and more frequently it is meant to refer to the
name of a profession.
As observed by Malakoff and Hakuta (1991),
Translation is typically viewed as a valuable skill that is available only
to the highly trained and linguistically sophisticated bilinguals who
come out of interpreter and translator training schools. It is not a skill
that is generally considered to be within the repertoire of just any bilin-
gual, much less children, much less minority-language children. Yet,
studies have found that children can both interpret and translate materi-
als that are within their comprehension and vocabulary (Malakoff and
Hakuta 1991: 144).

It seems then that translation as a human skill is a socially misunderstood

phenomenon. On the one hand, the skill of translation is granted to a prac-
ticing translator, on the other hand it is expected of bilinguals (anybody
who can use two languages for communication). Translation as a profes-
sional skill and the professional translator has been the focus of investiga-
 Establishing common grounds 31

tion for Translation Studies but the fact that translation ability as a poten-
tial to translate is open to all bilinguals (whether natural by birth or ac-
quired by socio-cultural immersion or formal language teaching) has not
been considered as an object of research able to contribute anything to our
understanding of translation as a human skill (see Toury 1995, Krings
1986b, Harris 1992). The common fact that in multilingual communities
translation performed by bilinguals who have not received any formal
training (i.e. natural translators) is a part of everyday life and performs a
communicative function thanks to which such communities co-exist has
not earned a lot of scholarly interest. What is worse, the outcome of this
untrained translation ability has been described as deviant from the stan-
dards set up for professional translators and left out as being unable to
contribute anything to the course of evolution of translation as a human
skill. In consequence there is a dearth of empirical research into transla-
tion expertise development (PACTE 2003) although there are voices that
point out the need to investigate the development of the human ability to
translate (Cronin 2005, Kiraly 1995, 2005a). Some scholars have made
valuable suggestions as to how the development of the human ability to
translate might develop (Kussmaul 1995, Gile 1995, Toury 1995, Shreve
1997) and their ideas have been widely accepted by the Translation Stud-
ies (hence TS) community, still empirical validation is not available and
the literature is based on experience-based assumptions about the devel-
opment of translation expertise (see Chesterman and Wagner 2004).
Translation teaching pedagogy is anxious to receive a model it could
safely adopt for the practical purposes of translator training (Kelly 2005,
Cronin 2005, Tennet 2005). A model that would respect the continuity of
the developmental process and encompass all forms and facets of transla-
tion as a social communicative phenomenon generated by all communica-
tors who assume the role of the translator is very much desired.

1.6.1. Need for a developmental perspective

Indeed, it seems only fair to admit that translation is a broad phenomenon.

Bilingual children in multilingual communities frequently act as language
brokers (for a review of literature see Morales and Hanson 2005) and
community interpreters for their relatives and neighbours. They are asked
to translate because they can do it, and they do it even if it requires to take
on grown up roles as communicators. Foreign language students are often
32 Chapter 1

asked to translate texts from and into their foreign language because they
are socially expected to be able to do it and if they say they are not able to
produce a translation they feel their knowledge of the foreign language is
undermined. Translation as a human skill is socially expected of people
who have access to two languages. In consequence, the term translator
should be able to encompass a range of abilities. In other words, a translator
is never a stable entity but he or she is always positioned at some point on
the developmental continuum from being a novice to being a professional.
Although, as observed by Shreve (1997) even the term professional transla-
tor is problematic, in his own words,

I’d like at this point to dispel the notion that professional translation is
synonymous with either graduation from translation schools or the sell-
ing of translation services on the open market. In the literature, there is
some significant confusion about what professional translation [and
thus professional translator – added by me] means (Shreve 1997: 125).

As mentioned above Translation Studies have primarily focused on the pro-

fessional translator but much more has been said about what the translator
should be like than how one becomes a professional translator. Snell-
Hornby (1992) in her article “The professional translator of tomorrow: lan-
guage specialist or all-round expert” set up very high standards claiming
that the translator should rather resemble “the intellectual polymath, and a
polyglot as well”. To quote, “Our ultimate aim is to develop latent linguistic
(and cultural) talent into the professional competence of an expert, who, as
a specialist in the fields of language, culture and communication, can work
as a generalist in a great many areas where translation is required” (Snell-
Hornby 1992: 22). Carrying on Snell-Hornby says, “obviously our young
graduates can at best be seen as potential experts of this kind, for all we
know, in the professional life of a translator nothing is as vital as experi-
ence”. Indeed, the word “experience” has become, and still is the magic
word, in the debate on what constitutes a professional translator. However,
it has never been specified what the vital ingredients of experience are from
which the professional translator emerges. As observed by Shreve (1997:
128) if the assumption is made that translators learn from their experience,
the relevant questions are these: ‘What are they learning?’ and ‘How are
they learning it?’ These questions formulated 12 years ago are still pending
some plausible systematic empirically based answers. To keep telling the
aspiring translators that ‘practice makes perfect’ is clearly not enough.
 Establishing common grounds 33

Walters (2005) shared some observations which show the clash be-
tween high expectations set for professional translators and the human as-
pects of the skill.
The interpreter’s/translator’s identity is presumed in some contexts and for
some genres to be anonymous, objective, and even omniscient – no mean
task for a mere mortal who has sometimes been compared to an airport
control tower operator. In reality, the interpreter/translator is human, male
or female, with a fixed age, professionally trained, and may have come to
work with a host of personal and circumstantial qualities that have greater
or lesser relevance to the task in hand (Walters 2005: 212).

It is precisely the human aspect of the translator indicated by Walters

(2005) which possibly has been overlooked in the study of translation
skill development. For much of the TS research the translator’s identity,
that is his or her psycholinguistic profile, is frequently unspecified and
undefined which consequently might lead to misunderstandings in the in-
terpretation of findings provided by various studies. Snell-Hornby (2006:
123) commented on the misleading claims of Krings’ (1986) research who
used the method of thinking aloud (TAP studies) trying to investigate
what is going on in the translator’s mind saying that Krings (1986) did not
investigate professional translators but language students. One can con-
clude that L2 learners are then contrary to the common expectations un-
able to translate, and do not deserve to be called translators. Yet one could
ask, what can they be called when they in fact are fully able produce a
translation? Like natural translators and language brokers L2 learners can
act as translators who use their natural ability to translate.
Furthermore, a lot of research into language processing in translation
is using both the terms ‘translator’ and ‘translation’ to any kind of inter-
lingual task, such as single word translation or the simple transcoding of
sentences, and text translation. This overextension of the term is of course
morphologically motivated but just as it was the case with the use of the
term ‘bilingual’3, it can lead to a misinterpretation of research results. A
lot of studies which report findings on the process of translation do not
investigate professional translators or interpreters but use bilinguals
sometimes with a very brief bilingual history of their second language ac-
–––––––––
3
Mostly expected to mean natural balanced bilinguals who as shown by Grosjean
(1982) and others are rare or non-existent, but also used to refer to anybody who can
speak two languages.
34 Chapter 1

quisition. Obviously there must be vast differences between somebody

performing a translation task who is only a fluent bilingual, a novice
translator undergoing training, a qualified translator beginning his/her ca-
reer and a professional translator with several years of experience (Hoff-
man 1997). Investigating how they all cope with the task can bring impor-
tant insights into the evolution of translation as a human skill.
In view of the above considerations, however, about the developmental
nature of translation as a human skill, and to be consistent with their impli-
cations for the person who is the agent, the translator, it is only fair to sug-
gest that the individual performance of the person who translates will re-
flect the place on the developmental continuum of translation as a human
skill. This developmental perspective makes room for all kinds of transla-
tors and for all kinds of translations produced. The translator-in-the-making
is always somewhere on the developmental continuum from being able to
make use of the natural ability to translate throughout the process of devel-
oping specific skills towards achieving translation competence and exper-
tise. Yet, this developmental view is not explicitly admitted. It is of utmost
importance in this work to emphasize that in this study the translator is
viewed as a developmental entity. He or she has to cover a certain route of
knowledge acquisition and building up his/her repertoire of skills to be able
to produce translation up to professional standards. It is a developmental
route that is never completely finished (see Shreve 1997, PACTE 2003) as
when providing translation services one is constantly forced to learn not
only new words but generally acquire knowledge. As it is known from cog-
nitive studies acquiring new knowledge and learning new skills will always
have a restructuring effect on the existing knowledge and skills (Anderson
1980, 1986, 2005, Donald 2007b).
Taking this broadminded view allows one to see the breath of transla-
tion as a human skill and might lead us closer to understanding its multi-
layered complex developmental nature. After all it is difficult to imagine
where we would now be in terms of humanity and civilization if it was not
for the human ability to translate, to transmit knowledge and disseminate
achievements and advances to all or at least many citizens of the world.

1.6.2. The role of translators

It is difficult to talk about the role translators have played throughout his-
tory (Delisle and Woodsworth 1995) without running the risk of being
 Establishing common grounds 35

pompous. Their presence confirms the powerful need to communicate

which is ingrained in people and the results of their work are indispensa-
ble for a feeling of the continuity of our civilization and for transmitting
knowledge. If it was not for translators we would not know the works of
great philosophers from ancient Greece, for example, and the name of
Agora, a place where the idea of democracy was born would be alien and
meaningless to the modern world. It is beyond imagination to envisage
how much impoverished we would have been not knowing the works of
great classic writers, novels by Hugo, Dickens, Dostoyevsky, and we
know them because they were translated and by that made available, ac-
cessible to users of other languages than the original language they were
written in. It is translators who provide a vital link between cultures and
nations divided by languages throughout history, although they them-
selves remain invisible otherwise than in the translations which they pro-
duce (Venuti 1995).
As put by Gentzler (1993: 1) translation as social practice is “as old as
the tower of Babel”. If one is willing to take the biblical perspective,
translation services became required with the fall of the Tower of Babel
where the tongues were mixed to create confusion and since then humans
became forever destined to translation and subjected to translation, “to the
law of a translation both necessary and impossible” (Derrida 1992: 226).
Derrida continues on the same page saying that after the fall of Babel,
“Translation becomes law, duty, and debt, but the debt one can no longer
discharge”. No matter if one is ready to accept that the fall of Babel at
least symbolically marked the beginnings of the translator profession, it is
not infrequently referred to by translation theorists. Steiner (1975), for ex-
ample entitled his influential work on translation, “After Babel”. Needless
to say the story of the Tower of Babel itself has become known worldwide
through translation. In a way one might say that the history of human civi-
lization is also a history of translation and much that we know about it we
know through translation and this means we owe our thanks to translators.
In today’s world translators are in growing demand. They accompany
politicians on their foreign visits, translate literary works as soon as they
are written by acknowledged original writers, work in international busi-
ness relations, translate official documents and news items as soon as they
are released by news agencies throughout the world. It is through transla-
tion that new technological advances are shared and knowledge and ex-
pertise achieved in one country can be disseminated and used to improve
36 Chapter 1

the life of people in others, no matter how geographically remote the places
are. It might be trivial to say that while languages make us citizens of our
own countries, translators make us citizens of the world, trivial but true.
Since translators are indeed social agents (i.e. gatekeepers see Chau
1999: 233) responsible for what is saved and consequently accessible to
others, as well as for what remains for ever lost in translation and there-
fore inaccessible to others, a viable question to ask is what constitutes
their expertise. What is the route that one has to cover from being able to
use two languages to being able to translate between them with fluency
and confidence? Can all the people who are bilingual (in the sense that
they can use two languages in everyday life) become successful transla-
tors? What skills are needed to translate with confidence and produce
competent fully functional translations? Unfortunately, there are no ready
made scientifically approved answers to these questions and although
translation has always been part and parcel of human life, and continues
to be in acutely growing demand, translators themselves have always re-
mained in the shadow of their work. We know the great classic books and
films but we do not register who translated them. We simply deposit trust
in translators as we do in any other specialists and professionals to whom
we have to resort, that they did their job well. Most practicing translators
are aware of that trust and indeed produce excellent translations, but there
are also those who unlawfully call themselves professional translators
since it is legally unregulated in some countries including Poland and the
US where no official accreditation is needed to provide translation ser-
vices. As a result many translations which appear in print are produced by
natural translators who might not be themselves aware that they abuse the
social trust of a client who seeks translation services and is usually not in
a position to assess the quality of the translation which he or she receives.
It is not uncommon to come across strange or funny translations, like for
example on a menu in a Polish restaurant where according to a neatly pre-
sented translation one can order, ‘trout suffocated in herbs’ and to go with
it, perhaps a glass of ‘sparking wine’ (Whyatt 2003). Although the issue
of translation quality is complex and multilayered (House 1977, 2009)
there is a close connection between the stage at which a given translator is
on the developmental continuum of his/her skills and the quality of work
he/she is able to produce (Shreve 1997, Cronin 2005). Since the word
‘translator’ is extremely stretchable as it was pointed out in the section
above, and since we are all interested in translation products which are of
 Establishing common grounds 37

high communicative standards, the demand to investigate the route from a

novice to a professional seems well justified in view of the fact that so far
it has unfortunately remained largely under investigated (PACTE 2003).

1.7. The scope of systematic research in translation expertise development

The reasons for the relative lack of research into translation expertise de-
velopment are complex and require taking a historical interdisciplinary
perspective when looking at those disciplines which have, or should have
a viable interest in the study of translation in general. First and foremost,
the discipline of Translation Studies is entirely devoted to the study of
translation in its all possible aspects concerning both theory and practice.
Since, however translation is a form of communication and in its verbal
form is dealing with language it belongs to the area of applied linguistics,
sociolinguistics, psycholinguistics and cognitive psychology. Let us
briefly inquire to what extent the topic of translation as a human skill has
been investigated in the three most relevant disciplines: linguistics, Trans-
lation Studies, psycholinguistics and cognitive psychology.

1.7.1. Linguistics

Linguistics is a discipline with well established roots devoted to the study

of language with a broad spectrum of interest including theoretical and
applied linguistics. Historically, it also encompassed the study of transla-
tion but in the early 1980s it was considered to provide a too narrow re-
search paradigm to study multiple aspects involved in translation (Berman
1989). Following the establishment of the new independent discipline
named Translation Studies (Snell-Hornby 1988) the topic of translation in
both theoretical and empirical studies carried out by linguists has rarely
received attention. Although major debates echoed von Humboldt’s con-
viction that there must be a midpoint shared by all languages (i.e. Chom-
sky 19654), the focus was always on structures rather than on people who
use them. Contrastive analysis (Lado 1957) studied how the postulate of
Universal Grammar (UG) was differently realized in various languages.
The similarities and differences between languages were studied to pre-
–––––––––
4
Chomsky (1965: 202) however in an enigmatic way stated that “although lan-
guages are to a significant extent cast in the same mould, there is little reason to suspect
that reasonable procedures of translation are in general possible”.
38 Chapter 1

dict a learner’s difficulties (Arabski 1979, Fisiak 1991), but the human
skill to translate was rarely used as a window on the two opposing para-
digms: UG versus linguistic relativity (with few exceptions, i.e.
Lewandowska-Tomaszczyk 1985). Linguistically confined reasoning when
applied to translation allowed only for theoretical discussions of equiva-
lence and untranslatability without reference to actual translation practice
(Hatim and Basil 1990).
According to Berman (1989) theoretical linguists have no real interest
in translation studies although linguistics continues in a way to insist that
translation is a proper object of linguistic investigation as it is able to offer
a conceptual analytical framework for the study of translation. However, “it
defines translation in such an abstract way that it ignores almost entirely the
written and textual aspects of the act, not to mention its cultural and histori-
cal dimensions” (Fawcett 1997: 144). Fawcett (1997) argues though that
linguistic discourse although not able to circumscribe translation, “has a
role to play and a voice which will not be silenced” (Fawcett 1997: 144).
Indeed, the relationship between linguistics and the study of translation
has been a troubled one (Fawcet 1997). It is quite risky to guess how many
linguists, whether theoretical or applied, would see benefits in the study of
translation? Looking at linguistics and Translation Studies one is likely to
notice that the reluctance is not one-sided but rather mutual. Snell-Hornby
(2006) in her recent publication, The Turns of Translation Studies expressed
her worries that some translation scholars might want to put an end to the
troubled relationship between TS and Linguistics, saying that,

This impression and particularly the observation that the pendulum is

swinging back to the past, was for me confirmed at the close of the
Third EST Congress in 2001, when there was an informal general ses-
sion to give participants the opportunity to comment on the contents
and results of the conference. The younger generation in particular were
invited to present their opinions. Most striking for anyone familiar with
the course of the debate over the last thirty years was the tendency no-
ticeable both in the topics of the conference programme and in the
comments of that closing session, “Back to Linguistics” (cf. Snell-
Hornby 2002). Is the translational wheel to be reinvented yet again?
Despite the promise of “new tools and methods” (Chesterman 2002), it
might seem so indeed (Snell-Hornby 2006: 151).

The above quote shows, on the one hand a genuine concern about continu-
ing the study of translation with the turns (i.e. the cultural turn and the func-
 Establishing common grounds 39

tional turn) the independent discipline of Translation Studies has taken

since it freed itself from the exclusively linguistic paradigm. On the other
hand however, it fails to see that the linguistics of today is no longer the
“straitjacket of the time” when the approach to translation was product ori-
ented and the translation process was considered as a purely linguistic op-
eration in which the aim is achieving equivalence of meaning through a
transformation of structures (Catford 1965, Nida 1964, Koller 1972). The
impossibility to capture the essence of translation even only as a specific
case of source-language-text-induced target-language text production
(Neubert 1985: 18 quoted after Schäffner 1998: 83) led to the emergence of
a new discipline to study the multi-faceted phenomenon called translation.
Still, translation intrinsically involves the use of language or languages,
both the use of language and translation are goal oriented and functional
and serve a common aim of communicating with others. Many disciplines
which originated in linguistics and other related disciplines, like for exam-
ple psycholinguistic, cognitive linguistics, sociolinguistics, text linguistics
or neurolingustics are focused on vital aspects of communication through
language and as such they have a lot to offer to the study of translation as a
social phenomenon, as a cognitive-communicative activity and as a human
skill. Translation Studies have borrowed various concepts from linguistics
(see Lewandowska-Tomaszczyk 1985: 59) and from neighbouring disci-
plines throughout its short history5 (for a detailed review see Malmkjaer
2005). As stated by Chriss (2006) translators are applied linguists, language
professionals and as expressed by Hatim and Mason (1990) in their fore-
word to Discourse and the translator writing on translation is a contribu-
tion to “this important area of applied linguistics research”.
The cooperation and a possible synergy especially in the area of empiri-
cal research has recently been voiced by some translation scholars (De
Groot and Christoffels 2005, Gile 1997, Shreve 1997, Walters 2005, Chmiel
2010) and the time is indeed ripe to bury the hatchet between TS and lin-
guistics. The benefits are most likely to be mutual and might prove wrong
the common conviction expressed by Uwajeh (1994) in the following way:
“If the importance of linguistics for translation is obvious nowadays to lin-
guists and non-linguists alike, it is generally not appreciated by linguists that
–––––––––
5
For example Reiss (1981) and her text typology after Büchler’s organon model of
functions of language, Fillmore’s scenes-and-frames theory (Snell-Hornby 1988),
Toury’s (1995) concept of ‘native translator’ possibly encouraged by Chomsky’s (1965)
native speaker, to give only three examples.
40 Chapter 1

linguistics itself could benefit from the findings of translatology” (Uwajeh,

1994: 287). If so far linguistics has not displayed much interest in the human
ability to use languages in order to translate the next most viable discipline
to take up genuine interest in translation skill/skills is TS itself.

1.7.2. Translation Studies

Translation Studies as a fully independent discipline (with currently a wide

range of interest, see Baker and Saldanha 2009) started to establish itself in
the 1980s following a seminal work of James Holmes (see the collection of
his lectures edited by Broeck 1994). Following the departure from linguis-
tics, translation scholars devoted considerable attention to these aspects of
translation which had not, and could not be investigated within the con-
straints of linguistics (see Gentzler 1993, Snell-Hornby 1988, Toury 1995,
Hatim and Mason 1990). These new areas of investigation included intercul-
tural issues (Snell-Hornby 1988, Bassnett and Lefevere 1990), the question
of norms (Hermans 1991, Toury 1995), functional approach with the skopos
theory (Reiss and Vermeer 1984), which shifted attention from the SL
(source language) text to the TL (target language) text steering far from the
much hated concept of equivalence (see Snell-Hornby 2006: 152, Bassnett
1996). New schools of thought focusing on the function that translated texts
play in the target culture, such as the “Manipulation School” or the school of
Descriptive Translation Studies and Deconstructionism appeared, and new
theories were formulated (see Holz Mänttäri 1984).
With the map of the discipline of TS drafted out by Holmes (1988) the
new research area “should emerge as an empirical science: main split into
Pure vs. Applied branches” (Toury 1995: 9) implying in this way “a
proper division of labour between various kinds of scholarly activity”
(Toury 1995: 9) in the same manner it was done in linguistics. Toury
(1995) in his influential book, Descriptive Translation Studies and Be-
yond insisted that the division between the pure and applied branches
should be kept clear. In his own words,
I would hardly subscribe to the view that (epitomized by Peter New-
mark but shared by so many) that “translation theory’s main concern is
to determine appropriate translation methods” (Newmark 1981: 19);
definitely not any more than “linguistics main concern is to determine
appropriate ways of language use”. Strong as this conviction is, how-
ever, it does not preclude the possibility of drawing conclusions from
 Establishing common grounds 41

theoretical reasoning, or scientific findings, to actual behaviour, be its

orientation retrospective (such as translation criticism) or prospective
(such as translator training or translation planning. This possibility does
exist, of course. However, drawing conclusions is up to the practitio-
ners, not the scholars (Toury 1995: 17).

This determined view of the theory vs. practice divide might seem dated at
the present time where most TS scholars view the study of translation as a
perfect ground to observe how theory and practice feed on each other and
stimulate new ideas (Kussmaul 1995, Chesterman and Wagner 2004). Yet,
Toury (1995) was ahead of his own time when he pointed out that none of
the Applied Extensions of TS can “draw on Translation Studies alone” and
pointed out that the area of translator training, for example apart from draw-
ing on pure TS would be modified by a theory of teaching and learning.
This pioneering observation has for years at the worst remained unno-
ticed and at the best was taken up in the works like PhD dissertations that
have never been published. As observed by Cronin (2005),

Translation theoreticians had in previous decades tended to neglect

translation pedagogy for considerations of translation, text, history, ab-
stracted from the teaching process. Presentations on pedagogy at trans-
lation conferences were devoted either to a scornful repudiation of the-
ory in the name of experience or to thought deadening outlines of
course syllabi which told little if anything about how courses were de-
livered or what their deeper theoretical underpinnings were (Cronin
2005: 250).

Toury’s (1995: 19) remark on the “inherent heterogeneity” of the applied

TS granted by the fact that each of the branches is an “extension ‘into the
world’ of the discipline” implies the need for an interdisciplinary approach
to translator training. His further comment, however in which he claims
that applied TS “cannot be anything but prescriptive” is now questionable6.
Toury when advocating the prescriptive nature of applied extensions said,

–––––––––
6
At the time, however many of the ‘new’ theories took over old ideas (Ljudskanov
1969; Seleskovitch 1976, 1978) which offered purely theoretical constructs with stages
taken for granted based on envisaging “what a translator might do” (Fawcett 1997: 139)
or what he/she should do when working on a translation (i.e. Toury’s concept of a ‘na-
tive translator’).
42 Chapter 1

They are not intended to account either for possibilities or likelihoods

or for facts of actual behaviour, but rather to set norms in a more or less
conscious way. In brief, to tell others what they should have done
and/or should be doing, if they accept these norms (or, very often, the
authority of their proponents) and submit to them (Toury’s 1995: 19).

The prescriptive attitude in a way has for years persisted in TS (see Walters
2005 for the criticism of the prescriptive bias in TS) and it has overshad-
owed the pioneering call for the interdisciplinary approach. If it had been
noticed and if a truly interdisciplinary effort had been made to investigate
the development of translation expertise, the prescriptive attitude might
have given way to more empirically based research (see PACTE 2003).
This was after all what Holmes appealed for when he said, “It seems to me
that before we can know how to train translators, we have to know what
takes place in the translation process” (1988: 95-96). Although the effort to
unravel all the intricacies of the translation process continues and remains
an open area of investigation Holmes’ conviction holds valid.
On the other hand, and frequently in the eyes of representatives of other
disciplines (i.e. Walters 2005) from the very beginning Translation Studies
had a strong practical bias and concentrated on the practicalities of profes-
sional translation with implications for translator training programmes. This
practical bias, according to Walters, for example, “may have blocked or de-
layed the investigation of more theoretically and empirically challenging
questions” (Walters 2005: 209). TS as a discipline was interested not in any
kind of translation but in professional translation, although there still is lit-
tle agreement on what in fact constitutes professional translation (see
Shreve 1997). In consequence, Translation Studies were not interested in
the early stages of translation expertise development or in its origins as a
human skill (see chapter 3 on natural translators). The developmental na-
ture of translation as a skill, and its consequences for translation as a prod-
uct have not been adequately researched to dispel the common misconcep-
tions about the human ability to translate discussed earlier in this chapter.
Cronin (2005) noted that from the 1940s onwards many translation schools
were established but contrary to expectations courses for translators have
not resulted in distinctive approaches to teaching translation. It took fifty
years as Cronin continues, “it was not until the 1990s, at the end of the
twentieth century, that serious monographs began to appear which looked
at the teaching of translation not only as a practical but as a theoretical
problem” (Cronin 2005: 250). There still is a lot of misunderstanding about
 Establishing common grounds 43

what are the essential ingredients of translation expertise, and the old di-
lemmas whether it is indeed an art or a craft, a matter of talent and natural
gift or hard work, practice and perseverance are still debated (Tabakowska
2003, Pieńkos 2003, Piotrowska 2007). It seems justifiable to suggest that
to gain grounds for challenging these dilemmas one should try to retrace
the course of the development of translation as a human skill from predis-
position to expertise. Recent projects (e.g., the PACTE and the TransComp
project) conducted by TS scholars committed to investigating how transla-
tion competence develops in trainees who aspire to become professional
translators will hopefully advance our understanding of the developmental
nature of translation skill. Let us look at psycholinguistics and cognitive
psychology to see whether these disciplines have displayed interest in an
all-inclusive approach to translation skill.

1.7.3. Psycholinguistics and cognitive psychology

In her 1997 article entitled, “The cognitive study of translation and inter-
pretation” De Groot (1997: 26) noted that translation has not been consid-
ered as a subject worthy of investigation in the field of applied linguistics,
psycholinguistics and cognitive psychology. Several reasons were pointed
out for this neglect including, for example the fact that translation was not
emphasized in mainstream journals and that it was too quickly labeled as a
‘special skill’ reserved for highly trained professionals (see also Malakoff
and Hakuta 1991) and therefore judged by many “too complex to grapple
with” (De Groot 1997: 26). Nearly a decade later the niche has not been
filled by psycholinguistic studies of translation in general apart from some
isolated case studies in the form of PhD dissertations (Hejwowski 1992,
Whyatt 2000) which remained unpublished and thus generally not available
to a wider research community. In consequence, in 2005 Walters admitted
that, “There have been surprisingly few systematic and scientific studies of
interpretation and translation (Shlesinger 2000), particularly in the fields of
applied linguistics and psycholinguistics” (2005: 209). Further reasons for
the absence of the interest in the study of translation as an instance of bilin-
gual language performance according to Walters are to be found in “the
sheer complexity of the phenomenon involving hidden social agendas, non-
explicit pragmatic considerations, unbridgeable cross-linguistic gaps and
awesome memory and time demands” (Walters 2005: 209). In 2006 in an
article, “Language control in bilinguals: monolingual tasks and simultane-
44 Chapter 1

ous interpreting” published in Bilingualism: Language and Cognition 9 (2),

De Groot and Christoffels noted that research on translation and simultane-
ous interpreting has not been a part of ‘the mainstream psycholinguistic
work on language control’ expressing at the same time their belief that the
investigation in these areas is “likely to inform and qualify models of lan-
guage control based on monolingual-task studies” (De Groot and Christof-
fels 2006: 189). The authors suggested that the study of translating between
two languages of a bilingual person (here the term bilingual is used to refer
to all the people who are able to use two languages for communication),
“where the term TRANSLATING covers all forms of language use, written
and oral, where a message expressed in one language (the SOURCE lan-
guage) is rephrased into another language (the TARGET language)” could
become a complementary source of data for a common approach in study-
ing bilingualism based on “testing the bilingual participants in monolingual
tasks” (2006: 189). The examples given above and many others show that
the study of translation has been vastly absent in the areas which deal with
these aspects of language which are intrinsic to the act of translating, bilin-
gual knowledge, bilingual memory, language control and language process-
ing and cognitive aspects of language acquisition and use. Nevertheless, De
Groot is quite emphatic in her conviction that studying translation has a lot
to offer saying,

All in all, I believe cognitive psychology should embrace translation as

an object of study. Doing so would be bound to increase our under-
standing of the human intellectual potential. In their turn, the new in-
sights in translation performance to be acquired from these intensified
research efforts could result in practical recommendations for both the
training and the professional practice of translators and interpreters
(1997: 31-32).

Indeed and what seems a welcome change translation as a valid area of

investigation has begun to feature in mainstream books and journals on
bilingualism. For example “The handbook of bilingualism” edited by
Kroll and De Groot (2005) includes in part IV ‘Aspects and Implications
of bilingualism’, section ‘Cognitive consequences’ a chapter, “Simultane-
ous interpreting: cognitive perspective” by Christoffels and De Groot.
Still, the article by De Groot and Christoffels (2006) is an exception
where psycholinguists look at translation (actually at simultaneous inter-
preting) as a human skill.
 Establishing common grounds 45

The above trace of interest in translation as a skill carries a promise of

what James Holmes advocated over 30 years ago when he called for a
collective effort, “teamwork between specialists in a variety of fields –
text studies, linguistics (particularly psycho- and socio-linguistics), liter-
ary studies, psychology, and sociology. And with the involvement of prac-
ticing translators” (Holmes/Broeck 1994: 101). Similar willingness to in-
tegrate findings and cooperate is voiced on the part of the TS community
(Jääskeläinen and Tirkkonen-Condit 1991, Tirkkonen-Condit and
Jääskeläinen 2000, Tabakowska 2001, Hejwowski 2004, Cronin 2005,
Kiraly 1997) although it remains largely singular.
In consequence there has been little scientifically valid cooperation be-
tween different research communities, i.e., applied linguistics, psycholin-
guistics, cognitive psychology and Translation Studies although there has
been a common practice of borrowing research methods or the object of
study. To illustrate, translation process studies borrowed the method of
thinking aloud from introspective psychology (Börsch 1986) and cognitive
psychology studied word translation to investigate the structure and organi-
zation of bilingual mental lexicon (Gollan and Kroll 2001). Yet, there have
been little if any cooperation and mutual exchange of the results and impli-
cations of empirical studies carried out in different areas. A 1997 publica-
tion entitled Cognitive processes in translation and interpreting edited by
Shreve et al. was an exception and a precious expression of the need for in-
terdisciplinary studies in translation process research but the issue of trans-
lation expertise development remained implicit and theoretically debated
rather than empirically explored. It is even possible that it is Translation
Studies which despite its interdisciplinary premises have shown more reluc-
tance for a possible synergy and willingness to draw from and share with
other disciplines. Possible reasons include the constantly growing aware-
ness of the complexity of issues involved in language processing in transla-
tion and the reluctance to try and decompose the process in order to see
how various sub-skills blend together in the multitasking intrinsic to a
translation activity. A more recent publication, and a follow-up to Cognitive
processes in translation and interpreting entitled Translation and cognition
(Shreve and Angelone 2010) is another attempt to encourage interdiscipli-
nary cooperation between cognitive studies and translation process studies.
Still translation expertise research remains marginal. On the part of Transla-
tion Studies, it remains to be acknowledged that applied linguistics in gen-
eral and more specifically its sub-disciplines such as psycholinguistics, so-
46 Chapter 1

ciolinguistics or neurolinguistics as well as discourse approaches to func-

tional linguistics, translation and language teaching/learning may have a lot
to offer to the field which ever since its beginnings has searched for more
feasible research methodologies.
It seems that what is really needed to study translation as a broad cluster
concept is a biodiversity of mind (Tymoczko 2005). It is generally agreed
by now that translation is a highly complex cognitive process embedded in
the social and cultural context but its underlying cause is in language barri-
ers which have to be removed via language processing at the level of its
physical (verbal) and conceptual (nonverbal) representation, which in-
volves an interplay of cognitive and psycholinguistic operations with many
non-linguistic aspects involved. Needless to say the final product, the mes-
sage which lifts the language barriers also arrives in the form of language.
We do need to look at what applied linguistics, psycholinguistics and cog-
nitive studies have to offer to understand “what takes place in the transla-
tion process” (Holmes 1988: 95-96), before we can know how translation
expertise develops. As it stands now over 20 years after Holmes’ seminal
work was published,

There are few if any “hints” for translators to follow; there are no com-
monly accepted inventories of techniques, strategies or procedures that
need to be acquired through classroom exercises and implemented in
clearly categorized problem situations; there are no translation methods,
which, if carefully learned and applied, will lead the translator to the
“right” solutions. Translation is the epitome of an ill-structured domain
– the translator’s personally mediated yet personal history of experience
makes as the interface for a succession of unique occurrences of inter-
cultural communication (Kiraly 2005a: 122).

The above view may sound harsh, but perhaps a harsh assessment is the
best way to provoke more research into translation expertise develop-
ment. At the time when “Translation is the Language of Europe”7, and
when it becomes more and more appreciated that translation has always
been the language of the world, what is needed is a collaborative effort
and open-mindedness to retrace the course of evolution of translation as a
human skill taking into account all possible manifestations of the human
ability to translate. The major responsibility in this respect lies on the part

–––––––––
7
Said once by Umberto Eco and frequently quoted in the context of the EU.
 Establishing common grounds 47

of Translation Studies which with no detriment to itself can become an

all-inclusive discipline devoted to the study of all forms and facets of
translation.

1.8. Conclusions

To sum up, this chapter was devoted to establishing common grounds.

The issues which were, subjectively, considered important for contextual-
izing the discussion on the genesis and evolution of translation as a hu-
man skill were presented and wherever possible supported by the opin-
ions of other scholars to make them a part of the ongoing debate among
scholars from the discipline of Translation Studies and other related disci-
plines. Translation as a complex cluster concept demands an open minded
broad perspective to recognize and acknowledge its various forms and
functions. Some effort was made to ensure the clarity of vital concepts
such as translation ability, translation skill and translation competence and
expertise, which are believed to be essential if one wants to avoid misun-
derstandings in the way these terms can be, and indeed have been used.
The major framework has been set out for the developmental perspective
in which the progression from translation as a predisposition, a mere po-
tential to translate shared by all human language users (also if the lan-
guage is non verbal like sign language or any other socially shared system
of communication) opens the developmental continuum. First the human
translating mind will realize this potential in the form of ability (usually
untrained) like in the case of natural translators to be discussed in chapter
3. Then the untrained ability may through repeated experience develop
into a skill which allows for a more fluent and more self-conscious per-
formance. Then the developing translator may decide to make a conscious
and deliberate effort to further the translation skill and progress towards
translation competence to be discussed in chapter 4. Finally following ex-
tensive practice the skill supported by competence may reach the level of
expertise. The consequences of the actual position on the developmental
continuum of translation as a human skill will be visible in the quality of
the translation performance. The distinction between the developmental
stages as well as the common lack of understanding of the development
which is needed to produce fully competent translation products meeting
professional quality standards results in the social misconceptions about
translation as a human skill.
48 Chapter 1

The framework is by no means new and it might even seem to have

always been a tacit assumption present in the work of Toury (1995), Gile
(1995), Kussmaul (1995), Kiraly (1995, 1997, 2002), Cronin (2005) and
many others involved in translator training as theorists and practitioners.
Unfortunately however, the above appreciation of translation skill devel-
opment in its continuity has not been taken up by researchers (PACTE
2003) and the genesis of translation as a complex professional expertise is
yet to be retraced and uncovered in its full evolutionary spectrum. A lot of
empirical and experimental effort is needed to perceive the skill of trans-
lating as an outcome of the social need and human desire to communicate
with others. It is my hope that looking at how the skill is built on bilingual
foundations and how it is demonstrated by those who have not received
any special training for it, and by those who are trained, or otherwise de-
velop translation competence and work as practicing translators, will
show the evolutionary continuity of translation expertise development. By
taking an empirical approach grounded on theoretical premises my aim is
not to de-mystify translation as a special skill reserved for the gifted few,
“the language elite” (Ivanova 1998: 92) but to draw attention to some as-
pects of translation expertise which if given more attention might help to
understand the seemingly mysterious nature of translation as a human
skill. In this approach I do not treat translation as an art or a craft but as
an intellectual skill to be researched and explored in its developmental
continuity. Like Mona Baker (1992) said in In other words, I believe that
professional translators should not be satisfied with saying that they trans-
late because they have a flair for it but just like other professionals they
should be able to explain what they do and how they do it as well as how
they have acquired and developed their translation expertise.
It seems natural to start the investigation with the bilingual founda-
tions which make translation proper possible by providing access to two
languages which one person is able to use for communication. Every
translator is a bilingual (either acquired or natural) who uses the potential
of the human translating mind to enable otherwise impossible communi-
cation when language and culture are a barrier which cannot be removed
otherwise than through translation. Chapter 2 will focus on these very
foundations, the bilingual knowledge of a potential translator with atten-
tion paid to the recent advances in the study of bilingualism, language and
cognition which are valid and revealing for sketching out the psycholin-
guistic profile of a translator-to-be.
Chapter 2

Whoever learns a new language becomes a new person

Rosi Landi (1973: 33)

Bilingual foundations of translation ability

This chapter will examine critical issues in the bilingual knowledge of a

translator-to-be. Since every translator is first a bilingual but not every bi-
lingual will choose to become a practicing translator a thorough insight
into the nature of bilingual knowledge is a necessary starting point. In the
evolutionary perspective on the development of translation expertise the
knowledge of two languages is only a starting point, a mere but essential
prerequisite. The purpose of this chapter is therefore to throw some light
on issues involved in the bilingual proficiency of a translator-to-be. The
issues raised in this chapter include: the nature of the linguistic knowl-
edge of L1 and L2, bilingual1 language users, interaction between two
languages in one mind, bilingual memory, language control and the cogni-
tive effects of bilingualism. Although it is generally agreed that sufficient
mastery of two languages does not guarantee that a bilingual person
(whether a natural bilingual or an L2 user) will make a good translator,
“bilingual competence constitutes a psycholinguistic foundation upon
which it is possible to develop translation competence” (Presas 2000: 20).

2.1. Bilingual foundations and Translation Studies

Although many issues are debatable in the development of translation as a

human skill at least one is certain and refers to all confused concepts dis-
cussed in chapter one. Translation ability, translation skill and translation
competence and expertise require the knowledge of at least two different
languages. Whether one calls it a mere prerequisite or a vital requirement
–––––––––
1
I use the term ‘bilingual’ in a broad liberal sense of anybody who can communicate
on an every day basis in two (or more languages) although I remain aware that it is not a
commonly shared view especially in Translation Studies. To quote, “The term ‘bilin-
gual’ is very much abused and the number of people who are truly bilingual is very
small” (Samuelsson-Brown 2010: 25).
50 Chapter 2

to translate, bilingual knowledge is the decisive factor to be able to con-

sider interlingual translation as an activity which one can perform. The
fact is so obvious that it has frequently been taken for granted together
with its consequences for the development of translation as a skill and
translation as professional expertise. As observed by Presas (2000: 27),
“comparative studies of monolingual and bilingual individuals reveal that
the acquisition of a second language involves the development of certain
cognitive features which are of interest for translation teaching”. Such ca-
pacities as lateral thinking, flexibility and the ability to make remote as-
sociations (Ben Zeev 1977, Appel and Muysken 1996) coupled with
“greater skill in handling the linguistic code, due to the fact that bilinguals
learn to separate the mental content of the lexical element from its graphic
or aural form at a very early stage” (Presas 2000: 27 see also Bialystok
2001) seem to be what is needed for translation as a human skill.
The standard requirement to enter translator training programmes is
that those who want to become translators or interpreters should have
mastered their second or foreign language (L2) to native-like proficiency.
The mastery of the first language is taken for granted and rarely given
more consideration. There are two striking consequences which underlie
the native-like proficiency requirement. First, it is implied that mastering
an L2 is a finite state, and secondly it assumes that every native speaker is
highly proficient in his or her L1 in the way that is best suited to acquire
translation skills (see chapter one). In view of the present knowledge
about language and bilingualism both assumptions are false. To quote,
“The individual’s first language, taken for granted in SLA2 research, is
complex and shifting. The L1 construct is an abstraction, a snapshot of a
moving target” (Cook 2007: 208). The linguistic knowledge of one’s L1 is
never finite and fixed so how then can the knowledge of L2 be expected
to be fixed and set as native-like proficiency?
Although today the general agreement points to the complexity of
translation as a cognitive process and many translation scholars do not fail
to acknowledge the bilingual foundations of translation competence not
much research has tried to draw insights from how the bilingual founda-
tions are laid and focused on the desired outcome: sufficient mastery of
the translator’s working languages (Presas 2000: 20). However, it is rarely
specified what such mastery in fact includes or admitted that achieving it
–––––––––
2
Second Language Acquisition.
 Bilingual foundations of translation ability 51

is a tough task. Taylor’s (1990: 1) observation may serve as an example:

“Translation or interpretation teachers may prefer to work on translation
or interpreting skills as separate entities from language ability, considered
as a sinc qua non but taken for granted”. Dodds (1999: 58) evaluates this
attitude as “an unrealistic position of convenience assumed by those of us
who would ideally delegate the laborious aspects of the job, i.e. teaching
the language, and be left with the more interesting task, i.e. translation
and interpretation”. This apt remark points to the mutual dependence be-
tween bilingual knowledge and translation as a human skill.
The relevance of bilingual knowledge for our understanding of the evolu-
tion of translation expertise is what is understood well in Machine Transla-
tion (MT) research, much less attention, however has been paid to it in Hu-
man Translation Research. The following quote can serve as an illustration,

Human crafted rules for creating an MT system capable of translating

any kind of text, for example, are considered to require an effort in the
order of 500 to 1,000 person years, and building a specialized bilingual
system (in the order of 10,000 concepts) would require approximately
100 person years (Ke Ping 2009: 167).

Have Translation Studies put enough effort into investigating the bilingual
foundations of translation competence? Are there many studies devoted to
the evolution of translation as a human skill or to that particular stage in
the process of evolution called bilingualism conducted from the perspec-
tive of a developing translator? Or do we still believe that translators are
born and not made (Nida 1981: 402, 2001: 89)?
Mona Baker in her introduction to In other words wrote:

if translation is ever to become a profession in the full sense of the

world, translators need something other than the current mixture of in-
tuition and practice to enable them to reflect on what they are doing and
how they do it. They will need, above all, to acquire a sound knowledge
of the raw material with which they work to understand what language
is and how it comes to function for its users (Baker 1992: 4).

Presas (2000) notices a gap in the way translation skill is viewed as de-
void of its bilingual foundations saying:

It is undoubtedly true that during the past decade interest in how learn-
ers learn has grown, and the resulting studies have shed much light on
52 Chapter 2

the problems faced by novice translators and on the strategies they em-
ploy to deal with them. However, it may well be that not enough atten-
tion has been paid to the root of the problems they experience, nor to
the bases of translation competence. I would suggest that both might
stem from their bilingualism (Presas 2000: 19).

The above quote is reflected in Walters (2005) view that looking into the
translator’s language acquisition history can provide revealing data for the
development of translation as a human skill. It is a trivial observation that
solid foundations make strong buildings. For a layman, foundations though
are something that one does not see and may be unaware of their vital im-
portance. For an expert, the quality of the foundations is essential in the long
run as it guarantees long-term benefits and in contrast to other elements of a
construction, it is something that cannot be changed without demolishing
the constructed building. To continue this simile it should be also added that
the way the entire structure is later on used and exploited will also bear con-
sequences on the overall condition of the foundations. In other words trans-
lation expertise is built on bilingual knowledge and there is a lasting rela-
tionship based on mutual benefit between them both throughout the transla-
tor’s developmental journey from a novice to a professional. For this reason
only bilingual knowledge deserves more detailed attention. However, the
acquisition and use of the second or foreign language has to take into ac-
count a wider picture of the nature of human language/languages so that the
processes and difficulties in the use of one’s L1 and L2 which are frequently
reflected in translation products become more tangible.

2.2. The nature of human language and the knowledge of two (or more)
languages

“Most people, including some highly educated ones, seem to have little
idea of just how complex language is” (Cattell 2006: 160). The question,
‘what does it mean to know two languages?’ evokes a more fundamental
question of ‘what does it mean to know a language?’ As a consequence all
the epistemological gaps in our understanding of linguistic knowledge
(Saleemi 2006: 13) will also refer to our understanding of bilingual knowl-
edge. Issues such as the relationship between thought and human language
(see Chomsky 2006, Pavlenko 2005, 2009) remain current challenges for
linguists (including psycho and neurolinguists) and language philosophers
alike. Yet, paradoxically as it may seem it is learning another language that
 Bilingual foundations of translation ability 53

has a revealing effect on one’s knowledge of the native language (see Ke-
sckes and Papp 2000). Without subscribing to any rigid theory of language
and going into unnecessary details, let us establish some basic assumptions
about linguistic knowledge which are believed essential when discussing
the bilingual foundations of translation expertise.
Human language is first and foremost a code subserving a powerful
need to communicate, that is to exchange information in order to achieve
some intended aim or in simple terms to get things done. It is a highly
complex system based on the shared knowledge of signs (letters, sounds,
words) and rules (grammar at all linguistic levels) for their combination
used to share meaning within a given language community. Language as a
code for communication or mind-sharing (Donald 2001) can, on the one
hand, be represented as knowledge (Chomsky 1964), on the other as so-
cial practice (Gardner 1979: 193, Pavlenko et al. 2001) which makes lin-
guistic forms inseparable from their socio-culturally established mean-
ings, connotations or innuendos. The complex nature of meaning in lan-
guage reflects this dual nature of human language, as knowledge and as
social practice. The linguistic knowledge of one’s native language is sub-
ject to socially agreed rules of linguistic behaviour (cf. Paradis 2009 and
his notion of ‘discourse grammar’) which form some kinds of conven-
tions marking out a socially conditioned network of mutual expectations
between users of the same language. Interpreting contextually relevant
meaning is a dynamic process of inferential comprehension (Sperber and
Wilson 1989) which apart from ruled governed word combinations and
the semantic values ascribed to individual words involves the extralin-
guistic factors of the whole communicative situation as well as one’s en-
tire cognitive repertoire of lifelong accumulated knowledge (Fauconnier
1998). To quote:

The more I think about language the more aware I become of all the
backstage cognition needed for understanding and meaningful use of
language. The forms become only a kind of prop, a powerful means of
prompting dynamic on-line constructions of meaning that go far beyond
anything explicitly provided by the lexical and grammatical forms
(Fauconnier 1998: 251).

This ‘backstage cognition’ is what one acquires together with the first
language and what evolves and becomes redefined throughout one’s life
parallel to the life experiences (Paradis 2007). Language in this context is
54 Chapter 2

primarily a tool for communication and the way the tool is used in actual
communicative situations is determined not only by linguistic competence
(implicit knowledge of grammar in the broad sense including phonology,
morphology, syntax and the lexicon) but by communicative competence
(Hymes 1971). Defined by Canale and Swain (1980) it includes four es-
sential components:

1. grammatical competence: words and rules

2. sociolinguistic competence: appropriateness
3. discourse competence: cohesion and coherence
4. strategic competence: appropriate use of communication strategies

It is not enough to know the rules and the words which can be combined
according to the rules to use language for communicative purposes. Since
the use of language is intentional (we speak when we want to speak) and
purposeful (it is used to achieve something, even if it is talking to one-
self), and mostly geared at interacting with others (including oneself as
other as in the case of talking to oneself), its use is governed by what
Paradis calls ‘discourse grammar’, the tacit socially conditioned rules of
language behaviour which guide the language user in the choice of appro-
priate linguistic means from the rich inventory offered by linguistic
(grammatical) competence. The pragmatic component is frequently deci-
sive for how successful a particular communicative encounter is. Yet,
there are other important components which contribute to the way lan-
guage is used to communicate with others, namely discourse competence
which refers to the speaker’s ability to structure an utterance or the entire
interaction, or a piece of writing. To guarantee, or at least safeguard a
successful exchange of information (even if the vague word successful
means in fact satisfactory for the participants) the language users can use
various communicative strategies which they themselves judge appropri-
ate in a given situation. As noted by Paradis (2009) the use of language is
intentional and driven by motivation, a strong desire to communicate with
others. The interplay of all the components of communicative competence
in language use usually remains so implicitly ingrained that many people
in everyday exchanges are hardly aware of it (Cattell 2006: 160).
In first language acquisition (L1) we acquire all the components de-
termining language use simultaneously and holistically. In our native lan-
guage all the aspects of one’s linguistic knowledge are acquired inciden-
 Bilingual foundations of translation ability 55

tally that is without conscious effort and remain encoded at a level which
for most is deeper than our conscious knowledge (see Paradis 2009). In
other words we follow the rules without being aware of them since when
acquiring our first language most communicative operations become
automatized to the point that we apply the rules in the form of procedures
without being aware of the complex computations that are behind them.
As pointed out by cognitive psychologists and sociolinguists the use of
language is a part of more complex cognitive structures and the proce-
dures of its use are encoded together with the prototypical scenarios of
usage. These situational concepts have been termed as frames, scripts or
schemas and constitute what Fauconnier (1998: 251) called the backstage
cognition essential in language use.
In second language acquisition (L2) excluding natural simultaneous
bilinguals, however the situation is completely different. When a mono-
lingual comes to learn another language, the complexity of language both
as knowledge and as social practice might appear overwhelming. Al-
though the research devoted to how people come to appropriate another
language has been branded Second Language Acquisition Studies (SLA)
some scholars including Paradis (2004, 2009) insist that the word acquisi-
tion should be reserved to native language as people do not acquire an-
other language but learn it making a conscious effort to internalize all the
relevant knowledge. Consequently, Paradis (2004, 2009) is emphatic
about recognizing the distinction between the procedural and declarative
status of all the knowledge components involved in language acquisition
and use. Linguistic competence is implicit and acquired incidentally that
is without conscious effort to acquire it and is represented by procedural
knowledge (knowing how) which is sustained by procedural memory. Na-
tive language users rely primarily on linguistic competence that is they
use language without being aware of all the computational procedures
which are involved in stringing words together to make meaningful utter-
ances. The other three components include pragmatic knowledge,
metalinguistic knowledge and motivation are also necessary for effective
communication and each component “relies on its own specific neural
substrate, which is susceptible to selective impairment” (Paradis 2009: x).
While linguistic competence relies primarily on the left hemisphere,
pragamatic knowledge (Paradis’s (2009) ‘discourse grammar’) which re-
fers to socially appropriate use of language and contains all the subtle
data which help language users choose the most fitting expressions for a
56 Chapter 2

particular situation is subserved by areas of the right hemisphere.

Metalinguistic knowledge defined as the more technical knowledge of a
language (Bialystok 2001) being conscious is declarative in nature (know-
ing that) and is sustained by declarative memory. Motivation, the drive to
communicate engages the dopaminergic system and will have a signifi-
cant effect on language performance “modulated by a range of affective
factors that result in great variability among second language learners”
(Paradis 2009: x).
The overlap with the components of communicative competence dis-
tinguished by Canale and Swain (1980) is significant: linguistic compe-
tence overlaps with their grammatical competence, pragmatic knowledge
with sociolinguistic competence, metalinguistic knowledge is utilized in
discourse competence and strategic competence and motivation drive the
strategic behaviour of a language user (see Whyatt 2009a).
What is important, and perhaps most important, in Paradis’s (2009) ar-
gument is the fact that the way people employ the four components is sub-
stantially different for one’s native language and for one’s second lan-
guage/languages. In the use of our native language if we are motivated to
use it (motivation is the primary requirement because language use is vol-
untary) we mostly rely on our linguistic competence and pragmatic knowl-
edge. Metalinguistic knowledge, the knowledge and awareness of language
as a system of communication, the ability to focus on how we use it and re-
flect on its formal properties is available, but not necessarily referred to in
the actual use of L1. When using L2 however, L2 learners (those who learn
another language for future use) and L2 users who according to Cook
(2002) use another language for everyday communication tend to rely pri-
marily on metalinguistic knowledge especially if the second language has
been taught by formal instruction and learned as explicit declarative knowl-
edge (knowledge that) Paradis in his recent book, Declarative and Proce-
dural Determinants of Second Languages (2009) explicates that the more
formal the language instruction the more L2 users rely on their knowledge
of the rules they should apply to make correct sentences. However, if the
teaching methods are more communicative and if the learners are provided
with practice in communicative situations they are more likely to develop
implicit linguistic competence. When however the L2 users experience
gaps in their L2 linguistic competence they will resort to their metalinguis-
tic conscious knowledge of the L2, or as proved by the transfer research
(Odlin 1989) they will use their L1 knowledge to find a solution in their
 Bilingual foundations of translation ability 57

communicative endeavor. If the L2 is not learned by classroom instruction

but, for example by immersion in the L2 culture (like in the case of immi-
grants), L2 users rely on their pragmatic knowledge and their metalinguis-
tic abilities when lacking L2 linguistic competence.
Although we are still far from understanding how the human mind
copes with two language systems (to be demonstrated further on in this
chapter) acquiring or learning a foreign language is possibly one of the
most common experiences in today’s world. Scholars agree that the num-
ber of people who can speak more than one language most probably
vastly exceeds the number of people who are monolingual (Cattell 2006).
Furthermore, the constant growing need to learn foreign languages (see
the EU directive for 2010) is stimulating the search for new more effec-
tive teaching methods to speed up the learning process and make its out-
come more satisfactory. In response to modern multilingual tendencies
Kesckes and Papp (2000) suggested that the Chomskyan (1986) question,
‘What constitutes knowledge of language?’ should be replaced by ‘What
constitutes knowledge of languages?’.
Carroll (1994) reminds that although individuals on average by the
age of five develop the competence of a ‘native speaker’ of their native
language. This kind of development,

takes precedence over the acquisition of skills in reading, writing, and

certain more specialized skills. Individuals tend to become differentiated
in levels of those other skills only at ages beyond the age of five or so. By
the time of adulthood, however, the individual differences in various spe-
cialized language skills can become quite pronounced, and substantially
independent of each other (Carroll 1994: 145-146).

These idiosyncratic differences in the way people use language and ex-
hibit different skills in for example making speeches or writing letters, or
showing more preference for listening rather than speaking when in the
company of other people are also important factors which somehow have
not been given much attention in SLA studies. It is possible that these dif-
ferences in L1 use might be relevant when studying those who choose to
study foreign languages and then perhaps become translators. Kesckes
and Papp (2000) suggested that one of the best ways of understanding
one’s native language is indeed to study a foreign language. To under-
stand one’s L1 and L2 seems essential for translating between the two
languages. Therefore, the question how the knowledge of two languages
58 Chapter 2

is organized in one mind is the first to ask in the study of the genesis and
evolution of translation as a human skill.
What does it mean to be bilingual in the sense of being able to com-
municate with ease in two different languages? What are the bilingual
foundations on which the translation skill can be built? Let us devote
some attention to a bilingual (or multilingual) individual as our potential
translator-to-be.

2.3. A bilingual person

For many people the term bilingual refers to a person born and brought up
in a bilingual family for whom using two languages is a normal way of
life. Referred to as true bilinguals, natural bilinguals or simultaneous bi-
linguals, they acquire two languages by ‘immersion’, i.e., by natural reac-
tion to the sounds made by its environment in order to communicate with
it (Thiery 1978: 146). Yet and most probably the majority of people who
are bilingual in the sense that they can communicate in their two lan-
guages learn their second language later on in life, either by immersion in
the L2 culture or, at school as children, or as adults. As observed by Ke-
sckes and Papp (2000) in their introduction to Foreign Language Mother
Tongue intensive foreign language learning is a special case of multilin-
gual development. In the modern understanding of bilingualism every-
body “including all individuals who actively use, or attempt to use, more
than one language, even if they have not achieved fluency in the second
language (L2)” (Kroll and De Groot 1997: 170) is a bilingual person.
Thiery’s (1978: 146) opinion that “no matter how well one speaks a lan-
guage, if he has learned it by tuition, he cannot be considered a true bilin-
gual” has been discarded. As observed by Beatens Beardsmore (1982: 8),
‘true bilingual’ in Thiery’s sense is in fact a rare if not a ‘non-existent
species’. “The condition of being in at least two languages appears in-
creasingly to be the natural condition of having any language at all. More
precisely, the concept that a human being might be confined to one lan-
guage appears increasingly to be a fiction” (Holquist 2003: 21). Instead
the prevailing view is that the knowledge of more languages is “the per-
fectly normal condition of the human mind” (Cook 1991: 115).
Webster’s dictionary gives a simple definition, ‘bilingualsim’ a noun
describes the quality of ‘being bilingual’ where ‘bilingual’means:
 Bilingual foundations of translation ability 59

a. Using or able to use two languages, especially with equal or nearly

equal fluency.
b. Using two languages in some proportion in order to facilitate learning
by students who have a native proficiency in one language and are ac-
quiring proficiency in the other: bilingual training; bilingual education.

Such a liberal definition shows that one is dealing with a complex phe-
nomenon (see Pavlenko 2005 on treating bilingualism as a monolith) and
the different routes that one follows to become a bilingual person will
bear consequences on the way bilingual knowledge is organized and used.
Parallel to different language acquisition histories, there will be differ-
ences in “oral and written proficiencies as well as daily patterns of lan-
guage use” (Haritos 2003: 1). Weinreich (1953), the unquestioned precur-
sor of recognizing the consequences of different routes to bilingualism di-
vided bilinguals into subordinate, coordinate and compound, referring to
the way they have come to acquire their two languages. Weinreich (1953)
believed that the route of L2 acquisition is reflected in the way the mental
lexicon is organized. Natural bilinguals by growing up in one reality
where two different labels (L1 and L2 word) are attached to the same
concept (or mental representation) are compound bilinguals. L2 learners
who acquire their L2 via their L1 are initially subordinate bilinguals (they
access their L2 via L1) or coordinate bilinguals (they develop separate
conceptual stores for each language) but with advances in their L2 profi-
ciency they also become compound bilinguals (De Groot 2002).
Although the term bilingual is now used in a more inclusive manner,
there are still some misconceptions about what it means to be bilingual. On
the one hand, the monolingual view of bilingual knowledge places high ex-
pectations of native-like mastery of both linguistic systems. On the other
hand, empirical research shows that the native-like mastery of the two lin-
guistic systems (at all levels of language use including pragmatics) is rare
even among natural or early bilinguals (Grosjean 1982, Paradis 2009). As
observed by Ewert (2009: 57), “The requirement that bilinguals be native-
like in their two languages led to a view of a bilingual as two monolinguals
in one person, with two separate linguistic systems. As the proficiency re-
quirements were relaxed and wider segments of the human population be-
gan to be studied by bilingualism researchers, it became obvious that a per-
fect balance between the two languages is extremely rare”. Pavlenko
(2005) reminds that Green (1998) cautioned against approaching all bilin-
60 Chapter 2

guals in the same way because they may have different levels of expertise
and different competences in their two languages (Pavlenko 2005: 437). To
quote,“some researchers treat bilingualism as a monolithic phenomenon
and thus do not pay attention to linguistic trajectories of their study partici-
pants” (Pavlenko 2005: 437). These misconceptions do not only refer to a
layperson’s view but also bear consequences for the worldwide research
community. Therefore it is essential for those who study bilinguals to de-
fine the bilingual status of their study participants as they may differ con-
siderably in terms of their language expertise in their two languages which
may affect research results and lead to distorted conclusions (see Paradis
2005, Green 1998, Grosjean 2002, Pavlenko 2005). This word of caution is
usually safe-guarded by researchers describing their subjects as ‘balanced
bilinguals’ – bilinguals who have roughly equivalent abilities in the two
languages (Malakoff and Hakuta 1991). Others use quite vague terms, for
example, some researchers describe their subjects as ‘more or less fluent bi-
linguals’ or ‘somewhat more fluent bilinguals’. As pointed out by Heredia
(2008), without an objective measure to classify bilinguals according to
their second language competence or knowledge interpreting research re-
sults becomes problematic. “To date, there is no universally accepted
method of assessing bilingual proficiency” (Haritos 2003: 1). It is generally
undisputable that the major factor in establishing one’s bilingual knowledge
is the history of the process in which it was acquired.
The recognition of different routes to bilingualism draws attention to
the different starting points a prospective translator can set off from,
which will bear consequences on how his or her bilingual knowledge is
organized prior to the intentional development of translation ability to
reach the level of expertise. Yet, Translation Studies as a discipline has
not devoted due attention to the nature of bilingual knowledge or the lan-
guage acquisition history of potential as well as practicing successful
translators. Let us now consider natural bilinguals and L2 learners and us-
ers as prospective translators with the aim of throwing some light on the
intricate relationship between bilingual knowledge, language acquisition
history and the human ability to translate.

2.3.1. Natural bilinguals and translation ability

According to common expectations a true or natural bilingual is expected

to be equally fluent or ‘nearly equally fluent’ in both languages. Beatens
 Bilingual foundations of translation ability 61

Beardsmore (1982: 8) referred to such a person as an ‘ambilingual’, that

is a person “capable of functioning equally well in either of his languages
in all domains of activity and without any traces of the one language in
his use of the other”.
As cautioned by Grosjean (1982), himself a natural bilingual and a
scholar committed to the study of bilingualism, this layperson’s monolin-
gual view of bilingualism does not match the reality. Bilingual in his opin-
ion is a person who uses two languages in every day life and not a person
who knows the two languages equally well. According to Grosjean’s (2002)
Complemantarity Principle, it is essential to remember that a bilingual uses
his/her two languages (separately or together) for different purposes, in dif-
ferent domains of life and with different people. Because the needs and
uses of the two languages are different, a natural bilingual is in fact rarely
equally or completely fluent in the two languages (Grosjean 2002: 2). For
this reason, a bilingual should be perceived as “an integrated whole, a
unique and specific speaker-hearer, and not the sum of two monolinguals”
(Grosjean 1997: 163). “He or she has developed competencies (in the two
languages and possibly in a third system that is a combination of the first
two) to the extent required by his or her needs and those of the environ-
ment” (Grosjean 1997: 163). Therefore as suggested by Grosjean, it is im-
portant to view the use of both languages in terms of language modes
(Grosjean 2001) forming a conceptual continuum between a monolingual
mode (when a bilingual uses one of the languages) and a bilingual mode
(when both languages are partially or entirely involved, e.g., as in language
mixing, code switching and translation). This holistic view of bilingualism
stands in contrast to “the public misconceptions of bilinguals” being two
monolinguals in one mind. As Grosjean (2002) explains this monolingual
view of bilingual knowledge on the one hand makes many natural bilin-
guals feel inadequate and on the other the implied expectations of equal
language proficiency in both languages make fluent bilinguals, who have
acquired their second language via formal instruction, feel inferior and
unlikely to call themselves bilingual. Grosjean’s position is firm, “bilingual
is not the same as two monolinguals in one” (Grosjean 1997). To quote,

The reasons that bring languages into contact and hence foster bilingual-
ism are many: migrations of various kinds (economic, educational, politi-
cal, religious), nationalism and federalism, education and culture, trade
and commerce, intermarriage, etc. These factors create various linguistic
needs in people who are in contact with two or more languages and who
62 Chapter 2

develop competencies in their languages to the extent required by these

needs. In contact situations it is rare that all facets of life require the same
language (people would not be bilingual if that were so) or that they al-
ways demand two languages (language A and B at work, at home, with
friends, etc.). This leads to what I have called the complementarity prin-
ciple which I define as follows: Bilinguals usually acquire and use their
languages for different purposes, in different domains of life, with differ-
ent people. Different aspects of life normally require different languages.
It is precisely because the needs and uses of the languages are usually
quite different that bilinguals rarely develop equal and total fluency in
their languages. The level of fluency attained in a language (more pre-
cisely, in a language skill) will depend on the need for that language and
will be domain specific (2002: 4).

This empirically grounded understanding of how natural bilinguals use

their two languages disqualifies the ‘public’ expectation of equal mastery
of both linguistic systems and gives grounds to claim that because natural
bilinguals use their two languages in different communicative settings the
range of the vocabulary in both languages may in fact be substantially dif-
ferent or even poorer in some domains than in case of a monolingual per-
son. This consequence does not create favourable conditions for their
ability to translate. Translating in a professional sense requires rich vo-
cabulary in both languages as well as adeguate control to keep the two
languages clearly apart to avoid crosslinguistic interference. Nida (2002)
in his paper entitled “Translator’s confrontations with false ideas about
language” given in the European Commission states:
Too often people assume that the best translators are fluently bilingual
and bicultural, but actually this is not always true. In fact, some people
who are constantly shifting back and forth between two languages are not
always the best translators because they are not fully sensitive to what is
precisely the most fitting expression in a particular receptor language.
The best translators are those who are fully sensitive to usage in their own
mother tongue and extremely well informed about the cultures of the
source language from which they translate. Knowledge of the source cul-
ture often proves more important than linguistic expertise (Nida 2002).

The view that natural bilinguals may not make the best translators, how-
ever, does not mean that they are not able to translate. They are usually
very much appreciated natural translators (chapter 3 will explore these is-
sues) and their natural ability can of course be intentionally developed
into translation expertise (the focus of chapter 4). Although I am not
 Bilingual foundations of translation ability 63

aware of any statistical data concerning the ratio of professional transla-

tors who are natural bilinguals and those who learned their second (Lan-
guage B) or third (language C) working language in a foreign language
classroom, I am inclined to expect that the majority of practicing transla-
tors of today are in fact former language learners. If research shows that
even in natural bilinguals, and contrary to common expectations, there is
an imbalance between the two languages, it can be expected that this im-
balance will be more pronounced among those who learn their second
language and at least in theory can become translators.

2.3.2. Bilingual competence of L2 learners as potential translators

The majority of those who are not born in bilingual families gain access to
their second language later on in their life (Grosjean 1982, Hamers and
Blanc 1990), and most frequently after they have acquired, at least partially
their native language (L1 or language A in TS terms). Originally referred to
as coordinate bilinguals (Ervin and Osgood 1954, Weinreich 1953) depend-
ing on the onset of their L2 acquisition they can be classified as ‘early ver-
sus late’ bilinguals (Genesee and Nicoladis 2006, Paradis 2009).
People come to acquire communicative skills in their L2 in the proc-
ess of learning. It is generally acknowledged that learning a foreign lan-
guage to achieve a high level of proficiency (native-like standards of
communicative competence) takes a lot of time and effort on the part of
the learner and on the part of the teachers (see Whyatt 2009b). It has be-
come clear that whether via formal teaching and learning in schools or via
unstructured exposure to the way L2 is used, the learning process aims at
internalizing all the components of communicative competence. It is
somehow less frequently acknowledge that in fact the aim of L2 teaching
and learning is developing an extremely complex skill of communicating
in another language.
According to cognitive theories of L2 acquisition, “learning a lan-
guage entails a stage wise progression from initial awareness and active
manipulation of information and learning processes to full automaticity in
language use” (O’Malley and Chamot 1990: 217). In the information
processing model L2 learning is viewed as “the movement from con-
trolled to automatic processing via practice (repeated activation)” (quoted
after Mitchell and Myles 1998: 86). This progression from controlled to
automatic processing “results in a constant restructuring of the linguistic
64 Chapter 2

system of the L2 learner (…). When the shift occurs, controlled processes
which consume a large share of the students’ attentional resources are
freed to deal with a higher level of processing” (ibid.). In a similar man-
ner, Anderson (1980/2005, 1996) in his ATC (Adaptive Control of
Thought) model views learning a second language as a process through
which declarative knowledge about the language (knowledge that) be-
comes proceduralised into procedural knowledge (knowledge how) via
practice. To quote Anderson (1980),

When we learn a foreign language in a classroom situation, we are aware

of the rules of the language, especially just after a lesson that spells them
out. One might argue that our knowledge of the language at that time is
declarative. We speak the learned language by using general rule-
following procedures applied to the rules we have learned, rather than
speaking directly, as we do in our native language. Not surprisingly, ap-
plying this knowledge is a much slower and painful process than applying
the procedurally encoded knowledge of our own language. Eventually if
we are lucky, we can come to know a foreign language as well as we
know our native language. At that point, we often forget the rules of the
foreign language. It is as if the class taught declarative knowledge had
been transformed into a procedural form (Anderson 1980: 224).

Paradis (2009) is of a different opinion and claims that a clear distinction

should be made between the way we acquire our L1 and learn our L2.
While the knowledge of L1 is indeed procedural, the knowledge of L2 is
declarative and consciously learnt knowledge. In his view if a second lan-
guage is learned after the first language has been acquired, it will not re-
sult in implicit linguistic competence comparable to that of one’s native
language. Depending on the teaching method the knowledge of the sec-
ond language will result in the different application of the four compo-
nents of communicative competence in L2 production. If the teaching is
carried out via formal instruction, the learner’s declarative memory will
be involved and the learner will gain metalinguistic knowledge of the
facts about language (for example, that double negation is not allowed in
English). When the teaching methods applied are more direct and com-
municative, the learner’s procedural memory will be engaged and it may
result in acquiring implicit linguistic competence. Although Paradis does
not mention it, it can be expected that those people who learn their second
language by immersion (direct contact with the language as in the case of
immigrants) will with time develop some kind of implicit competence but
 Bilingual foundations of translation ability 65

in their L2 use will most likely rely on pragmatic knowledge. Undoubt-

edly in both routes of learning an L2 motivation is central to the ‘appro-
priation’ (term used by Paradis to refer to conscious L2 learning) and use
of second languages, and when present it will improve performance.
Since the implicit linguistic competence of L2 learners and users (I refer
to Cook’s (1992) distinction here) is very much incomplete and may even
be non-existent, when using the L2 speakers rely on metalinguistic
knowledge and pragmatic knowledge to compensate for the gaps. Al-
though the role of L1 in the process of learning an L2 has not been dis-
cussed by Paradis, it is very likely that L1 competence is also used to-
gether with metalinguistic knowledge to facilitate L2 performance at least
at the very early stages of foreign language learning. Since linguistic
competence and metalinguistic knowledge are subserved by different
kinds of memory they can be used simultaneously, e.g., metalinguistic
knowledge can be used to monitor L2 performance, spot production errors
and self-repair them. If the L2 users rely on their metalinguistic declara-
tive knowledge they use their L2 in a controlled (conscious) way, if the
L2 learners rely on their implicit procedural knowledge they use the lan-
guage in an automatic (not open to conscious control) way. As put by
Paradis (2009: x), “Practice will either speed up controlled processing or
promote implicit competence (or both, to different extents and at different
times)”.
Although the question of the automaticity of L2 knowledge has been
challenged by Paradis (2004, 2009) other scholars make slightly different
predictions about the so called ultimate attainment and automaticity in L2
production. Kroll and Linck (2007: 250) while agreeing that “increased
speed and accuracy alone does not indicate increased automaticity” refer
to Logan’s (1988) instance theory of automaticity according to which the
increased efficiency of processing can be explained by the memory re-
trieval process. According to Logan (1988) processing can take two
routes. One involves the computation of every step using some kind of al-
gorithm. Every computation however leaves a memory trace of the proc-
ess. The more frequently the computation is performed (increased prac-
tice) the stronger the memory trace is left. Future processing thus does not
have to compute the individual steps but involves the memory trace of the
steps that have to be performed. In this so called ‘race model’ automatic-
ity “is considered to be attained when the memory retrieval process is
completed more quickly than the algorithmic process, thereby increasing
66 Chapter 2

the efficiency of processing” (Kroll and Linck 2007: 250). This explana-
tion challenges Paradis’s skeptical view of automaticity in L2 and opens
up new avenues to investigate the development of cognitive procedures
which enable fluent performance in L2. Indeed,
even if we suspect that we have no chance to ever achieve near native
competence, all of us would probably still set it as our ideal goal, and
try to draw as near to it as possible. When seriously learning a foreign
language, we work on our accents, memorize grammatical patterns, try
to learn as many words as possible, try not to fall for false friends and
to avoid interferences. Taken together, this clearly means that bilingual
competence is what we ultimately aim for in foreign language learning
(Hentschel 2009: 16).

Needless to say, it is an aim which is achieved by few and it is possible to

speculate that perhaps better results in terms of L2 proficiency could be
achieved if the foreign language teaching methodology recognized the
dynamic nature of bilingual development in which there is a constant in-
teraction between the learner’s two languages (Cook 1992). It is possible
that this dynamic interaction, the ongoing battle for dominance between
the learner’s two languages if brought to the attention of the learner him-
self/herself could prove helpful in developing L2 proficiency (Witte
2009). However, L2 teaching methodologies still keep trying to make L2
learners acquire L2 just like they came to acquire their mother tongue.
The communicative or direct method, for example completely disregards
the fact that L2 learners are already L1 users in the belief that eliminating
L1 from the L2 learning context will enhance the development of linguis-
tic competence in L2. All is done in the hope that the use of L2 will be-
come as automatic as the use of L1 and therefore as native as for native
language users. This aim is frequently shared by L2 pedagogy and L2
learners themselves. Only recently, more scholars point to translation
tasks as a potential ally in engaging the learner’s both languages for the
benefit of enhancing their bilingual knowledge in its entirety where it in-
cludes much more than the linguistic knowledge (see chapter 3 for a de-
tailed discussion). It seems that the human ability to translate can indeed
be seen as the fifth skill of a bilingual person who can: speak either of the
two languages, write in both languages, read and exhibit listening com-
prehension as well as translate between the two languages. All the skills
are determined by the level of L2 proficiency. Translation, though in con-
trast with the other skills involves simultaneous conscious activation of
 Bilingual foundations of translation ability 67

the two languages and therefore it can be expected to be determined by

L2 proficiency and by other capacities to be discussed in the following
chapters.
To sum up, irrespective of the course of L2 acquisition or learning, all
bilinguals can use either of their two languages for communication, ex-
perience interference and are able to translate between them (Fabbro
1999). Recent psycholinguistic experiments and neurolinguistic studies
confirm that both languages remain active in the bilingual mind even if
only one is actually being used (Green 1993, Thierry and Wu 2007). Sec-
ond Language Acquisition Theories have continuously attempted to ex-
plain the nature of this unique co-existence of two languages in one mind.
The dynamic interaction between the two languages in the bilingual mind
is of interest in the discussion of translation as a human skill as it will al-
low to understand the translation process which on the one hand, involves
the conscious simultaneous activation of both systems and on the other
hand imposes a requirement to keep both systems apart in order to avoid
(or control) interference. Looking at the dynamic co-existence between
L1 and L2 throughout the process of L2 acquisition provides a deeper in-
sight into the nature of the bilingual knowledge of a potential translator
and is to some extent determining the course of evolution of the human
skill to translate.

2.4. The interaction of two languages in one mind – an L2 learner’s per-

spective

The question how the knowledge of two different linguistic systems is ac-
commodated in one mind is one of the most intriguing ones. Answering it
is most likely to bring a worldwide revolution in teaching and learning
foreign languages but for the time being there are only some speculations.
As pointed out by Paradis (2005: 411), however although research has
provided some interesting insights since Albert and Obler (1978) formu-
lated a number of fundamental questions for bilingualism research, we are
still probably quite far from being able to provide a complete understand-
ing of bilingualism and multilingualism.
In the context of L2 teaching via formal instruction the relationship
between the two languages is best perceived in terms of progressive dy-
namic interaction. The course of the way in which the two languages
will influence each other is determined by several factors. The two most
68 Chapter 2

frequently described include the bilingual age (hence the distinction be-
tween early and late bilinguals) and the teaching methods (explicit in-
struction versus more communicative methods). An extended list of fac-
tors could also include the level of L2 socialization, everyday contact
with the L2, frequency of L2 and L1 use, purpose of using L2, motiva-
tion, future prospects, etc. The dynamic nature of the way the native and
the foreign language influence each other has been discussed by SLA
scholars and psycholinguists and the debate over L1 – L2 interaction has
itself evolved from a conservative perspective into a more liberal ap-
proach. Although the “[g]hosts from the past like ‘language interfer-
ence’, ‘negative transfer’, etc. still make their presence felt” (Witte et al.
2009: 3) when the native language is mentioned in the foreign language
classroom, more politically correct taxonomy is being used, like ‘cross-
linguistic influence’, or ‘language contact’. The awareness of the cross-
linguistic interaction, no matter whether it is expressed in the form of
transfer or interference and understanding the mechanism behind it is
crucial for understanding the bilingual foundations of translation exper-
tise. Let us briefly review the nature of the dynamic L1 and L2 interac-
tion as revealed by transfer research.

2.4.1. Cross-linguistic interference

The concept of interference or cross-linguistic influence (Jarvis and Pav-

lenko 2008, Pavlenko 2009) is crucial in understanding the interaction be-
tween two or more languages in the learner’s mind. Interference can be
present at all levels from phonology and phonetics (articulation) through
lexicon and grammar (formulation) to the level of discourse (usage) in L2
performance. As put by Green (1993: 270), “Performance in L2 reflects
the nature of the system that has been constructed” in the mind of the L2
learner. The increasing level of proficiency in L2 is usually accompanied
by a decrease in L1 interference. Originally and in most general terms the
word transfer was used to show that L2 learners when unable to cope with
the complexities of their L2 use will use their L1 knowledge (Andersen
1983). The phenomenon of language interference stigmatized in the for-
eign language classroom for obvious reasons shows the dynamic interac-
tion between the two languages in a bilingual mind and although it is fre-
quently noticed in the form of negative transfer from L1 to L2 the reversal
is also possible although it has become a focus of interest only quite re-
 Bilingual foundations of translation ability 69

cently (Ewert 2009). Kesckes and Papp (2000) insist that both L1 and L2
affect each other in the sense that some aspects of L1 or L2 knowledge
can be transferred to each other.
As pointed out by Odlin (1989) it is not only the similarity between
the two languages that causes transfer (see Selinker 1992: 260 on Wein-
reich’s ‘interlingual identifications3’) but transfer can be also due to dif-
ferences between the learner’s two languages (Kellerman 1995). Although
as pointed out by Odlin (1989) a lot of transfer research focused on mor-
phology, syntax and lexis, the phenomenon is present at all levels of lan-
guage as well as at the underlying conceptual level (Pavlenko 2009). In
brief as noted by Kesckes and Papp (2000) transfer occurs at two levels,
structural where it affects the units and patterns of language and at the
higher conceptual level where it affects pragmatics and knowledge in
general and results in transferring discourse patterns, forms of address and
other language specific patterns of behaviour as well as L1 and L2 spe-
cific concepts. Yet, as claimed by Pavlenko (2005, 2009) pragmatic and
conceptual transfer has not received sufficient attention although it fre-
quently leads to more serious cross-cultural misunderstandings than for
example, lexical transfer (Arabski 2006). Translation practice has a rich
store of mistranslations caused by conceptual and pragmatic transfer
(Korzeniowska and Kuhiwczak 1994, Korzeniowska 1998, Schäffner
1998, Whyatt 2006c).

2.4.2. Three perspectives on bilingual coexistence in a single mind

In early SLA theories, e.g., Selinker’s (1972) interlanguage theory, trans-

fer and more precisely negative transfer was perceived only in one dimen-
sion as an L1 interference into L2 and has been stigmatized as leading to
errors (Sridhar 1994: 802). Transfer errors were believed to be typical of
the learner’s interlanguage, defined as a separate system, in between L1
and L2. Since all the systems L1, L2 and interlanguage were seen as sepa-
rate the interlanguage theory did not acknowledge the constant bidirec-
tional interaction between the learner’s L1 and L2. This clear cut isolation
of the language systems in the learner’s mind has become increasingly
questioned by researchers from various disciplines including cognitive
studies, psycholinguistics and neurolinguistics (see Cook 1992).
–––––––––
3
The identification of two items as the same across linguistic systems.
70 Chapter 2

Cook (1992) returning to Weinreich’s idea of the coordinate versus

compound distinction formulated the multicompetence theory noting that,
“[s]ince the first language and the other language or languages are in the
same mind, they must form a language super-system at some level rather
than be completely isolated systems” (Cook 2003: 2). The learner at the
first contact with the L2 perceives both L1 and L2 as separate systems
(coordinate) but in the process of learning L2 both languages come to in-
teract with one another along what Cook called the integration continuum.
The integration continuum demonstrated in Fig. 1 shows the possible rela-
tionships languages enter in the multicompetence theory.

L1

L1

L1
L2
L2 L2

Separation interconnection integration

Fig. 1. The integration continuum of possible relationships in multicom-

petence (adapted from Cook 2002: 11).

Multicompetence which was originally defined as a “compound state of a

mind with two grammars” (Cook 1991: 112), and redefined later on as
“the knowledge of more than one language in the same mind” (Cook
2002: 10) has an interesting contribution to offer in terms of dynamic
cross-linguistic interaction. First, as stressed by Cook multicompetence is
a different state of mind than monocompetence, second, the two lan-
guages are perceived as interconnected to a greater or lesser extent in the
mind of the L2 learner and this interdependency will bear consequences
for the way both languages are processed in one mind. To quote, “switch-
 Bilingual foundations of translation ability 71

ing from speaking the L1 to speaking the L2 may not be taking one
grammar off-line and replacing it with another, but weighting the parts of
one overall system, i.e., ‘wholistic multicompetence’” (Cook 1991: 115).
The relationships between the languages in multicompetence can vary
along several dimensions (Cook 2002: 12-13):

– Different relationships can apply to different areas of language.

– The relationships might change according to the stage of development
or of language attrition.
– The closeness of languages may affect the relationship.
– The relationship might vary from one person to another (quoted after
Ewert 2009: 37)

Pointing out the factors which may affect the level of integration/separation
between L1 and L2 will have consequences on how the L2 user processes
both languages also when translating between them. Hall, Cheng and Carl-
son (2006) criticized Cook’s model on three grounds (for details see Ewert
2009: 37): “(1) a view of L1 and L2 language knowledge as distinct sys-
tems; (2) the presumption of a qualitative distinction between multicompe-
tence and monocompetence; and (3) the assumption of homogeneity of
language knowledge across speakers and contexts” (Hall, Cheng and Carl-
son 2006: 220). Instead, Hall, Cheng and Carlson (2006) claim that the two
languages of a bilingual form one system in which forms belonging to dif-
ferent languages are nothing more than contextual variants.
In their reconceptualised multicompetence model Hall, Cheng and Carl-
son (2006) shift focus from the cognitive to a social-interactional perspec-
tive (Firth and Wagner 1997, Firth and Wagner 2007). All language knowl-
edge is always “provisional and sensitive to renegotiation and renewal”
(2006: 230). The dynamic view of language knowledge refers to all the lin-
guistic codes a person might know and it is essentially experience derived
knowledge. In this respect monolingual and bilingual speakers differ accord-
ing to the differences in their language experience. This view reminds very
much of Grosjean’s Complementarity Principle discussed in the section on
natural bilinguals. The quote that “all language knowledge is socially con-
tingent and dynamic no matter how many language codes one has access to”
confirms the analogy (Hall, Cheng and Carlson 2006: 229). Different lan-
guage experiences lead to differences in knowledge referring to L1 and L2.
The language knowledge of individuals is perceived as “dynamic constella-
72 Chapter 2

tions of linguistic resources” (Hall, Cheng and Carlson 2006: 226). There is
no distinction between linguistic competence and performance:

rather than a prerequisite to performance, language knowledge is an emer-

gent property of it, developing from its locally-situated uses in culturally-
framed and discursively patterned communicative activities. Language
structures, as conventionally conceptualized, are simply post-hoc observa-
tions of the continually shifting patterns and schemas we employ to nego-
tiate specific contexts of action (Hall, Cheng and Carlson 2006: 228).

Both perspectives on the constantly dynamic relationship between L1 and

L2 knowledge in the mind of the L2 learner allow to see bilingual knowl-
edge as linguistic resources which are at the language learner’s/user’s (or
translator’s) disposal. The assumption that every language experience will
have a reciprocal effect on the knowledge structures is in line with the
cognitive principles. Learning an L2 via practice (repeated activation)
“results in a constant restructuring of the linguistic system of the L2
learner” (McLauglin and Heredia (1996) quoted after Mitchell and Myles
1998: 86). However, it is practice in communicative situations which in-
ternalizes grammar by placing on the user certain requirements to comply
with the rules of discourse these communicative situations are a part of
(cf. Paradis 2009). As pointed out by Ewert (2009: 43), “[t]hese inher-
ently dynamic and variable constellations of linguistic resources are sub-
ject to stabilizing forces of societal norms that value constancy”. Thus,
“stability in language knowledge has its roots in socioculturally contextu-
alized activity” (Hall, Cheng and Carlson 2006: 229).
The dynamics in bilingual knowledge is further developed by employ-
ing the mathematical theory of dynamic systems which change over time
(van Geert 1994: 50). Herdina and Jessner (2002) formulated a Dynamic
Model of Multilingualism (DMM) in which there is room for all aspects of
the co-existence of two languages in one mind. First, the two languages of
a bilingual person are seen as systems nested in a larger multilingual sys-
tem which itself is nested in the system of human cognition. The person
who is learning a foreign language is also a dynamic system in interaction
with his or her social environment. The different linguistic systems of a bi-
lingual speaker are interconnected and interdependent and therefore subject
to bidirectional transfer at all levels of language from phonology to dis-
course. Recent studies in the structure of the bilingual mental lexicon (to be
discussed below in more detail) point to a more complex interaction at the
 Bilingual foundations of translation ability 73

structural and what was previously disregarded at the conceptual level

(Kroll and Linck 2007: 246). It seems that in the language learner’s mind
there is a constant battle for dominance between the two language systems
and even very proficient L2 users and learners cannot switch their L1 off.
In this battle it is the language which is most frequently used that will be
more readily available to the point that the actual label L1 and L2 do not re-
flect language dominance but only the order of language acquisition. As
observed by Heredia (2008) in the case of immigrants, for example it is in
fact the L2 that is dominant and frequently the L1 by not being used falls
into attrition. This perspective points to the social factors which condition
the interaction between both languages and general cognitive processing in
a language user who is constantly engaged in culturally situated (either in
L1 or L2) social interaction (Ewert 2009: 46).
Following the recognition of the interdependencies connecting bilin-
gual knowledge in the L2 learner’s/user’s mind, the term language inter-
ference or negative transfer is replaced by more liberal terms like cross-
linguistic influence taking off in this way the blame from the L2
learner/user for being unable to suppress interference. Other scholars, e.g.,
Kesckes and Papp (2000) keep using the word transfer as the best descrip-
tion of the actual processes of L1 and L2 interaction. In this work I use all
three terms: influence, interference and transfer as describing instances of
cross-linguistic interaction.
To sum up answering the question whether bilingual knowledge is a
single system or two partially integrated systems (Ewert 2009: 39) is not
essential for recognizing its importance as the foundations of translation
ability, competence and expertise. What is the most significant aspect of
the bilingual knowledge of potential translators is its dynamic nature and
the constant interaction of the two linguistic systems in one mind. This in-
teraction or cross-language competition (Kroll and Linck 2007) is what
bilinguals experience all the time. As put by Grosjean (2001)4 they move
between their monolingual and bilingual modes and choose the mode they
will communicate in according to the socio-pragmatic context. Sometimes
they will fail to control interference, sometimes they will code-switch or
blend and mix languages. However, what is natural and accepted in eve-
ryday communication is not acceptable in professional translation. The
first contrastive requirement is that professional translation places definite
–––––––––
4
See also De Groot and Christoffels (2006).
74 Chapter 2

demands on keeping the two languages apart and is very much intolerant
to cross-linguistic interference. This implies that a kind of control over the
bilingual’s two languages needs to be operating in the bilingual mind.
Séguinot (1997: 117) observed that “interferences occur all the time be-
tween semantically and graphically related words both within and across
languages. Some mechanism is required to suppress unwanted connec-
tions as well as activate those that are required…”. The implication of
such a mechanism points to the need for cognitive control in bilingual
language performance (Rodriguez-Fornells, et al. 2006) which would en-
sure the retrieval of information relevant for the task to be performed
(e.g., a target language word) while suppressing all the irrelevant informa-
tion (e.g., a source language word) in the long term memory. The fact that
the amount of negative transfer especially at the structural level decreases
parallel to growing L2 proficiency while the amount of positive transfer at
the conceptual level increases points to the increased control over the two
systems. Consequently, developing the skill of translation will involve
learning to control undesired interference and keeping the two languages
safely apart. This kind of cognitive control over bilingual knowledge does
not come easily and certainly not without the conscious awareness of
cross-linguistic interference as a part and parcel of bilingual language
processing. The notion of language control suggested by Green (1986)
deserves a more detailed discussion as an important aspect of language
processing for a prospective translator.

2.5. The notion of language control

Although the leading question behind Green’s model of language control

(IC Model – Inhibitory Control Model) was to describe how the mind
knows which language to use in a particular moment (see Penfield 1953
and Penfield and Roberts 1959 on the language switch mechanism5), the
issue of control can be perceived as global (inhibiting one language while
using the other) or local, e.g., inhibiting some outputs of the system while
activating others, as for example in choosing one lexical item out of sev-
eral competing equivalents in language production or in the process of

–––––––––
5
Contrary to Penfield and Roberts (1959) and their simple language switch mecha-
nism the model presented by Green is a variant of the model proposed by Morton (1980)
and was initially restricted to the recognition and production of words in two languages.
 Bilingual foundations of translation ability 75

translation (see De Groot and Christoffels 2006). Such a perception of

language control is important for the bilingual’s ability to translate which
involves constant switching between the two languages as well as control-
ling interference. As stressed by Green (1993) language control will use a
considerable amount of mental resources therefore it is possible that it
contributes to the overall mental effort needed to translate. The high en-
ergy demands are pronounced in oral translation and especially tangible in
simultaneous interpreting. Although there are obviously also other factors
influencing high energy expenditure in interpreting, the need for language
control is a contributing factor which however has not received much at-
tention in the empirical studies of translation and interpreting.
Green’s model rests on the assumption that for a bilingual speaker al-
though only one language is selected the other one nevertheless is not
completely deactivated (Ervin and Osgood 1954; Preston and Albert
1969). This dual activation is believed to account for the fact that bilin-
guals take longer to name objects (Altenberg and Cairns 1983 on lexical
decision tasks) and are prone to errors of interference (Grosjean 1982)
which to be avoided place sometimes heavy demands on the cognitive re-
sources of the bilingual speaker and are difficult to avoid especially when
the resources are taxed by stress or fatigue (Dornic 1980, Clyne 1980).
According to Green (2000: 376) normal fluent speech can be seen as a
successful avoidance of errors and slips and mistakes demonstrate the
temporary disruption of varying degrees of severity in the form of lan-
guage which is produced and ultimately point to a control failure6. Green
(2000: 376) illustrates his observation with the following example:

Transient failures include errors when we blend two or more words to-
gether. Within a language we find blends such as “strying” (blended
from “trying” and “striving”) and across languages we find ones such as
“Springling” (blended from “spring” and “Früchling”. In such cases, it
is evident that the normal speaker can recognize that an error has been
made and can also produce an appropriate utterance (Green 2000: 376).

What is more the joint activation of the two languages allows the bilin-
gual to code-switch between expressions from either of the two languages
depending which one reaches the activation threshold first. Code-
–––––––––
6
As pointed out by Walters, the construct was borrowed from cybernetics and infor-
mation processing where it was simply defined as ‘avoidance of error’ (Walters 2005: 68).
76 Chapter 2

switching and mixing languages (inserting an L2 form in an otherwise L1

utterance) is frequently done by bilinguals when in the company of other
bilinguals showing that the need for language control is determined by the
actual communicative circumstances. Usually, however, and in more for-
mal communicative interactions bilinguals tend to keep their two lan-
guages apart and to do so Green is emphatic that they have to execute
language control to inhibit undesired output and interference. In other
words Green aims to explain most things that bilinguals do: keep their
two languages apart, mix them in code-switching, translate and experi-
ence interference.
Yet, as pointed out by Green (1993: 255) most L2 acquisition theories
“fail to address the control issue explicitly” and despite the recent ad-
vances in the psycholinguistic studies of bilingualism still not enough is
known about how the use of two or more languages is controlled by the
L2 learner at various levels of the language acquisition process. If how-
ever learning a foreign or second language is considered as a case of ac-
quiring a complex cognitive skill and the use of language is perceived
within the general theory of action (Davidson 1980), the notion of control
is indeed essential to understand both the gradual mastery of the skill and
its subsequent use. To quote:

One of the fundamental features of the human bilingual brain is its ca-
pacity to control which language to use at a given moment and in a
given context, both for speech comprehension and production. This
specific cognitive mechanism, referred to as the “language control” or
“language selection” mechanism, allows bilinguals, for instance, to
communicate in one language rather than in another, and to switch
back and forth between languages during the same conversation, de-
pending on the preferred language of the interlocutors. It also allows
to implicitly identify the language of heard or written words and to
produce words in a selected target language, while minimizing the in-
terferences from the nontarget language (e.g., by preventing interfer-
ences from the native L1 during production in a weaker L2) (Abu-
talebi et al. 2007: 1).

Green (1986, 1993, 1998) in his inhibitory control model focused on the
control requirements in using language (see Walters 2005: 70) and sug-
gested that “non-pathological language use not only requires intact lan-
guage (sub) systems and intact connections between them but also the
means to activate and inhibit these systems and to inhibit inappropriate
 Bilingual foundations of translation ability 77

outputs of the systems” (quoted after De Groot and Christoiffels 2006:

190). Green suggests (1986) that speech production can be understood as
skilled action in general. “In particular, the selection of a word, like the
selection of a particular action, involves regulating a single underlying
variable of the amount of activation. Choosing an appropriate word re-
quires ensuring that its activation exceeds that of its competitors” (Green
2000: 407). Therefore a notion of activation and inhibition based on the
assumption that the two languages are constantly present in the mind of
the bilingual speaker is essential for understanding language control. Ac-
tivation and inhibition require energy and use up the available resources7.
If the resources are diminished by such factors as stress, fatigue or by ex-
cessive processing demands, performance failure may occur as a result of
insufficient control (see also Paradis 1998).
Green distinguishes three stages of activation of the two linguistic sys-
tems: selected (when in use), active (not in use but able to exert effects on
processing) and dormant (stored in the long term memory but not used and
therefore not having much effect on on-going processing). Since both lan-
guages can or have to be active at the same time in the case of translation or
interpreting the language performance is more prone to “involuntary intru-
sion” (Green 1993: 263) that is interference. However, simultaneous activa-
tion can also facilitate language processing due to spreading activation
across languages. In other communicative interactions, it can help inter-
locutors to avoid disruption in their speech production by code switching
which bilinguals do if the context allows (Grosjean 2001). In tasks which
require control of interference as is the case in translation attentional re-
sources are required and used up since the intervention of the supervisory
attentional system (see Norman and Shallice 1986) or the central executive
in the working memory model (Baddeley 1986, 1996) is needed to ensure
strict control over language production. What is stressed by Green and cur-
rently confirmed by psycholinguistic experiments (De Groot and Christof-
fels 2006) language selection is only partially a matter of increased activa-
tion of one language but mostly it operates by suppressing the activation of
the other language. Hence, Green named his model the Inhibitory Control
model (IC model in Fig. 2).
–––––––––
7
Green observes that most functional models of speech production ignore this en-
ergy requirement “and yet we would not normally consider the description of a work-
ing device as complete without some account of how it is powered” (Green 2000: 407-
408).
78 Chapter 2

Word Conceptual/intentional Word

input system input
L1 L2

specifier

Word Resource Word

output generator output
L1 L2

Phonological/orthographic assembly

Speech/writing output

Fig. 2. The functional control circuit in bilingual production (Green’s In-

hibitory Control Model (1986) adapted from Walters 2005: 70).

In his model described by Walters (2005: 67) as a ‘general purpose process-

ing model’ Green takes on the distinction of three independent functional
processing components: conceptualization, formulation, articulation which
operate during language production (see Levelt 2001). The IC model fo-
cuses on lexical processing and is prepared to encompass “a range of phe-
nomena in both bilingual and language impaired performance, in particular
speech disfluencies, code-switching, and translation” (Walters 2005: 67).
There, are however two conditions that have to be met to bring about lan-
guage control: explicit intention and language tags on both word meanings
(lemmas8) and word forms (lexemes). Green similar to La Hej (2005 quoted
–––––––––
8
in psycholinguistics the lemma is in fact an intermediary level between word mean-
ing and the word’s form (lexeme) and apart from the meaning component it includes in-
formation about the syntactic behaviour of the word before the phonological form is se-
lected (i.e. the meaning is lexicalized). Many scholars however do not distinguish between
the semantic level (the level of conceptual meaning and lemmas and treat them as one level
in contrast with the lexical level (see Caramazza 1997 and Paradis 2009).
 Bilingual foundations of translation ability 79

after De Groot and Christoffels 2006: 195) draws a parallel between lan-
guage tags in bilinguals and tags in monolinguals for style and register9. Ini-
tially (Green 1986, 1983) the model was based on a fairly simple operating
principle. In L2 production there is increased activation of both L2 mean-
ings (lemmas) and forms (lexemes) while L1 lexemes are suppressed at the
level of articulation. In interference or in code switching, as mentioned ear-
lier, there is a competition at the level of phonological assembly and the
word which reaches the activation level first is the one which is pro-
duced/articulated. In his later elaboration of the IC model Green (in Price et
al. 1999) introduces the condition of multi-tasking to his notion of control in
bilingual processing (see Walters 2005). Drawing an analogy between bilin-
gual production and the Stroop effect10 Green explains that when asked to
translate a printed word, “bilinguals have to avoid naming the printed word
and, instead, produce a translation equivalent as a response” indicating in
this way that the concept of selection, competition and suppression/ inhibi-
tion are the major processing notions in bilingual production. To account for
the versatile multi-level nature of control Green uses the procedural notions
of schemas and goals. Schemas, following Norman and Shallice (1986), are
best perceived as “networks detailing action sequences… that individuals
may construct or adapt on the spot in order to achieve a specific task…”
whereby language is perceived as “a form of communicative action” (after
Walters 2005: 72). Green (in Price et al. 1999) explains that schemas are
procedural in nature in the sense that they use the declarative representation
of the instructions and become ‘methods’ to achieve the intended goals by
gaining the control of action. Green claims that the control is multi-level
(global or local) and the schemas are task specific: low level (articulation),
intermediate (lexical decision tasks and translation) and high level (business
–––––––––
9
La Heij (2005) suggested that a language cue is present at the preverbal stage
which specifies which language is going to be used. Lemmas are also tagged for lan-
guage so the outcome of the “complex access, simple selection process” depends on the
overlap between the cue and the information in lemmas.
10
The Stroop effect in psychology named after John Ridley Stroop (1935) demon-
strates the reaction time taken by a task in which a person is asked to name the colour
printed in a name denoting a different colour, i.e. the yellow colour ink is used in the
word red. The person asked to name the colour has to suppress the meaning of the word
‘red’ and say ‘yellow’ naming the actual colour. The Stroop test has consistently shown
that naming of a colour takes longer and is prone to more errors when there is a concep-
tual mismatch between the printed word and the colour than when the colour and the
word refer to the same colour (i.e. the word “blue” is printed in blue ink).
80 Chapter 2

meetings, written production and conversational interaction). Although the

reason why translation requires an intermediate level of control is not ex-
plained by Green, it is possible to expect that he refers in his example only
to single word translation and not to translation as an act of communication
which as such would have to require a high level of control.
The goal to carry out a “linguistic act” is controlled by a coordinated
effort involving three components: the executive or Supervisory Atten-
tional System (SAS see Norman and Shallice 1986) which is responsible
for non-automatic behaviour and specifies which language is to be used.
This central specification is then passed to the task schemas (as instruc-
tion ‘to translate’, ‘to name pictures’, etc.) which “compete to control out-
put from the lexico-semantic system” by adjusting the activation levels of
representations and by inhibiting outputs” (Walters 2005: 72). At this
point, the primary question is what drives the intentional system. Walters
(2005: 73) claims that it is the social and pragmatic motivations (cf. Para-
dis 2004, 2009), the situational context and the participants (Grosjean
1997) that drive the control model11.
To sum up, Green’s notion of language control can “provide an alter-
native means for understanding both the development and the breakdown
of performance in L2” (Gollan and Kroll 2001: 331). The notion of con-
trol allows to explain the competence – performance divide in L2 learners
who make mistakes. Although as observed by Gollan and Kroll (2001:
332) a reservation has to be made that performance errors can be a mix-
ture of insufficient control and poor representation of the weaker lan-
guage. Yet, the same notion of language control is valid for native lan-
guage users who make slips although they “have intact cognitive and lin-
guistic representations” with unimpaired links between them. In language
use as in any other skilled action slips occur because of temporary prob-
lems with control which cannot be executed appropriately if the resources

–––––––––
11
Walters (2005), SPPL (Socio-Pragmatic Psycho-Linguistic) model attempts to in-
tegrate the sociopragmatic and psycholinguistic aspects of bilingualism which have a po-
tential to point to some areas in SL learning which overlook the close connection be-
tween the social identity and cognitive perspective involved in the acquisition and use of
one’s bilingual knowledge. The model also favours the holistic view of language as-
sumed by Hymes (1971) in his notion of communicative competence. The SPPL model
of bilingualism is grounded on the assumption that the L1 and L2 information is rooted
in the social world, and language choices are possible at every stage of production from
intention (conceptualization) to articulation.
 Bilingual foundations of translation ability 81

are insufficient (Green 1993). Séguinot (1989) studying translation per-

formance in professionals noted that mistakes are made when the infor-
mation processing demands exceed the capacity for the control of actions.
The main asset of Green’s IC model is the focus on cognitive re-
sources and energy requirements. In this respect the notion of control fits
in with the cognitive view of the mind as a limited capacity processor
(McLaughlin and Heredia 1996: 213). According to the cognitive princi-
ple learning is perceived as a movement from controlled to automatic
processing via practice that is the repeated activation of the newly ac-
quired knowledge (see Anderson and his Adaptive Control of Thought
model, 1983, 1985). Learning a new language is considered as learning
any other complex skill where practice propels the continuous movement
from controlled processing which requires a lot of attentional expenditure
to automatic processing which provides more mental resources for other
aspects of language use, such as for example concentrating more on the
content, self monitoring, etc.
The notion of language control in bilingual processing opens other is-
sues inherent in different uses of language. Translation, which is the focus
of the discussion in this work, will obviously place different requirements
within Green’s model than, for example writing an essay in English or par-
ticipating in social interaction. The metacognitive control over the transla-
tion process requires efficient mental effort management in which language
control may consume a considerable amount of cognitive resources espe-
cially in novice translators. The importance of the bilingual’s ability to con-
trol the two languages at the lexical level can be expanded to the need to
control all other linguistic levels from phonology to discourse. Avoiding in-
terference is what translators learn to do throughout the course of develop-
ing their translation proficiency. Green’s model of inhibitory control has
several implications for the study of translation as a human skill. First of
all, it accounts for the limited cognitive resources which have to be well
managed to perform the task (Whyatt 2010). Gile (1995/2009) used the
term coordination effort to refer to the need to control the translation proc-
ess, which as suggested by De Groot and Christoffels (2006) performs the
function similar to Green’s concept of language control. Secondly, the need
to control languages at the same time implies that language switching
which occurs in the translation process all the time will have its cost in
terms of mental energy and attentional demands of the translation task. As
pointed out by Costa and Santesteban (2004) the language switch cost will
82 Chapter 2

differ depending on the level of L2 proficiency and language dominance.

Paradoxically, however, behavioural research shows that the cost of switch-
ing back to one’s stronger language (usually L1) is higher, possibly show-
ing that undoing the stronger inhibition takes a relatively longer time (see
Pavlenko 2009: 13). Thirdly, it assumes constant cooperation between the
working memory of the translator and his/her long term memory where all
the bilingual knowledge is deposited and might be at any time of the trans-
lation process urgently needed. When code-switching in translation, the
translator voluntarily and consciously aims to keep the two languages apart
during the formulation and articulation (or written text production). Yet, if
the language control mechanism is not efficient enough the TL text might
still show interference which contrary to code-switching is involuntary (see
Presas 2000: 26 quoting Beatens Beardsmore 1982: 110) and most proba-
bly on a subconscious level (I will return to the notion of control and con-
sciousness in chapter five). Finally, as pointed out by Bialystok et al. (2004)
language control exercised in bilingual language use may have a positive
effect on cognitive control in general (see Prior and MacWhinney 2010 on
language switching and task switching) which is vital for mental effort
management in the translation process in which language switching and
task switching occur in proportions unmatched by other uses of a bilin-
gual’s two languages (Gile 1995, Whyatt 2010). It is also possible that be-
cause of the strict requirements for efficient language control in the transla-
tion process, translation tasks can be used in language teaching to enhance
language control in bilinguals (Whyatt 2009b), an issue to be discussed in
detail in chapter 3. Having introduced the notion of language control in bi-
lingual language use (also in translation) let us provide some insight into
the structure and organization of the bilingual’s lexico-semantic system to
fully understand the control requirements in a bilingual language user and a
potential translator over his/her bilingual memory.

2.6. Bilingual memory

Linguistic competence as a component of communicative competence in-

volves the knowledge of grammar and vocabulary. Grammatical rules are a
finite set of patterns similar to mathematical algorithms which specify all the
possible combinations of words which are considered acceptable in a given
language. Knowing two languages presupposes the knowledge of two
grammars and two lexicons. As observed by Vermeer (1992: 147), “Knowing
 Bilingual foundations of translation ability 83

words is the key to understanding and being understood. Children acquire

words first, and next the grammar of a language. The bulk of learning a new
language consists of learning new words: grammatical knowledge does not
make for great proficiency in a language”. Since any linguistic action in-
volves the use of vocabulary and translation tasks place some special de-
mands on the selection of lexical items it is essential to devote some attention
to the presently available understanding of the bilingual mental lexicon. As
observed by De Groot (2002) psycholinguistic research into bilingualism has
been dominated by two related questions referring to the representation and
processing of vocabulary by bilingual speakers. The question of representa-
tion, “tries to decide whether bilingual memory contains two separate lan-
guage specific stocks of knowledge, one for each of the bilingual’s two lan-
guages, or, instead, a single stock that is shared between the two languages”
(De Groot 2002: 32). Research into lexical processing tries to decide whether
lexical processing is language selective and thus handled only by the ‘con-
textually-appropriate language subsystem’ or whether lexical processing is
essentially language non-selective and thus handled by both of the bilingual’s
language subsystems which are assumed to form one unified system.

2.6.1. The hierarchical structure of the bilingual lexicon

Research into the storage and access to the bilingual lexicon has focused
on two aspects of lexical knowledge: the word’s form (surface features,
sound, spelling) and meaning (semantic content), and other aspects like
for example, syntactic information about the possible combinations it can
form with other words or pragmatic information about its contextual con-
straints in terms of usage have not been taken into account. In a way then,
research into the structure and processing has chosen to treat the bilin-
gual’s mental lexicon as similar to a bilingual dictionary. Nevertheless,
experimenting with this limited lexical information has allowed research-
ers to arrive at a basic conclusion that bilingual lexical memory is hierar-
chical in nature and consists of at least two layers of memory representa-
tions (or nodes), one layer stores meanings of words (lemmas) and the
other layer stores their forms12 (lexemes) (De Groot 2002).
–––––––––
12
Evidence comes from monolingual studies (Smith 1997) which showed dissocia-
tion effects between forms (lexical factors) and meanings (semantic factors). Reaction
time experiments, for example showed that response time was shorter for strings of let-
ters which were words than for non-words (Besner et al. 1990).
84 Chapter 2

In the case of bilinguals the hierarchical structure of the mental lexicon

has to account for the question whether the two languages are segregated or
integrated at both levels, or at one of the levels. In other words, does the bi-
lingual mind associate the English word girl with the same semantic repre-
sentation (meaning, concept, mental image) as the meaning evoked by the
Polish word dziewczyna or the French word fille? If this is the case then the
bilingual lexicon is integrated at the layer of meanings (shared conceptual
representation) but separated at the layer of forms (separate lexical repre-
sentations). As pointed out by Paradis (2007) the bilingual memory in ef-
fect contains three stores (a three-store hypothesis): the common underly-
ing conceptual base (CUCB) which is shared between the two languages
(see Kesckes and Papp 2000) and two lexical stores, for the bilingual’s L1
and L2. Evidence for the CUCB and separate lexical stores comes from
translation-priming13 or cross-language priming experiments which showed
no effects of translation priming in data-driven tasks (focused on process-
ing word forms, e.g. in fragment completion when the subjects are asked to
fill in the missing letters in a word (el- - ph - nt = elephant, example from
De Groot 2002: 35, see also Cieślicka 2005) or in lexical decision tasks
which measure response times to strings of letters which form words or
non-words. Gerard and Scarborough (1989) and Kirsner et al. (1984) re-
ported the lack of differences in the response time (null-effects of transla-
tion priming) when a response to the word apple was preceded by its
French translation equivalent pomme or a semantically unrelated femme (at
least if the translation pairs consist of non-cognates that is completely dif-
ferent forms in the two languages). These findings allowed to conclude that
the word forms are stored in two separate memory representations. In con-
trast a translation priming effect was obtained in conceptually-driven tasks
(focusing on meaning such as free recall or a semantic decision task where
participants are asked to categorize the words as e.g.,, concrete or abstract)
for both cognate and non-cognate words. Semantic priming studies showed
that a word (target) is processed (recognized) faster if it is preceded by a
semantically related word (prime) than when it follows a semantically unre-
lated word. The semantic-priming effect was obtained when both prime and
target were from the same language (Neely 1991) and when they were from

–––––––––
13
the term ‘priming’ refers to the finding that the prior presentation of a word (the
‘prime’) speeds up the processing of the same word (repetition priming) or another word
(called the target) (e.g. Forbach et al. 1974).
 Bilingual foundations of translation ability 85

the bilingual’s two languages (Chen and Ng 1989; De Groot and Nas
1991). These findings of spreading activation in the conceptual/meaning
store suggested that meaning representation of translation equivalents are
shared between the two languages and form a single conceptual store, a
common underlying conceptual base (CUCB). The question which imme-
diately arises is: What are the connections between the three stores if the
priming effect was obtained in conceptually driven tasks but not in data-
driven tasks?
There has been a considerable amount of research examining the con-
nections between lexical and conceptual representations (Kroll and Stew-
art 1994). The starting point was taken from Weinreich’s (1953) classifi-
cation of bilinguals on the grounds of their different language acquisition
histories (see also Ervin and Osgood 1954) into coordinate, subordinate
and compound bilinguals. Potter et al. (1984) assumed that the different
sociolinguistic experiences of different bilinguals will lead to different or-
ganization of the three components and presented two models: the word-
association model (Wienreich’ subordinative model) and the concept me-
diation model. As shown in Fig. 3 the word association model predicts the
route of L2 acquisition via L1 and demonstrates that the underlying con-
ceptual representations can be accessed only via the bilingual’s first lan-
guage. On hearing an L2 word, for example ‘door’ one does not see the ob-
ject (access its mental representation/concept) but first has to retrieve the
Polish translation equivalent ‘drzwi’ and only then can access the meaning.
As pointed out by De Groot (2002: 37), “the L2 word is in fact assigned the
L1 word’s meaning”. Consequently, the conceptual capacity of the bilin-
gual within the word association model equals the conceptual capacity of a
monolingual L1 speaker. This kind of organization is characteristic of the
initial stages of L2 learning although the direct communicative methods of
teaching are aimed at minimizing the role of L1 in the process of L2 acqui-
sition. By providing direct exposure to L2 efforts are made to make the de-
veloping bilingual lexicon fit the concept mediation model characteristic of
natural bilinguals.
86 Chapter 2

CONCEPTS

L1 L2

Fig. 3. A word association model (subordinative bilingualism) (adapted

from De Groot 2002: 37).

In the concept-mediation model (Weinreich’s compound model) both L1

and L2 lexical items have direct access to the conceptual layer (Fig.4).

CONCEPTS

L1 L2

Fig. 4. The concept-mediation model (compound bilingualism) (adapted

from De Groot 2002: 37).

It was suggested that with time and practice the word-association links
can be gradually replaced by the direct concept-mediation links. This so
called intermediate (Potter et a. 1984) or developmental model (Kroll and
De Groot 1997) linked the nature of lexical and conceptual connections to
the bilingual’s level of proficiency (Gollan and Kroll 2001: 325). It was
assumed that a word-association memory structure predicts longer re-
sponse times (RTs) than the concept mediation model14. Yet, Potter et al.
(1984) failed to obtain differences in reaction times between the non-

–––––––––
14
Note that in Potter’s et al (1984) model the labels L1 and L2 indicate the order in
which the two languages were acquired as well as the dominance (relative strength) of
the two languages. However Heredia (1997) noted that the dominance issue is always
relative to the way both languages are used. If the initially weaker L2 becomes the lan-
guage of everyday communication it can become more dominant (stronger) than L1.
 Bilingual foundations of translation ability 87

fluent and fluent bilinguals on picture naming that is a task known to acti-
vate access to the conceptual memory, and word translation (L1 into L2)
tasks. Chen and Leung (1989) also compared RTs on picture naming and
translation tasks between non-fluent and fluent bilinguals and found that
the beginners responded consistently with the predictions of the word-
association model that is the response time for L2 to L1 translation was
faster than picture naming in L2. The proficient bilinguals’ responses on
the other hand, showed the concept-mediation retrieval patterns. Further
research led to a proposal of the revised hierarchical model also referred
to as the asymmetry model.

2.6.2. The revised hierarchical model (RHM)

Kroll and colleagues (Kroll 1993, Kroll and Sholl 1992, Kroll and Stewart
1994) concentrated on the strength of the connections between the three
components of the bilingual lexicon and revised the hierarchical model to
account for the asymmetry of lexical processing (see Fig. 5).

CONCEPTS

L1 L2

Fig. 5. The revised hierarchical model (RHM) (adapted from De Groot

2002: 39).

The model reflects the different level of proficiency (command of the two
languages) or the ‘language imbalance’ due to “differential experience
with the two languages, with L1 having been used more than L2” (De
Groot 2002: 40). In a way the RHM combines the word-association and
the concept-mediation model adding connections between the three com-
ponents which differ in strength. A strong unidirectional connection holds
between the conceptual layer and the L1 lexical layer whereas the link be-
tween the conceptual layer and the L2 forms is weak. The strong link be-
tween L2 to L1 remains and reflects the frequent practice of learning L2
88 Chapter 2

words through associating them with their L1 translations (Kroll and

Stewart 1994: 158). The connection from L1 forms to L2 forms is weak in
comparison with the strong link between L1 forms and concepts.
De Groot (2002: 40) points to the developmental nature of the RHM
which predicts that, “the processing of L2 by non-fluent bilinguals will
parasitize L1 word forms more often than will L2 processing by proficient
bilinguals” (note that a clear reference to the developmental aspects of
cross-linguistic influence is made in this quote). What is interesting, the
model predicts direction effects on word translation speed and it is as-
sumed that L2 to L1 translation will be faster (the strong connection be-
tween L2 and L1 forms) than L1 to L2 which is always conceptually me-
diated and takes longer. Kroll and Sholl (1992) and Sholl et al. (1995)
demonstrated the direction effect on RTs in translation which involved
manipulating a meaning-related variable (clustering into semantic catego-
ries vs randomized items) and obtained a facilitating effect of the seman-
tic priming of an earlier picture naming task on subsequent L1 to L2
translation and the lack of the priming effect on L2 to L1 translation. Con-
firming again that L2 to L1 translation of isolated words utilizes lexical
links. However, a significant number of studies presented results which
are problematic for the asymmetry model pointing in this way to some de-
ficiencies and limitations of the model.

2.6.3. Deficiencies and limitations of the RHM

De Groot et al. (1994) reported null-effects of translation direction on RTs

or even reported the opposite effect of faster translation from L1 to L2. It is
therefore possible that the conceptual memory store is activated when
translation takes place irrespective of the direction (L2 to L1 or L1 to L2).
Further challenges for the RHM were reported from the studies of the word
type effect (Heredia 1997, Cieślicka and Ekert 2009) on the speed of lexical
processing. The processing of concrete words was consistent with the RHM
whereas the processing of abstract words contradicted the model (Heredia
1997, De Groot and Keijzer 2000). Heredia (1997) responding to the RHM
of the bilingual lexicon suggested that, “regardless of which language is
learnt first, the more active (dominant) language determines which lexicon
is accessed faster”. The asymmetrical model cannot account for the effect
of word type (concrete/abstract) and word frequency on processing because
it does not focus on how the processing precedes but on processing in gen-
 Bilingual foundations of translation ability 89

eral (Cieślicka and Ekert 2009). Words are processed differently depending
on such features as concreteness, frequency or familiarity and recency of
use. Further challenges came from studies which showed lexical processing
inconsistent with the RHM involving cognates (words that are similar in
form across the two languages) as well as low-frequency and high-
frequency words suggesting that there are other determinants which influ-
ence access to the bilingual lexicon (for a detailed review of determinants
of bilingual memory see De Groot 2002).
A further challenge to the three-component model comes from the fact
that the two words in translation pairs, more often than not, do not share
meaning completely15 (a frequent problem in translation practice see Per-
dek 2011) whereas both models the word-association model and the con-
cept mediation model seem to imply that the meaning assigned to an L2
word is the meaning of the corresponding translation equivalent in L1. This
observation led to focusing more on the conceptual layer of the bilingual
lexicon. De Groot (1992, 2002) suggested that accounting for the cross-
language non-equivalence of meaning requires to assume that the mental
representations have to be perceived as being composed of “sets of more
primitive meaning elements” (De Groot 2002: 48). Then the RHM can ac-
count for the non-equivalence by assuming that “the two words in a transla-
tion pair do not have to share exactly the same set of such elementary
meaning elements” (conceptual feature model see Kroll and De Groot
1997: 187 or distributed feature model in De Groot 2002: 49).
CONCEPTS

L1 L2

Fig. 6. The distributed feature model (adapted from De Groot 2002: 49).

–––––––––
15
This non-equivalence of meaning across languages was interpreted in early bilin-
gual memory research as indicative of separate (coordinate) conceptual stores for a bi-
lingual’s two languages (Weinreich 1953, see also Kolers 1963 for indications of inter-
lingual associations).
90 Chapter 2

The figure above shows the lack of overlap between all the meaning com-
ponents16 although the exact definition of the componential elements is not
discussed by De Groot it is probably best referred to as the semantic fea-
tures model (Aitchison 1996) or the perception of words as ‘bundles of
meaning’ (Kussmaul 1995) in Translation Studies literature. The major as-
set of the conceptual feature model is that it further complicates the RHM
and it grants due attention to the conceptual layer of the bilingual mental
lexicon. The common underlying conceptual base (as observed by Paradis
(2007) the place where language takes meaning from) is not static and its
transformation goes together with developing language proficiency. As
stressed by Pavlenko (1999) it changes all the time throughout life’s ex-
periences and differs between individuals. Exploring the area of bilinguals’
conceptual competence and adding data on bilinguals’ conceptual profi-
ciency is therefore a needed complement to the rich data on lexical process-
ing and representation. It is this layer in the bilingual lexicon that is of ut-
most importance in translation and the knowledge of how it is organized
and linked to the lexical stores of L1 and L2 is essential to understand lan-
guage processing in translation carried out by people at different stages of
the developmental continuum of translation as a human skill.
Paradis (1997, 2009) and Pavlenko (1999), among others call for the
need to distinguish between semantic and conceptual representations in
the bilingual lexicon whereas the three-component model considers these
two types of knowledge as belonging to the store of concepts. To quote
De Groot (2002: 48), “This may turn out to be a crucial flaw of the model,
although conflating them [the semantic and conceptual meaning, BW]
may be the logical consequence of the facts that pinpointing the differ-
ence between semantic and conceptual knowledge is a tedious task, and
that both types of knowledge that are plausible originate from one and the
same source”. Altough, De Groot does not specify this source. Green
pointed out that “there is a need to distinguish between the semantic and
the conceptual” saying that “What is envisaged is not always expressed
and what is uttered does not fully determine sense” (Green 1993: 253).
Pavlenko (1999, 2009) draws a clear distinction between semantic knowl-
edge and conceptual representation. While a semantic component is char-
–––––––––
16
An example is the Polish word pietruszka (which can refer to the green leaves and
the root of the vegetable as well) and English ‘parsley’ (which refers only to the green
leaves), or Polish drugie śniadanie (literally ‘second breakfast’) and its potential out of
context ‘partial’ equivalent in English: ‘brunch’.
 Bilingual foundations of translation ability 91

acterized by polisemy and provides “explicitely available information

which relates the word to other words [its position in the word-web as de-
fined by Aitchison (1996), BW], idioms and conventionalized expressions
in the language”, the conceptual component contains “non-linguistic
multi-codal information, which includes imagery, schemas, motor pro-
grammes, and auditory, tactile and somatosensory representations, based
on experiential world knowledge” (Pavlenko 1999: 212, quoted after De
Groot 2002: 47-48). Paradis (2009) goes even further and distinguishes
between vocabulary and lexicon. The vocabulary is a list of words and
their meanings and people come to learn the form-meaning associations
by seeing and hearing the words in their both languages. The word-
meaning relationships are not stored in the procedural memory like the
rest of one’s implicit linguistic knowledge but in the declarative memory
as explicit knowledge. “The lexicon is a network of interrelated elements
with their inherent morphosyntactic properties. In addition to their spe-
cific lexical semantic constraints, both the interrelationships between
lexical items and their syntactic properties differ from one language to the
next (and are implicit)” (Paradis 2009: 16). The lexicon is what people
“acquire through use, from encountering words in different sentential con-
texts” (Paradis 2009: 17).
The distinction is crucial and rests on the learning versus acquisition
process (discussed in section 2.2) with the consequences anchored on the
fact that vocabulary being a part of explicit knowledge is subserved by
the declarative memory whereas the lexicon being a part of implicit
knowledge is subserved by the procedural memory. In effect, “[n]ative
speakers acquire a lexicon, and with time, in school, or by sheer observa-
tion, they learn new words with their associated meanings. Second lan-
guage learners usually gain knowledge of a vocabulary before they ac-
quire a lexicon, and often explicitly learn, at least partially, the syntactic
properties of (some) words. As a result, most L2 learners have at their
disposal a number of sound meaning (and written word-meaning) associa-
tions but lack the competence related to (often also the conscious knowl-
edge of) their morphosyntactic properties, which may vary from those of
their native language” (Paradis 2009: 17). Consequently, the psycholin-
guistic experiments which provided evidence for the RHM “can only
speak to the representation of individuals’ vocabulary, not their lexicon”
(Paradis 2009: 18). The above reservations suggest that the structure of a
monolingual and bilingual lexical representation contains more layers
92 Chapter 2

(e.g., the layer of non-linguistic concepts, lexemes (forms) and lemmas

which contain information about the words’ grammatical behaviour). It is
also quite likely that the content of separate layers may be much richer
than suggested in the RHM. What is more, recent evidence from research
focusing on concepts in the bilingual lexicon points to much more com-
plex conceptual organization in the bilingual memory (Pavlenko 2009:
126) which will affect language processing in bilinguals’ in spontaneous
speech, creative writing and translation.
To sum up the hierarchical structure of the bilingual lexicon is undis-
puted as such, although as pointed out by Pavlenko (2009) the common
conceptual store remains to be explored to reveal cross-linguistic differ-
ences. Among the most insightful conclusions is the fact that the bilingual
lexicon is not a static formation but is developmental in nature. Depending
on such factors as the language acquisition history, level of proficiency in
L2 and everyday use of the L2 the strength of the connections between the
three components (or stores) may vary amongst bilinguals. In other words,
to quote Kroll and De Groot (1997: 170), “the architecture of the bilingual’s
mind may be a reflection of the level of expertise in the second language,
and the context in which the second language was acquired”. To account
for the challenges to the RHM reporting word type effect, frequency and
saliency of words in both languages De Groot (2002) suggested that the
state of ‘mixed representations’ that is the coexistence of word-association
representations as well as concept-mediation representations within one and
the same multilingual mind might be possible. Numerous studies (see Kroll
and De Groot 1997 for an extensive overview) including translation tasks
support the developmental shift which takes place “in representation and
processing as a function of increasing L2 proficiency” (Kroll and De Groot
1997: 176) and reliance on lexical processing is gradually replaced by con-
ceptually mediated processing. This however does not mean that lexical
links die out or disappear (Whyatt 2006b). What is more, as stressed by
Gollan and Kroll (2001), a specific language experience and intensive prac-
tice, for example in translation, may lead to an increased awareness of the
degree of overlap (equivalence, partial equivalence, or conceptual gaps) at
the conceptual level shared by the two languages in the bilingual mind.
Consequently, intensive practice in language use might lead to a re-
organization of the strength of connections between the three components
of the bilingual lexicon (Presas 2000, Whyatt 2006a) and will improve the
process of accessing word forms and meanings in language production or
 Bilingual foundations of translation ability 93

translation due to increased language control (Green 1993). To encompass

all these variables De Groot (2002) admits that the bilingual lexicon can in
fact have mixed connections between the three stores with some being
compound and others coordinate in nature. What is more with developing
L2 proficiency and growing language experience coordinate connections
might become compound and compound connections might become coor-
dinate. However, as pointed out by Kroll and Linck (2007: 239) psycholin-
guistic research on lexical access in the L2 studied the performance of
‘relatively skilled adult bilinguals’, “with fewer studies that specifically ad-
dress the changes that occur in language processing as L2 skill develops”.
What remains fairly undisputed is that although the activiation of L1 de-
creases with growing proficiency in the L2, growing skill in the L2 “does
not correspond to a switching off of the L1” (Kroll and Linck 2007: 246-
247), hence the need for an efficient language control mechanism (see sec-
tion 2.5) in lexical access both in L2 or L1 production and in translation.
Ivanova (1998: 96), for example observed that, “translation requires
deeper-level processing which is beneficial to increasing the fluency of
lexical retrieval” (see also Kroll 1993).

2.6.4. Bilingual memory and translation

Psycholinguists are not unanimous on whether lexical access is language

selective (the language is specified prior to retrieving a word from mem-
ory), for example the RHM (Kroll and Stewart 1994, Gollan and Kroll
2001) or nonselective (bilingual lexicon does not have two separate func-
tionally independent lexical representations but one integrated lexical
store) like in the Bilingual Interactive Activation model (Dijkstra and Van
Heuven 2002). Undoubtedly, bilingual lexical access (see Levelt 2001 for
a detailed analysis of the lexical access process, and Costa and Sant-
esteban 2004 on lexical access in bilingual speech production) is of pri-
mary importance in translation tasks in which one constantly reaches into
language specific lexical stores of the bilingual lexicon to find a needed
lexical item in the target language. Accessing a desired word in translation
is dictated not only by the meaning relationships at the conceptual level
but also by the language specific contextual requirements and formal (i.e.,
grammatical) constraints. Interpreting meaning incorporates not only ac-
cess to concepts but also the frequently subtle knowledge of word combi-
nations and syntactic patterns (Lewandowska-Tomaszczyk 1990, 2004).
94 Chapter 2

In this respect the research on bilingual memory in the form of the RHM
does not provide all the subtle data included in lexical access in transla-
tion practice which is always contextualized and positioned in a specific
socio-pragmatic situation dictating specific requirements. As observed by
De Groot (1997: 42) the availability of context is likely to alter the predic-
tion of the RHM on the concreteness effect in lexical access with no dif-
ference between accessing concrete and abstract words. On the other
hand, being able to pick up contextual clues especially at the macro level
and use them in lexical access during the translation process is a cognitive
skill which develops with translation experience (to be discussed in chap-
ter three). Initially, it seems that untrained natural translators’ performance
exhibits the tendency for word-to-word (horizontal) translating along
lexical links (Krings 1986a, Lörscher 1991). Vertical translating (includ-
ing conceptual meaning analysis) has been ascribed to more experienced
translators (De Groot 1997, Seguinot 1997).
The dynamic interactive nature of bilingual memory and bilingual
knowledge points to the constant need to inhibit undesired competing
lexical items both crosslinguistically and within one language. What is
more, recent psycholinguistic studies (e.g., Thierry and Wu 2007) using
neurolinguistic methods (brain potentials) reveal ‘unconscious translation’
during foreign language text comprehension which confirms the simulta-
neous activation of both languages not only during language production
(Marian and Spivey 2003). This crosslinguistic simultaneous activation is
useful in translation17. In TS Think Aloud Protocol studies provided a lot
of evidence for the competition among TL potential candidates that is
lexical items which are made available to the translator (e.g., activated in
his/her bilingual memory) who then has to select (decide on) the best
match for the target language text. This tedious process of assessing po-
tential equivalents, rejecting the misfits and selecting the most appropriate
one in terms of not only meaning but contextual appropriateness may sig-
–––––––––
17
Paradis et al. (1982) reported on an intriguing case of crosslinguistic activation in
bilingual aphasics who when recovering were able to translate from one of their lan-
guages into the other but they were at the same time unable to use the other language for
spontaneous speech production. The authors termed this phenomenon paradoxical trans-
lation and concluded that translation is a cognitive task dissociable from understanding
and speaking two languages. Another viable explanation is that when translating the pa-
tients used their metalinguistic knowledge and translated words using lexical links (see
De Groot and Christoffels 2006: 190).
 Bilingual foundations of translation ability 95

nificantly contribute to high energy expenditure in translation tasks. As

reminded by Green (1986, 1993) the cost of language control and con-
stant switching between languages in terms of cognitive resources, which
are in limited supply, is not to be underestimated and has to have a share
reserved in mental effort management during the translation process. On
the other hand, the close interaction of the two languages in the bilingual
memory is likely to facilitate the translation process by spreading activa-
tion, providing that both lexical stores and the common underlying con-
ceptual base have rich inventories of lexical items and mental representa-
tions. It is to be expected that through translation practice the bilingual
mental lexicon will grow not only in terms of items which are available
but also in terms of the changing strength of the connections between
lexical forms and conceptual representations. Paradis (2009) even sug-
gested that experienced conference interpreters throughout their working
life collect many crosslinguistic equivalents which can be quickly recalled
from memory and facilitate the translation process. Yet, his suggestion
that experienced interpreters use only lexical links without conceptual
processing seems questionable as it implies that it is possible to translate
without interpreting the meaning (conceptual content of the SL utterance)
relying only on the metalinguistic lexical associations. It seems more
plausible to expect that both lexical and conceptual processing comple-
ment each other especially in familiar situations and contexts.
This brings the issue of the place of bilingual knowledge seen as a part
of general knowledge structures. The relationship between language and
knowledge in the bilingual mind has already been signaled by the three
store hypothesis and the postulate of the common underlying conceptual
base which both L1 and L2 share. Yet, many questions need to be asked
about the impact of translation experience on both lexical availability and
conceptual proficiency of the translator at various stages of the develop-
mental continuum discussed in chapter one. Fast and fluent lexical access
is very much desired in the translation process (see Bell 1991) and the
need is even more urging in interpreting practice. Gile (2009: 227) put
forward his Gravitational Model of lexical availability in interpreting
which draws on the frequency of use and recency effects reported in lexi-
cal access research (Prat et al 2007). Bell (1991) described his concept of
the frequent lexis store and the lexical search mechanism relying also on
the frequency and recency effects for lexical access. Although these pro-
posals are derived from the authors’ extensive experience as scholars and
96 Chapter 2

practitioners they nevertheless await empirical validation. Finally there is

a question of the effect of translation experience on the structure and or-
ganization of the bilingual memory. In this scenario a translator at any
stage of translation skill development is a constant language learner and
in his or her work new vocabulary items are encountered on an everyday
basis. Translating them requires careful measurement of the degree of
conceptual overlap between an SL lexical item and its potential semantic
equivalent chosen for the TL text. Perfect matches occur especially in
concrete words (e.g., English ‘apple’ for Polish jabłko) but are in the mi-
nority as compared with those which are considered approximate equiva-
lents (e.g., ‘glass’ for szklanka). Abstract words are definitely more prob-
lematic as their meaning is largely influenced by the culture of the respec-
tive languages (see Wierzbicka 1991 on her account of the Polish word
tęsknota and the English partial equivalent ‘longing’). Needless to say the
problems with lexical access will increase with non-literal figurative ex-
pressions as for example in ‘You are the apple of my eye’, which when
translated into Polish will most likely mean, Jesteś oczkiem w mojej
głowie. In this example as well as in metaphors, and especially conceptual
metaphors, the need for conceptual proficiency becomes of the utmost
importance.
Most practitioners would agree that translation practice is a good
source of language experience when it comes to subtle data on words and
their meanings. It is not surprising that translation tasks may raise an L2
learner’s awareness of the complex nature of meaning. What on the sur-
face might seem to be a pair of translation equivalents turns out to share
only some aspects of meaning (Whyatt 2006b). For example, for a begin-
ning Polish learner of English the Polish word seler and English word
‘celery’ are a perfect pair of translation equivalents which share the same
concept (the vegetable called ‘celery’). Yet, using both terms as transla-
tion equivalents would lead to a conceptual misunderstanding as the Pol-
ish word seler refers to the root of the vegetable which is the most fre-
quently used part in Polish cuisine, whereas the English word ‘celery’ re-
fers to the stems (or sticks) which in Polish are referred to as seler na-
ciowy and still remain a novelty in Polish kitchens. The English equiva-
lent of the Polish ‘seler’ is the word ‘celeriac’ which is rarely used as the
root part of the vegetable is not a part of the standard British diet (Whyatt
2007b). Experiencing these intricate asymmetries in which meaning is
mapped on lexical items in both languages of a bilingual person is fre-
 Bilingual foundations of translation ability 97

quently a by-product of the translation experience. Yet, it is frequently ob-

served that in L2 teaching verbal fluency is valued higher that conceptual
proficiency in L2 use. It is probably not an overstatement to say that one
can be fluent and come across as a proficient L2 learner and user but at
the same time be very much unaware of confusing concepts which are be-
hind the L2 words. Danesi (1992) pointed to the fact that some L2 learn-
ers may adhere to the rules of L2 grammar, show high verbal fluency and
come across as well organized communicators, but their discourse will
lack the conceptual appropriateness exhibited by native speakers of L2.
Therefore, verbal fluency does not have to reflect the learner’s conceptual
proficiency. This can be possibly reflected in L2 learners’ weak ability to
understand and interpret implicit meanings in L2 as well as figurative ex-
pressions (Cieślicka 2004 demonstrated it in advanced Polish learners of
English) and metaphors (Danesi 2003)18, as well as in interpreting hu-
mour and irony (Bromborek-Dyzman et al. 2010). As it is precisely in
these non-literal manifestations of language use that native speakers treat
linguistic means as a kind of prop for the complex “dynamic on-line, con-
structions of meaning that go far beyond anything explicitly provided by
the lexical and grammatical forms” (Fauconnier 1998: 251). Pavlenko
(2005) noted that conceptual proficiency is problematic even for ad-
vanced L2 users. Paradis (2009) noted that immersion students, for exam-
ple are notorious for fluency over accuracy. Kroll and Linck (2007) re-
ported on a study by Tokowicz et al. (2004) in which native English stu-
dents of Spanish as their L2 who spent some time in Spain made more er-
rors in spoken word translation giving any semantically related word if
they could not produce a target translation. Students who did not study
Spanish abroad either produced the exact translation equivalent or said, ‘I
don’t know’ valuing accuracy over fluency.
Raising awareness of the intricate connections in the bilingual mental
lexicon is an essential part of developing as a translator. Toury (1995)
pointed out that it is the ability to establish similarities and differences be-
tween the two languages (what he termed interlingualism) that is an es-
sential part of translation proficiency development. Yet, neither the term
interlingualism nor the process of becoming an interlingual individual
(Whyatt 2009c) in the process of developing translation expertise has at-

–––––––––
18
see Harden ( 2009) on translating conceptual metaphors and Pisarska (2004) on
linguistic aspects of metaphor translation.
98 Chapter 2

tracted much attention. Presas (2000) remains one of few scholars who
discussed the possible transformation in the functional reorganization of
bilingual knowledge as a consequence of developing translation compe-
tence (see also Whyatt 2006b).
The discussion of the bilingual foundations of translation ability
would be incomplete without relating the emerging complex picture of bi-
lingual knowledge to the L2 learner’s/user’s general cognitive structures.
This might demonstrate some important prerequisites vital for the devel-
opment of translation skill in its evolutionary continuity.

2.7. Cognitive effects of bilingualism

The amount of literature on the cognitive consequences of bilingualism is

impressive (see Bialystok 2005 for a review) and it has documented ac-
counts from a detrimental effect of bilingualism to a beneficial influence of
learning another language on the learner’s general cognitive abilities. Gol-
lan et al (2005) showed that since bilinguals use two languages they use
each of their languages less than monolingual speakers and in consequence
might have smaller vocabularies in each of their two languages than the
monolingual speakers of L1 and L2 (notice the monolingual view of bilin-
gualism). Bilinguals were reported to show the effects of cross-language
competition in the frequently experienced tip-of-the tongue phenomenon
than monolinguals (Gollan and Acenas 2004). Today’s view of bilingualism
is that of an enriching intellectual experience. Wittgenstein said, ‘the limits
of my language mean the limits of my world’19. Whorf (1956), the father of
linguistic relativity and himself “an avid language learner” (Pavlenko 2005:
436) indicated in his writings that learning another language “has the power
of transforming or enhancing the speaker’s worldview” (Pavlenko 2005:
436). As pointed out by Rosi Landi (1973: 33), whoever learns a new lan-
guage becomes a new person. Among the most frequently mentioned bene-
fits, there is enhanced metalinguistic awareness (Bialystok 2001) which
makes language learners more sensitive to their L1. As noted by Kesckes
and Papp (2000), language students frequently observe that the best way to
understand one’s own language is to learn another language.
Research into the developmental aspects of bilingualism (Clark 1978:
36) points to the fact that being exposed to two or more linguistic systems
–––––––––
19
Ludwig Wittgenstein Tractatus Logico-Philosophicus (1922).
 Bilingual foundations of translation ability 99

can make the structural patterns of language more noticeable and encourage
attention to the systematic features of language. As a result bilinguals show
increased metalinguistic awareness and ability (Tunmer and Myhill 1984).
Metalinguistic knowledge following Bialystok (2001) can be defined as a
more technical knowledge of the language. Although the actual role of the
learners metalinguistic abilities in L2 appropriation has not been extensively
studied, Roehr (2008) found a positive correlation between L2 proficiency
and metalinguistic knowledge. As reported by numerous studies bilinguals
are at an advantage over monolinguals when it comes to metalinguistic
awareness by the virtue of being exposed to two linguistic systems. The
metalinguistic knowledge, awareness and ability (three constructs men-
tioned by Bialystok 2001) is not linked to one language but it is transferable
in nature. In other words if somebody is a naturally good reflective commu-
nicator both in speech and writing in one’s native language and is able to use
language as a tool, appropriately and up to the point, fluent and accurate in
expressing ideas in compliance with the pragmatic code of language behav-
iour, all these skills will become transferred to the L2 use once the linguistic
level of L2 proficiency is achieved. As adhered by Paradis (2004/2009) and
discussed in section 2.2 the role of metalinguistic knowledge in L2 use by
those who acquire a second language through formal instruction and later on
in life as adults is frequently the primary source of knowledge about the L2
system. L2 users then will resort to their metalinguistic knowledge when
they experience gaps in their L2 linguistic competence. Consequently, it is
possible that the need to establish similarities and differences between the
learner’s two languages is a constant psycholinguistic process which ac-
companies second language acquisition no matter whether it is carried out in
a natural bilingual family setting or in an L2 language classroom. This proc-
ess in the bilingual mind probably is more subconscious in natural bilingual-
ism than in acquired bilingualism, more subconscious if teaching an L2 be-
gins early in one’s childhood and more conscious in adult L2 learners. If the
teaching methods are communicative rather than relying on explicit instruc-
tion, the level of an L2 learner’s meatlinguistic knowledge will differ. It will
be different in language course participants who learn a foreign language for
practical communicative reasons and in EFL students at university level
who study the language to become teachers or translators. Both these pro-
fessions necessitate high levels of metalinguistic knowledge. Although the
role of metalinguistic knowledge has not received much attention in Transla-
tion Studies, it remains undisputed that metalinguistic awareness and ability
100 Chapter 2

are part and parcel of a translator’s expertise where language is used as a

tool, and the use of language is reflected upon and judged as adequate, ap-
propriate, natural, inappropriate, inadequate, or unacceptable. As suggested
by Malakoff and Hakuta (1991) it is possible that metalinguistic knowledge
and the ability to use it in translation is one of the factors which distinguish
natural translators from professional translators. I shall return to these issues
in more detail in chapter 3.
Other positive effects of bilingualism include better performance on
tasks which require the suppression of irrelevant information and executive
control (Bialystok 2001). Recent findings on the cognitive effects of bilin-
gualism point to interesting and promising results. Bialystok et al. (2004)
noted that elderly bilinguals did much better than elderly monolinguals on
cognitive tasks which require the inhibition of irrelevant information. Kroll
and Linck (2007) attributed the age defying effect of bilingualism to the
constant exercise in the suppression of irrelevant lexical items. To quote,
“If one language must be inhibited to use the other, then bilinguals poten-
tially develop expertise in inhibitory control that then extends beyond lin-
guistic tasks to domain general cognitive performance” (Kroll and Linck
2007: 262). Recent neurolinguistic studies showed that the mechanism re-
sponsible for language control is the same as for the control of action in
general (De Groot and Christoffels 2006). Price et al. (1999) used the pre-
dictions of Green’s inhibitory control model (1986, 1993) and demonstrated
that subjects who were asked to translate words from German into English
showed activation in the area of the brain (recorded in Magnetic Resonance
Imaging – MRI) responsible for the control of action. This enhanced and
well trained skill of attention control (see Gopher 1992) in bilinguals is in-
dispensable in translation which involves multitasking in the conditions
where both languages are activated and while decisions on lexical selection
have to be made fairly quickly (or very quickly in simultaneous interpret-
ing). Gile (1995/2009) proposed the efforts model pointing out that apart
from comprehension and production effort, there is memory effort and co-
ordination effort which supervises the division of labour in the translating
mind. I shall return to these issues in chapter 5.
Further cognitive consequences of bilingualism include factors in-
volved in creative processes (Appel and Muysken 1996) which are essen-
tial for refining the skill of translation to reach the level of expertise
(Kussmaul 1991). Nida (1981, 2001) saw creative thinking as an essential
part of translators’ cognitive make-up. Research into the cognitive effects
 Bilingual foundations of translation ability 101

of bilingualism shows that the essential features of creative processes are

stimulated by acquiring another language. Bilinguals are considered to be
cognitively more flexible, better at lateral thinking and more efficient in
making remote associations as well as having “greater skill in handling
the linguistic code” (Presas 2000: 27) due to their metalinguistic aware-
ness. This view is consistent with Lambert’s (1978) opinion expressed
over thirty years ago in the following words: “It may be that their [the
translatotors’ BW] bilinguality, a prerequisite for membership in the pro-
fession, has the effect of providing them with special forms of intelli-
gence, sensitivity, and skills at finding out what is meant and what is im-
plied” (Lambert 1978: 132). This suggests another important cognitive
consequence of becoming bilingual that is the constant need to negotiate
relationships and connections between a bilingual’s two languages and the
general knowledge store. Both the process of learning an L2 and the early
translation experience show that this relationship is in fact very complex.
The general question which still remains to be addressed is the rela-
tionship between language and knowledge, or in a bilingual context
knowledge and the two languages of a bilingual person. Wilss (1996)
pointed out that the role of language in general education is probably still
underestimated. What remains undisputable is the fact that once we ac-
quire language it is through language that we acquire knowledge of vari-
ous areas and subjects. At schools knowledge is imparted through lan-
guage, acquired and tested through language during tests and exams. Ac-
cording to the principle of functional encoding (see Beatens Beardsmore
1982) the knowledge acquired or learned in one language will be more
readily available in the language in which it was acquired than in the
other language in which the bilingual person might simply lack words to
express it (cf. Grosjean’s Complementarity Principle 2002 discussed in
section 2.3.1), or access to the specific vocabulary might be delayed due
to the recency effect according to which words used less frequently take
longer to retrieve in the lexical access process.
Kolers (1973) conducted an experiment in which he wanted to meas-
ure brain function during the translation process. His subjects learned to
recite their alphabet backwards in one language but they failed to be able
to do the same with similar fluency in their other language. Although the
subjects whose two languages share the same script (the same letters in
the alphabet), German and French speakers with English as their second
language were better at reciting the alphabet backwards in English than
102 Chapter 2

the native speakers of Arabic and Korean. This shows that the skill
learned in one language will not necessarily be equally accessible in the
other language. Knowledge acquired in one language, e.g., mathematical
skills will be more readily available in the language in which it was ac-
quired. Korzeniowska (1998) points that even at a very high level of bi-
lingualism people usually prefer to count in one language rather than the
other. For this reason Cummins (1991) points to the need to differentiate
between language proficiency and academic proficiency as determinants
of educational success. Cummins (1991) suggested that a distinction
should be made between global language proficiency and academic profi-
ciency. Deriving his observation from a bilingual education context he
concluded that the educational success of a bilingual learner does not de-
pend on the Basic Interpersonal Communicative Skills (BICS) which take
about 2 years to acquire, but it is more dependent on Cognitive Academic
Language Proficiency (CALP) which takes 5-7 years to acquire. His ob-
servation is valid for the present discussion in the light of literacy related
knowledge. It is also significant in the perception of translation as a
knowledge-based activity (Wilss 1996) in which bilingual knowledge is a
vital but not the only kind of knowledge applied. These processing diffi-
culties referring to language imbalance and asymmetrical connections be-
tween the lexical stores of the two languages and the general knowledge
store (CUCB) will have to be negotiated in developing translators and
handled in expert translators. I shall discuss my proposal of how it might
be achieved in chapter 5 by proposing a Knowledge Integration Network
(KIN) according to which the process of developing translation expertise
is a process of learning to integrate knowledge (language knowledge and
general knowledge).

2.8. Conclusions

The purpose of chapter two was to discuss the bilingual foundations on

which translation expertise can be built. Translation Studies have taken for
granted that potential translators have mastered their two languages to na-
tive-like standards remaining quite unaware of the complex body of knowl-
edge and skills involved in the process. It is fundamental in this work that
both terms ‘a bilingual’ (defined in a broad sense (see Kroll and Linck
(2007: 238) referring to all individuals who can use more than one lan-
guage for communication) and a ‘translator’ should be perceived as devel-
 Bilingual foundations of translation ability 103

opmental constructs and not as constant stable entities. The failure to accept
that both bilinguals and translators are constantly subject to developmental
changes in the level of their L2 proficiency and generally in terms of their
bilingual knowledge is possibly the root of misunderstandings about bilin-
gualism (Grosjean 2002) and translation as a skill socially expected of bi-
linguals (Beatens Beardsmore 1982). Although nowadays the majority of
the world’s population is bilingual or multilingual (Li Wei 2000) the need
for translation services is still continuing to grow.
The nature of language in general as a sophisticated and intricate sys-
tem of communication assigned to a particular language community shows
that one is dealing with an overwhelming amount of knowledge and skill.
Language as knowledge can be easily defined as a rule governed code in
which words can be combined in sentences which can carry meaning. Lan-
guage as social practice (Sfard 1998), however is much more difficult to
define and its rules are in fact reliant on subtle data that all users of the
same language come to share and understand by interacting with one an-
other in society. The complexity of language becomes apparent when one
comes to acquire or learn another language. In natural bilinguals two lin-
guistic systems develop simultaneously and are used in a complementary
manner in response to the current communicative needs or situations which
require the use of one or the other language. For obvious reasons nobody
can speak two languages at the same time. In consequence, natural bilin-
guals are not, as it is still commonly expected equally proficient in their
two languages (Grosjean’s Complementarity Principle 2002). If they were
they would potentially make the best translators. As it is, in their bilingual
knowledge they probably have excellent conceptual proficiency but might
lack lexical items especially in areas outside their customary language use.
When a foreign language is learned after the native language has been,
at least partially acquired, it is initially learned as a set of knowledge, an al-
ternative code for communication. When two languages start to live in one
mind, they constantly compete for dominance. L2 learners and users ex-
perience interference at all levels of language, i.e., when lacking compe-
tence in some areas (grammar, lexicon, discourse patterns) they may trans-
fer their L1 knowledge and do the best they can to communicate their in-
tended meanings. As suggested by Cook (1992) they develop multicompe-
tence. Their bilingual memory starts to store native and foreign words for
the same concepts. When speaking in their L2 learners initially translate
their L1 words into their L2 translation equivalents. In comprehension they
104 Chapter 2

frequently have to first translate what they hear into their L1 words and
only then they can understand what is being communicated. With growing
proficiency the processes of comprehension and production gain speed and
the lexical links are used less while the conceptual connections become es-
tablished and bring visible benefits for the speed of L2 processing. The L2
learners get better at inhibiting their L1 when speaking in their L2 that is
they get better at language control (Green 1986, 1993) and their L2 per-
formance becomes somehow speeded-up or automatized to the point that
they are even able to monitor their performance and spot and repair transfer
errors. Yet, they are still frequently unable to match native language per-
formance especially in terms of conceptual proficiency. Even if they sound
fluent they frequently lack the L2 conceptual competence, they make word
usage errors, or sound too formal or too informal in a particular situation.
Even if they have acquired an extensive knowledge of the L2, they still fal-
ter when it comes to L2 as social practice and their language behavior
might be perceived as odd by native L2 users (e.g., Polish L2 users are fre-
quently perceived as impolite failing to respect the English concept of pri-
vacy). What is more, recent studies into L2 users’ L1 reveal the impact of
L2 on the use of L1 (Poulisse and Bongaerts 1994, Cook 2003). Ewert
(2009) investigated the impact of learning a second language on the knowl-
edge of one’s native language and concluded that, “there is a qualitative
change in L1 knowledge with increased L2 proficiency and length of expo-
sure to the L2” (Ewert 2009: 147). As a result the L1 of an L2 user is not
the same as the L1 of a monolingual native speaker.
Bilingual knowledge, however, is not a sum of knowledge ascribed to
L1 and L2 monolinguals at all levels, linguistic and conceptual. A bilin-
gual person is a unique individual who can speak and understand either of
his/her language, who experiences language interference, can switch from
one language to the other or mix and blend them in one sentence as well
as use the two languages to translate between them. The fact that all bilin-
guals experience interference shows that it is impossible to switch off one
of the two languages which, if not kept under control, will show up in ar-
ticle misuse or in an unintended lexical or conceptual transfer error.
Translation as an activity requiring the use of two languages is likely to
reveal a lot of information about the relationship between the two lan-
guages of a bilingual individual. As chapter 3 will demonstrate when
translation is performed by untrained natural translators it shows all the
week aspects of one’s bilingual knowledge. By the need to activate both
 Bilingual foundations of translation ability 105

language systems held together by the common underlying conceptual

base, translation might create favourable circumstances for cross-
linguistic interference at all levels. Can untrained translators exercise suf-
ficient cognitive control to keep the two languages safely apart and avoid
interference? Is the frequently low conceptual proficiency of L2 learners
visible in their translation performance? Is translation really the fifth skill
of all bilinguals? Do we need to train translators or is natural translation
ability sufficient for effective intercultural communication? These and
other related questions will be dealt with in chapter three which aims to
discuss how the human skill to translate is manifested as an untrained
ability of bilinguals acting as natural translators.
Chapter 3

… contrary to popular opinion, translation has little

to do with fluency, and bilinguals range from being
very poor to being very competent translators.
Grosjean (1982: 257)

Translation as an untrained ability

The purpose of this chapter is to discuss the initial stages in translation

skill development when translation is performed as an untrained ability by
people who can communicate in their two languages. Maria González
Davies (2004) in her book entitled, Multiple Voices in the Translation
Classroom, described this stage as unconscious incompetence in the sense
that those who perform translation are unaware of the intricacies involved
in the process (González Davies 2004: 40). In today’s multilingual com-
munities translation is an everyday occurrence and the ability to translate
is socially expected of bilinguals. Translation then is the fifth skill apart
from the four skills of speaking, understanding, listening and writing in
the two languages of a bilingual person. As suggested in chapter one peo-
ple are predisposed to communicate with others and this predisposition
includes communicating via translation. The chapter will focus on two
groups of translators who most frequently demonstrate their untrained
ability to translate, natural translators and L2 learners. Discussing how the
human ability to translate is manifested by those users of L1 and L2 who
translate without any prior training for it will demonstrate the use of bi-
lingual knowledge for interlingual communication. It will also prepare the
ground for discussing the further development of translation as a trained
skill expected of professional practicing translators (chapter 4).

3.1. Dispute over bilingual knowledge and translation capacity

In 1978 Harris and Sherwood published an article entitled “Translation as

an innate skill”. The article reported on case histories of bilingual children
(see Harris 1980 on “How a three-year-old translates”) who without any
training or instruction were performing the so called ‘natural translation’.
108 Chapter 3

Natural translation understood as, “the translating done in everyday cir-

cumstances by people who have had no special training for it” (quoted after
Toury 1995: 241) led to a hypothesis that “the basic ability to translate is an
innate verbal skill” (Harris and Sherwood 1978: 155). The developmental
stages in the unfolding of translational behaviour in naturally bilingual
children were suggested by Harris and Sherwood (1978: 155) as follows:

1) Pre-translation characterized by an occurrence of single word labels

from the two languages naturally spoken in the home environment of
the child accompanied by a conceptual bridge, a non-verbal entity
(compound bilingual memory model). The characteristic features re-
ported by Harris involved “the unconsciousness of the whole process
and the small scale and low rank” (single words, after Toury 1995:
243), and the fact that it was considered functionally redundant.

2) Autotranslation characterized by spontaneous functionally redundant

“translating of what one has just said in one language into another”
and directed either to oneself (intrapersonal autotranslation) or to oth-
ers (interpersonal autotranslation)

3) Transduction that is socially functional translation where “the transla-

tor acts as an intermediary between two other people” either within
the family or outside the family (Harris and Sherwood 1978: 165).

The above successive stages of the tentative model were claimed by Har-
ris to be universal and taken as evidence that translation can indeed be re-
garded as, “humanly innate” resulting in a “specialized predisposition in
children” (Harris and Sherwood 1978: 168).
The claims made by Harris (1977) and Harris and Sherwood (1978)
together with the statement that natural translation should be a primary
source of data for “the scientific study of translating” instigated a heated
debate about the nature versus nurture factors in the skill of translation
between Harris and translation scholars within a newly established disci-
pline of Translation Studies (Krings 1992, Harris 1992). Toury (1995) in
his seminal book, Descriptive Translation Studies and Beyond, said, “To
be sure, the Harris-Sherwood model was not devised as an overall ac-
count of the emergence and development of translating as a human skill.
Rather it was confined, right from the start, to very young natural transla-
 Translation as an untrained ability … 109

tors. Even within those limits the adequacy of the model seems question-
able” (Toury 1995: 244).
The major criticism which followed included little concern for the cir-
cumstances under which the transition from one phase to another occurred,
concentrating on one single variable, that of age and equating biological
age with bilingual age in order to conclude that “translating is coextensive
with bilingualism” (Harris and Sherwood 1978: 155) and proceeds from
spontaneous and functionally redundant translating (see also Albert and
Obler 1978: 218) to socially functional translating. Other translation schol-
ars such as Lörscher (1986), Wilss (1982) and Krings (1992) expressed
their concerns that translating as a skill allowing for fluent translation can-
not be conceived as given by the mere fact of being bilingual. To quote,
“the identification of translating as a skill with mere bilingualism seems an
unwarranted oversimplification. After all, there is no answer in it to the
crucial question of what it is that brings forth the unfolding of the skill (that
is, its realization in actual behaviour) and the way (or ways?) it then devel-
ops” (Toury 1995: 245-246). Harris received severe criticism but the con-
cept of natural translation entered the debate about the genesis and evolu-
tion of translation as a human skill, and as expressed by De Groot (1997)
remains valid as what comes as natural does not need to be trained.

3.2. Translation as a natural ability of bilinguals

According to Harris and Sherwood (1978), “all children can translate in all
cultures, in all languages and registers, throughout history and from the
time the individual starts to acquire a second language” (cf. Malakoff and
Hakuta 1991: 144). In this respect natural translation is not a “learned skill,
such as learning a foreign language in school, but, rather, it is a skill which
is developed from a natural and existing base, similar to the development
that occurs in mother-tongue language abilities” (ibid.). As pointed out in
chapter 1 the word ability and skill are used as synonyms. Consistently, as
any natural ability it should be open to improvement under appropriate
guidance, although Harris (1977, 1980) does not elaborate on the issue of
further development in natural translation ability. Similar scant attention is
paid to individual differences among natural translators. The possibility that
natural translation is perhaps an intriguing game for developing bilingual
minds or a just a communicative strategy dictated by the circumstances is
not considered, yet by all means plausible (Malakoff 1992).
110 Chapter 3

3.2.1. Studies in natural translation

As pointed out by Malakoff and Hakuta (1991), “For many bilingual chil-
dren throughout the world, translation is an everyday activity, a part of their
lives as bilinguals (Grosjean 1982). Yet translation is a poorly understood
phenomenon” (Malakoff and Hakuta 1991: 142) and there is very little em-
pirical literature on translation as a natural ability of bilinguals, especially of
bilingual children. The evidence that is available either comes from “anec-
dotal” reports by linguists (Leopold 1939-1949), or indirectly from studies
which used translation as a research technique (Harley 1986, Swain Naiman
amd Dumas 1974, Kroll and Stewart 1994, De Groot 2002) without focus-
ing on the actual translation ability. Some of the reasons for lack of interest
in natural translation abilities were discussed in chapter 1. Malakoff and
Hakuta (1991: 144) quote a number of studies (e.g., Shannon 1987) which
reported on young children acting as natural translators supporting in this
way the existence of natural translation ability put forward by Harris (1977).
Malakoff and Hakuta (1991) tested Harris’s (1977, 1980) claim of natu-
ral translation (not to be confused with professional translation) as a skill.
The authors presented two studies of translation ability which provided
“empirical support that late elementary-school students are able to produce
good written and oral translations”. The study investigated Puerto Rican
Children in New Haven, Connecticut with respect to their translation abili-
ties. The subjects were divided in two groups for study 1 on properties of
translation ability and Study 2 on distribution of translation ability. The first
study investigated 16 subjects (8 girls and 8 boys fourth- and fifth-grade,
mean age 10.7 years) who apart from being bilingual were also reported to
have had some experience in translating for their family or peers. The sec-
ond study investigated a randomly selected group of subjects (27 girls and
25 boys) from fourth- and fifth-grade bilingual education classes. Prior to
the experiment the language proficiency of the subjects in both languages
was assessed by the Woodcock Language Proficiency Battery and an online
analogies task was created and administered in each language. Translation
tasks for the first study included word, sentence and story translations in
both directions from Spanish to English and from English to Spanish. Sen-
tences and stories were constructed to minimize vocabulary complexity and
to “provide ample opportunities for grammatical pitfalls that would cause
(intrusion errors, i.e., errors in which the source-language structure intrudes
into the translation)” (Malakoff and Hakuta 1991: 153). The word and sen-
 Translation as an untrained ability … 111

tence translations were done orally as an on-line task with the stimuli for
translation appearing on a computer screen. A subject was asked to read the
source material and provide the translation which was tape-recorded to en-
able assessment of the accuracy of the translation. Sentence translation was
carried out in a similar on-line fashion but an additional element was ma-
nipulated. Some sentences were administered in a straight and some in an
‘imagery’ condition where a sentence would appear on the screen first fol-
lowed by a “thought balloon” and an instruction for the subject “to make a
mental picture of the sentence”. The imagery condition was meant to defy
literal translation by expanding the “window space” for translation, whereby
the meaning of the sentence would be processed to a greater extent and
grammatical pitfalls through literal translation would be avoided (Malakoff
and Hakuta 1991: 154). The story translation was done in writing. In addi-
tion a word identification task was included as an attempt to ‘assess transla-
tion proficiency’. The subjects in study 2 were asked to translate in the writ-
ten format the same story which was translated by the subjects in Study 1.
The translations by both groups of subjects were coded and compared in a
detailed error analysis.
The results obtained in study 1 allowed for the following conclusions:

1) The subjects were extremely good translators and made few errors in
both source-target directions (however, morphological errors were ig-
nored as in the authors’ opinion they did not affect major meanings of
the translations)

2) The imagery manipulation did not have an effect on whether literal or

non-literal translations were made (a vast majority included literal
translations irrespective of the direction. Malakoff and Hakuta (1991)
noticed that the lack of effect may be due to the “ineffectiveness of the
imagery manipulation” which proved tedious and tiring for the sub-
jects who had to “form images continually across a large number of
sentences” (Malakoff and Hakuta 1991: 155).

3) Translation was more efficient into English than into Spanish (so re-
flecting English language dominance)

4) For word translation the speed is predicted by proficiency in the target

language rather than in the source language, although the pattern was
112 Chapter 3

less clear for sentence translations. The authors concluded that possi-
bly SL proficiency plays “a greater role as the unit of language that
needs to be processed gets larger” (Malakoff and Hakuta 1991: 156).

The general conclusion of the study was that in addition to proficiency in

the two languages, “there appears to be a translation proficiency” predict-
ing translation speed in terms of accessibility of the two lexicons meas-
ured by the word-identification task. This conclusion led to a proposal
that translation skill apart from proficiency in two languages requires “an
additional component of accessibility of the two lexicons” (Malakoff and
Hakuta 1991: 156).
The results obtained in study 2 showed that the error patterns revealed
similarities between both groups. The less selected group in terms of
translation experience made fewer errors when translating from English
into Spanish. Malakoff and Hakuta (1991) explain this surprising result
saying that all Study 2 subjects were in bilingual education programmes
where both languages are used as the language of instruction. Study 1
subjects were not in the bilingual education programme at the time of the
experiment, although all but one subject were in a bilingual education
programme “at some point” and because of the subtractive1 nature of their
bilingualism in the community they “reach a plateau in their Spanish pro-
ficiency” (Malakoff and Hakuta 1991: 161). This in turn might have con-
tributed to the fact that the subjects in the more selected group were not
significantly better in their translation ability. Generally the study showed
that, translation skill is present in bilingual children. As the authors con-
cluded, “Bilingual children are able to translate, albeit with flaws, and
their translations reflect their understanding of the communicative impor-
tance of translation”. Making reference to the discussion of the bilingual
foundations of translation skill, it is possible to say that natural translators
focus on the conceptual level and the structural level (lexico-grammatical)
may be compromised. In a functional sense though natural translation
serves the communicative goal and bilinguals are able to perform com-
municatively adequate translations if one accepts the view that it is possi-
ble “to communicate meaning in the absence of correct sentence struc-
–––––––––
1
Depending on the social status of a bilingual’s two languages bilingualism has
been defined as additive (adding prestige, i.e. in the case of English for learners of Eng-
lish in Poland), or subtractive (perceived as minor and looked down upon, i.e. Polish for
Polish immigrants in the USA in the early 20th century).
 Translation as an untrained ability … 113

ture” (Malakoff and Hakuta 1991: 143). Unfortunately, it is unclear if the

subjects in the study were natural bilinguals (born in a bilingual family)
or whether they came to acquire English as their second language outside
their monolingual family. The latter case is more likely judging by the
range of errors reported in the study. The lack of clearly defined language
acquisition history does not undermine the obtained results and the study
demonstrated decisively that bilingual children whether natural or L2
learners/users are able to translate, that is, produce communicatively ade-
quate translations. To quote,

In natural translation, linguistic sophistication and explicit knowledge

of contrastive linguistics is generally not the norm, especially among
grade school children. Although children of this age can speak two (or
more) languages correctly, they do not have a conscious awareness of
the specific differences between language systems. Despite the absence
of such linguistic knowledge, children are able to communicate mean-
ing; although the meaning may be embedded in poor sentence structure
(Malakoff and Hakuta 1991: 150).

In professional translation such an approach is perceived as highly unpro-

fessional. A translator is more likely to miss the intention of the SL text
author than form an unacceptable sentence structure. Translation espe-
cially written as a purposeful communicative activity is about the transfer
of meaning from a SL text into a TL text but in the form that is acceptable
and appropriate for the TL users (i.e., “in the natural form of receptor lan-
guage” (Larson 1984: 17)).
This points to the structural difference between natural translation and
professional translation although both kinds of translation respond to
communicative needs and help to overcome language barriers. It cannot
be left unacknowledged that most natural translation is done orally in
every day communicative situations (cf. community interpreting) where
the extralinguistic factors such as the contextual setting, the communica-
tive purpose and the participants provide strong support for the produc-
tion and interpretation of utterances. Natural translation is done for in-
house use and its quality is not an issue to be discussed or analyzed. One
might say that the form becomes secondary and the actual words and
structures serve only as a kind of prop for the online construction of
meaning which relies heavily on the context and the backstage cognition
of the participants (see Fauconnier 1998) as well as on the cooperative
114 Chapter 3

principle and the maxims of communicative behaviour (Grice 1975, Gutt

1991, Sperber and Wilson 1986). The actual form may be grammatically
deficient and left as such which points to the lack of metalinguistic moni-
tor to ensure correctness. This however, is something the participants of
such interactions are prepared to ignore as we all sometimes ignore slips
of the tongue, vagueness and the poor sentence structure of our interlocu-
tors as long as the rich pragmatic context allows us to fish out their com-
municative intentions. Natural translations are usually performed in more
informal, more private contexts where the forms which carry meaning are
not a priority, where there is a high tolerance of linguistic inadequacies
and transfer errors. In public official situations (e.g., political visits, court-
room hearings or business negotiations) the context requires a high level
of formality, flawless unambiguous translation in terms of content and
form. These requirements cannot be met by natural translators as they are
not within the reach of the untrained ability to translate. Such require-
ments call for professional translation services.
Gómez (2006) tested Harris’s (1977) tenet that translation is a natural
skill in bilinguals looking at written translation. The subjects were natural
bilinguals (born to bilingual parents) and advanced native Spanish stu-
dents of English taking their last semester of a translation programme. All
the subjects were aged between 18 and 25 (average 22.5). Both groups
were involved in English to Spanish translation and vice versa. The re-
sults obtained in the study showed that natural bilinguals took longer
when translating in both directions and they used more printed sources,
whereas translation students used only the Internet. No bilinguals and
only 50% of the translation students read the SL text prior to translating it
into Spanish. About 50% of each group revised the text after they had fin-
ished translating it. However, the differences were not significant when
translating into English. Bilinguals made more pauses and corrected their
translations more frequently. They used twice as many full words (the au-
thors are not clear what they mean by full words but presumably it means
content words) when translating into English and three times as many full
words when translating into Spanish than the translation students. In the
translation quality assessment which followed all the translations were
graded by 3 evaluators. The translation students received better grades
when translating into Spanish (their native language) whereas bilinguals
scored better when translating into English. The author concluded that
translation training modifies the translation students’ behaviour which re-
 Translation as an untrained ability … 115

flects developing expertise in the L2 to L1 direction. The fact that the

changes are not reflected in the quality of L1 to L2 translation (Spanish to
English) was explained as probably due to their “meager command of
English, excellent for everyday usage, but still lacking for a professional
endeavour” (Martin 2006: 7).
Both studies (Malakoff and Hakuta 1991 and Gómez (2006) con-
firmed that bilinguals are able to translate in the sense that they can per-
form the task either in oral or in written modality. Both studies indicated
that the formal quality of translations produced by natural translators in-
dicates that apart from bilingual knowledge there are other capacities
which are involved, and which probably would improve the formal qual-
ity of the product. However, the results also indicated that translation pro-
ficiency in the sense of the capacity to produce not only communicatively
effective but formally adequate flawless translation “goes beyond the sum
of the two language proficiencies”. Malakoff and Hakuta (1991) further
suggested that “it may well be that translation ability is related to metalin-
guistic skills, a hypothesis that must be explored in future research”
(Malakoff and Hakuta 1991: 157). The third factor, strategies which
Malakoff and Hakuta mentioned as important for translation proficiency
will only improve the performance as long as language proficiency and
metalinguistic knowledge have reached an optimal level of attainment.
This possibility is corroborated by the results obtained by Gómez (2006)
with translation students who although equipped with a range of strategies
which they must have acquired through training scored worse than natural
bilinguals in a Spanish to English translation task. Yet, they performed
better in forward English to Spanish translation. These results point to the
importance of feeling at home with the two languages that one uses in
translation though as indicated in chapter two achieving this kind of ulti-
mate attainment is not an easy task.
Some translation scholars share the opinion of the composite character
of translation as a skill. Lörscher (1997: 2014) phrased the innate predispo-
sition as a “rudimentary ability to mediate” stressing that it cannot be per-
ceived as a guarantor of the skill which allows to translate with facility and
without heavy interference and in accordance with particular normative re-
quirements. To quote Toury (1995: 245) translation skill “is precisely a
matter of the development not unfolding of the skill” as this would simply
mean that every bilingual could become a skilled translator (Nida 1981).
Toury (1995) is decisive about the skill of translating saying that
116 Chapter 3

It would seem much more convincing to argue that some additional fac-
tors are needed in order to trigger off the “specialized predisposition”
for translating and set in motion – most probably, a certain combination
of personality and environmental circumstances (Toury 1995: 245-246).

This indication of additional factors points to the notion of translation as a

complex skill where language proficiency in both languages is necessary
but not sufficient to translate between them without formal flaws. Yet, to
say that translating is not a natural bilingual skill would mean to deny
empirical facts.

3.2.2. Natural translators are translators

“There are thousands of children in the United States referred to as lan-

guage brokers who have the responsibility of translating for their immi-
grant parents. Language brokers have broad roles as mediators and deci-
sion-makers” (De Ment et al. 2005: 255). They are asked to speak for oth-
ers because they can do it, they can interpret and they frequently have to
accept the grown up roles as multicultural communicators, interpreters
and translators (Shannon 1990). Interpreting occurs between all kinds of
languages (including sign language) whenever a message originally ex-
pressed in one language is on the spot retransmitted in another language
(cf. Anderson 1978: 218 and Carr et al. 1997: 211). Indeed, it is only quite
recently that researchers have started to pay attention to non-professional
translation commonly referred to as language brokering (see Morales and
Hanson 2005 for a review of the literature).
The social function natural translators are able to fulfill as interlingual
communicators should not be underestimated. If it was not for the interlin-
gual capacity of bilinguals to mediate meanings across language barriers
everyday life in multilingual communities would be impossible to imagine.
If for every interlingual encounter professional translators were needed
possibly half of the world’s population would have to undergo translator
training, and as it will be discussed in chapter four the training itself is only
the staring point in the development of translation as a professional skill.
What is more, in many respects natural translators make better community
interpreters than trained translators. Coming from bilingual communities
themselves they have a better understanding of the background knowledge
which is essential to understand what is being said and what needs to be
communicated, and what is likely to be misunderstood by an interlocutor
 Translation as an untrained ability … 117

who usually has a different social and cultural background (Bullock and
Harris 1997, Harris 1997). Of course, when examined under the cool eye of
a linguist these linguistic exchanges may seem imperfect, erroneous, full of
interferences and generally formally inadequate, but so is the spontaneous
speech of bilinguals (see Grosjean 2002) and sometimes even of monolin-
guals. What matters is getting the message across and since most natural
translation is done orally and falls under the category of community inter-
preting any ambiguities and misunderstandings can be clarified on the spot.
Like in any other social encounter the L1 speaker, the natural translator and
the L2 interlocutor will cooperate to communicate what is required in the
situation whether the situation involves talking to an immigration officer, a
medical doctor or any other speaker. Somehow it is deeply engrained in
human linguistic interaction that interlocutors do no give in until they feel
that the meaning has been negotiated and the communicative aim has been
achieved. Natural translators take translation for what it truly is, a commu-
nicative strategy, a tool that you have to use when language is a barrier,
when two parties that need to communicate information to each other lack
the knowledge of a common system of communication which allows to
share meaning/sense/ideas/intentions.
Natural translation then is a communicative strategy, a tool which
can be used with a range of skill and the only test of its communicative
quality is how it makes communication possible across language barri-
ers. Being natural it can lack all the professional refinements, all the
metalinguistic care for formal correctness or stylistic embellishment.
Natural translators are probably far from seeing translation as an opera-
tion on languages or even utterances, far from the structural attitude of
seeing language as a system because language as knowledge of the sys-
tem is secondary. Language as a social practice is of primary pragmatic
importance (Hickey 1998). This communicative, functional adequacy af-
ter all is precisely what Translation Studies see as the desired feature of
translation in general and of professional translation in particular (Ver-
meer and Reiss 1984 and their skopos theorie). Let us support the valid-
ity of this attitude to translation with a brief selection of quotes by those
translation scholars who made fundamental contributions to our under-
standing of translation:
Translation to put it briefly, is not just an exchange of words and struc-
tures, but a communicative process that takes into consideration the
reader of the translation within a specific culture (Kussmaul 1995: 1).
118 Chapter 3

In effect in translation we do not operate with sentences at all but with

utterances, i.e., units of discourse characterized by their value in com-
munication. In certain types of translation then, it is both possible and
necessary to aim at equivalence of pragmatic meaning at the expense of
semantic meaning. Pragmatic meaning overrides semantic meaning in
these cases. And we can then consider a translation a primary pragmatic
reconstruction of its original (House 1997: 31).

Translation is more a process of explanation, interpretation and refor-

mulation of ideas than a transformation of words: the role of language
is secondary, it is merely a vector or carrier of thoughts (Newmark
1988: 72).

Undoubtedly, the functional aim in natural translation is in line with the

above views and with the reservation made by Malakoff and Hakuta
(1991: 143) that it is possible “to communicate meaning in the absence
of correct sentence structure” in language whose form may be far from
“the natural form of receptor language” (Larson 1984: 17). More so,
since a lot of community interpreting done by natural translators is car-
ried out in stressful situations including police arrests, seeking medical
assistance, solving neighbourly conflicts, etc. the control over the use of
language is likely to be taxed anyway (see chapter two on language con-
trol) and more likely to lead to slips of the tongue and language mis-
takes. Because of the services they provide natural translators deserve to
be recognized as members of the translating community. As observed by
De Groot (1997),
the phenomenon of natural translation, being coextensive with bilin-
gualism, (…) may have implications for the training of translators and
interpreters. What comes naturally need not be trained. Therefore, ex-
amining natural translations to understand their characteristics and to
determine the knowledge and skills they manifest may lead to changes
in existing translation and interpretation curricula. The time gained
could then be used to devote more attention to those aspects of these
skills that do not come naturally but require formal instruction (De
Groot (1997: 29).

To sum up, it is possible that Harris’s (1977) tenet that translation is a

natural predisposition in bilinguals caused needless debates which could
have been avoided if the words ‘translation ability’, ‘translation skill’ and
‘translation competence’ were not used as interchangeable synonyms, but
as suggested in chapter 1 as points on the developmental continuum of
 Translation as an untrained ability … 119

translation as a human skill. Equating ability with skill and competence

leads to false conclusions that ability guarantees competence, or expertise.
It is much better to consider the three in terms of a developmental contin-
uum. As pointed out by Sternberg the continuity of the development as-
sumes that “human abilities are forms of developing expertise” (Sternberg
1998: 11). If Harris and Sherwood’s (1978) claim had been phrased,
‘translation as an innate predisposition’ then there could be “very little
quarrel with the argument that a predisposition for translating, which may
indeed be coextensive with bilingualism, is “part and parcel of mankind’s
basic linguistic equipment” (Wilss 1982: 39). The powerful social need to
communicate makes bilinguals act as intermediaries for those for whom
language is a communicative barrier. They use translation as a communi-
cative tool and it is perhaps not that translation is a natural innate skill but
a necessity which enables people to fulfill a stronger human need, the
need to communicate with others. Malakoff and Hakuta (1991) concluded
their study with practical implications pointing to the fact that since the
ability to translate is expected of bilingual children it can be used as a tool
for research and language proficiency assessment (see Swain, Dumas and
Naiman 1974). This natural ability can be then an ally in education where
it can be used as “a psycholinguistic and sociolinguistic hook into ampli-
fying the bilingual skills of students. It therefore should be used to en-
hance linguistic awareness and pride in bilingualism, particularly for mi-
nority bilingual children whose home language is not valued by the ma-
jority culture” (Malakoff and Hakuta 1991: 163).
The next section will be devoted to another unrecognized group of un-
trained translators, foreign language learners who display yet another
facet of translation as a natural ability. By examining this facet it is hoped
that some light can be shed on the initial stages of the evolution of trans-
lation as a human skill.

3.3. L2 learners as natural translators

If we accept that translation is a natural skill of bilinguals, there seems to

be no reason why this natural skill should be denied to those bilinguals
who come to appropriate their L2 via formal learning. In view of the all-
inclusive definition of a bilingual (Linc and Kroll 2007) this extension of
the term ‘natural translator’ seems justified. This section will examine the
nature of translations produced by L2 learners who, similar to natural bi-
120 Chapter 3

linguals and immigrant L2 users, frequently exercise their untrained abil-

ity to translate. Yet, translation as the fifth skill of an L2 learner has a
complex history in language teaching theories and methodologies (see
Vermes 2010 for a brief overview of the pros and cons).

3.3.1. Translation as the fifth skill of L2 learners

A lot has been written about the unfair treatment translation as a skill has
received from SLA theories and SL teaching methodologies. To quote,

Despite the widespread popular assumption that translation should play

a major and necessary part in the study of a foreign language, twentieth-
century theories of language teaching and learning have at best ignored
the role of translation, and at worst vilified it. From the turn of the cen-
tury onwards almost all influential theoretical works on language teach-
ing have assumed without argument that a new language (L2) should be
taught without reference to the student’s first language (L1) (Cook
2001: 117).

The rejection of translation tasks by L2 teaching methodologies is possibly

an after effect of the Grammar-Translation Method (based on the so-called
scholastic method involving reading foreign language texts with a dictionary
and writing down a translation). As observed in the quote above banning
translation tasks from the L2 classroom was also a straightforward conse-
quence of the SLA theories (see Cook 2001: 119) and most notably of the
Direct Method where teaching L2 excludes the use of the learner’s L1
(Whyatt 2009b: 181). To quote, “The Grammar-Translation Method with its
focus on formal accuracy, often using translation not only in exercises but
also for testing, tormenting generations of language learners with texts of
absurd complexity only to show them how little they knew, is without any
doubt an approach which does not merit a revival, (…)” (Witte et al.
2009: 1). When translation is understood as transcoding, as it was the case
in the Grammar-Translation Method (Howatt 1984) which was used to
teach foreign languages since the Roman Times (see Malakoff and Hakuta
1991), it entirely misses the concept of translation as intercultural communi-
cation. As observed by Malmkjaer (1998: 3), the Grammar Translation
Method might have worked well with scholarly, studious people with “an
analytical bent who enjoy learning grammatical systems” and whose main
aim was to learn to read foreign texts. However, when applied to the teach-
ing of modern languages in schools where students with varying abilities are
 Translation as an untrained ability … 121

grouped together, it failed. As observed by Howatt (1984: 133) the Gram-

mar Translation Method mirroring the teaching of Latin placed the emphasis
in the teaching of modern languages also on grammar and written language
with complete disregard for spoken language (for the history of the method
see Malmkjaer 1998). With priority given to speech in the early 19th century,
the usefulness of the Grammar Translation Method was questioned and to-
gether with the rejection of the L1 by communicative methodologies, trans-
lation as the fifth skill of a bilingual was banned from the L2 classroom.
As observed by Cook (in Baker 2001: 119) the Direct, Communicative
Method demanded that L2 learners’ attention should be focused on commu-
nication and meaning rather than on form in the hope that such an approach
will stimulate the subconscious acquisition of the L2 system (Krashen
1982). “Translation which implies conscious knowledge of two language
systems and the deliberate deployment of both, is not among the activities
compatible with this belief” (Cook in Baker 2001: 119). In consequence
with no value attached to the learner’s L1 translation had no role to play and
was even considered by some as potentially harmful to the L2 learning
process. Malmkjaer (1998) in her introduction to Translation and Language
Teaching reviews the objections raised by anti-translationalists including
Lado (1964) and Gatenby (1967) as to why translation tasks should not be a
part of foreign language teaching. Malmkjaer (1998: 6) compiles the follow-
ing list of arguments raised against the use of translation in the foreign lan-
guage classroom including the opinions that translation:

– is independent of the four skills which define language competence:

reading, writing, speaking and listening
– is radically different from the four skills
– takes up valuable time which could be used to teach these four skills
– is unnatural
– misleads students into thinking that expressions in two languages
correspond one-to-one
– prevents students from thinking in the foreign language
– produces interference
– is a bad test of language skills
– is only appropriate for training translators

As observed by Malmkjaer (1998) the ‘doggedness’ with which the above

arguments have survived now up to the 21st century shows that ‘there must
122 Chapter 3

be some truth in them’ possibly stemming from a common misconception

about the nature of translation (see chapter 1). However, “if the types of ex-
ercises used in language teaching resemble the types used in translators’
training programmes reasonably closely” (Malmkjaer 1998: 6) then the
above arguments ‘fall away, one by one’. If translation is “grounded in a
sound understanding of the principles which should underlie all translation
activity” (Stibbard 1998: 69), it can be a useful pedagogical tool.
The tides about the place of translation in L2 teaching are certainly
changing and the more welcoming approach is supported by the continu-
ous efforts to understand how two languages coexist in one mind (see
chapter 2). With more and more evidence that L2 learners might need to
organize their two linguistic systems whether subconsciously or more
consciously translation as a pedagogical tool is being re-discovered.

Translation is intuitive and necessary and inevitable in the process of

learning another language and also in the process of understanding an-
other language through an intermediary (Pellatt 2002: 126). For the fol-
lowing reasons: Translation is a reality in everyday life, and becomes
more so as society is increasingly globalized. We translate constantly in
non-professional contexts such as instructions, timetables, ingredients,
letters (Pellatt 2009: 345).

As observed by Pellatt and many others (see Witte et al. 2009) the benefits
of letting L2 learners employ their bilingual knowledge (not only L2 com-
petence) give learners an opportunity to become aware of the more subtle
data extremely relevant for their communicative competence, which, how-
ever, cannot be taught in classes where only the L2 is used. These reasons
combined with the social expectations that bilingual language users should
be able to translate should not be ignored. Why should we not prepare our
L2 learners for the tasks which are ahead of them as participants in multi-
lingual communities in which intercultural competence is an essential part
of communicative competence (Cook 2010)? Indeed, recent interest in the
benefits of translation tasks for developing bilingual competence supports
the need to reinstate translation as the fifth skill of L2 learners. Some L2
teachers admit that translation might have been banned by theorists but in
fact it has never left the L2 classroom, and certainly it has never left the L2
learner’s mind (Zojer 2009: 31, Witte 2009: 82). Translation, as pointed out
by Malakoff and Hakuta (1991) with reference to natural translators (lan-
guage brokers), is indeed a powerful ally in amplifying pride in bilingual-
 Translation as an untrained ability … 123

ism, which in the context of foreign language learning can be used to en-
hance L2 performance as well as to integrate multilingual communities. As
pointed out by Malmkjaer (1998) translation as a task cannot be done
unless a fair amount of reading and writing in the learner’s L2 has already
been done and a satisfactory level of proficiency in the basic skills has been
achieved. Translation can be viewed as a skill which integrates all other
language skills (Whyatt 2008).
In March 2008 a conference entitled “Translation in Second Language
Teaching and Learning” was organized by the National University of Ire-
land. It gathered translators, translation trainers and language teachers
who discussed the various uses and benefits of translation tasks in the L2
classroom. As the selection of articles in the volume with the conference
proceedings entitled, Translation in Second Language Teaching and
Learning (Wittie et al. 2009) shows that all of those who spoke about the
useful applications of translation share the modern understanding of the
translation process which has its communicative, cognitive as well as so-
cial requirements. Yet, the L2 learners who were the subjects of the stud-
ies presented by the conference participants were not discouraged by
translation tasks despite, “the sheer complexity of the phenomenon” (Wal-
ters 2005: 209). On the contrary, there are a significant number of studies
which report on the positive attitudes L2 learners have towards translation
as their fifth skill (Sewell 2005, Whyatt 2008, 2009b). These positive atti-
tudes themselves speak for the use of translation in the L2 classroom (see
also Cook 2010, Leonardi 2010).
More and more voices can be heard about using translation tasks as an
educational tool. Colina (2002) appeals for ‘fostering closer interaction
among SLA, language teaching and translation studies’. In her words, “in
2nd language acquisition, translation, as a form of language use, is not a
language transfer exercise or a search and replace operation, but an activ-
ity performed with the objective of achieving a particular communicative
function across cultural and linguistic barriers”. Anderman (1998: 45) re-
ports that in 19 out of 21 British universities “translation was taught as a
way of improving students’ linguistic proficiency, that translation is used
to consolidate L2 constructions for active use and monitor and improve
comprehension of L2”. Sewell (2005: 153) in her article entitled, “Stu-
dents buzz round the translation class like bees round the honey pot –
why?” analyzed five groups of reasons why she believes translation is
perceived as the answer to some very deep-seated impulses felt by many
124 Chapter 3

language-learners. These reasons included: the need for confidence and

self-esteem, the need not to lose face, the need to be rewarded, the need
for certainty, for closure, for autonomy and the needs arising from any in-
troversion in our personalities (Sewell 2005: 153). In her view translation
as a task is “immensely satisfying. It is analogous to cooking – you end
up with a product you can call your own, the fruit of your individual, sus-
tained labour. Your ST [source text – BW] is the basket of ingredients
which you “treat” in various ways in order to produce your delicious dish.
The production of the dish constitutes the “reward” or satisfaction you get
out of the process” (Sewell 2005: 156). It should be added, however, that
contrary to cooking there is no set recipe to follow in order to produce a
good translation. This open-ended nature of the activity adds to the chal-
lenge, stimulates creativity and slowly builds up self-confidence as a re-
sult of repeated exposure to doubt. It then does not come as a surprise that
for EFL students the task of translation is regarded as a kind of a brain
teaser, an intellectually demanding exercise, challenging but rewarding
and therefore enjoyable (see Whyatt 2008). Needless, to say in the proc-
ess of translating a text there is a constant mixture of solid knowledge and
intuition summoned to solve open ended problems.
Following Jensen (1998) challenging tasks are usually highly motivat-
ing and completing them is considered highly rewarding. Embedded in
such a positive aura, it is to be expected that the memory traces left in the
form of newly acquired knowledge will be much stronger that in the case
of mechanically performed grammatical or lexical exercises. The intellec-
tual involvement needed to translate a text, combined with gradual prob-
lem solving and decision making intrinsic to the translation process (see
Piotrowska 2007) can indeed be compared to playing a game, or a run-
ning event with hurdles which have to be overcome. Completing this
challenge brings self-satisfaction.
Whyatt (2008) offers an insight into areas which can benefit from the
use of translation in the second language classroom and suggests that the
relationship between bilingual knowledge and interlingual skills is that of
mutual benefit. Anderman (1998: 46) points to the fact that through trans-
lation students’ text awareness is raised. Further support for the reintro-
duction of translation as a valuable skill which can bear fruit in terms of
enhancing and advancing L2 learners’ bilingual competence comes from
respected and renowned scholars in applied linguistics and SLA. Malmk-
jaer (1998) quotes Cook (1996) speaking in favour of translation as a task
 Translation as an untrained ability … 125

which can promote multilingual competence and is therefore a valuable

teaching/learning tool. Harden (2009: 120) suggests that translation is
useful in helping students acquire conceptual fluency and metaphorical
competence which is problematic even for advanced L2 learners with a
relatively high level of verbal fluency (see chapter 2 on conceptual profi-
ciency). In Harden’s estimation communication in the foreign language
alone, being “driven and restricted by certain need, (…) more often than
not leaves insufficient room for reflection of the interaction between the
L1 and the L2. This interaction in the L2 learner’s mind is a well estab-
lished part of the learning process”. Harden (2009) points to the stimulat-
ing ‘cross-fertilization’ between the two languages of the L2 learner and
concludes in the following words: “Translation exercises, used properly,
can provide the opportunity to reflect on the conceptual frameworks of
both languages involved as they are not focused on immediate communi-
cative needs. The written language is not ephemeral and thus lends itself
to trials, errors, and retrials more easily than the spoken word” (Harden
2009: 130). Whyatt (2007b) suggests that through translation L2 learners
re-discover all the simple facts about language use and communication
which they have taken for granted. Through their translation experience
including in-class discussion and post-translation analyses “students in a
guided way gradually come to view language in a holistic way with its in-
tricate network of connections reaching far beyond its linguistic means
into culture, reality, imagination. With awakening the students’ self-
awareness of the orchestrated effort one has to make to carry meaning
cross-linguistically, the task of translating itself becomes for them a more
and more rewarding activity”. This positive attitude is something lan-
guage teachers could capitalize on (see Sewell 1996). On the other hand,
the fact that through structured translation practice L2 learners become
aware of what is expected of professional translators is something not to
be thrown away by translation trainers (Mackenzie 2004). Christiane
Nord (1991: 165-166) says that translation practice in the language class
could develop skills which will later become a part of translation compe-
tence including an awareness of contrastive structures, or the effective use
of dictionaries (Pym 2003). Although as pointed out by Pym this view
was not popular in the early 1990s where more voices were heard sup-
porting dividing language learning from translator training.
126 Chapter 3

House (1986: 182) did bravely insist that acquiring communicative

competence was the aim of both the language class and the teaching of
translation. Despite the trend of the 80s and early 90s when many trans-
lation schools, at least in Europe, were struggling to mark out their ter-
ritory with respect to the established departments of Modern Lan-
guages. Mary Snell-Hornby, for example, was urging translation
schools in Eastern Europe to “cut the umbilical cord” with the Modern
Language departments (1994: 433) (Pym 2003).

The above quote shows that even translation scholars have started to no-
tice that L2 teaching methodologies and translation training methodolo-
gies share the aim of optimum attainment in bilingual knowledge. There
are still some voices that translation and L2 education should be kept
apart but in view of the increasing understanding of both proficiencies
they appear unsupported by evidence. To reconcile conflicting views on
translation in language teaching Klaudy (2003: 133) suggests that a dis-
tinction should be made between pedagogical and real translation which
differ in terms of: the function, the object and the addressee (cf. Gile
1995: 22 distinguishes between school translation and professional trans-
lation). Although the distinction is methodologically useful it is not essen-
tial when analyzing translation as a developing skill. Since the focus in
this work is on the evolution of translation as a human skill analyzing
how L2 learners perform translation tasks and how their performance
changes with more exposure to translation activities and following the
feedback they receive from the teacher or their peers, can inform us about
the early stages of the skill and complement the knowledge gained from
natural translation studies.

3.3.2. Studies in L2 translation

In contrast to natural translators L2 learners, at least initially, see trans-

lation as a mater of converting languages, a simple transformation of
words and structures. When translating they are preoccupied with
searching for equivalent expressions, mostly at the semantic level, they
calculate equivalences and semantic overlaps trying to stay as close to
the original as possible. Not surprisingly translation is hard work and the
lack of linguistic self-confidence results in the frequent use of bilingual
dictionaries. This attitude is probably justified by the methods via which
language used to be taught, and still continues to be taught by some
 Translation as an untrained ability … 127

teachers. As noted by Krings (1986a), Lörscher (1986, 1991, 1992),

Kussmaul (1995), Gerloff (1986), Whyatt (2000, 2007a) and others L2
learners with very low or low level of proficiency will translate by asso-
ciation of L2 forms with their L1 translation equivalents and will do so
without reference to the conceptual/meaning layer of their mental lexi-
cons. Lörscher (1991) referred to this tendency as form-oriented transla-
tion. De Groot (1997) used the term horizontal translation in contrast to
vertical translation which is sense-oriented. Presas (2000: 25) used the
term associative translator to refer to translations characteristic of ama-
teur translators who translate by associating L1 and L2 forms, and in
consequence frequently mistranslate rather than translate falling into
traps of false friends and taking seemingly similar words as sharing
identical meanings (‘actually’ as equivalent to the Polish aktualnie
meaning ‘currently’). As can be expected L2 learners as translators at
least initially represent the type of translator perceived as amateur, un-
skilled producing awkward incomprehensible translations (branded
translationese) full of L1 interference and generally regarded as a highly
undesirable third language (Duff 1981). The blame for this kind of flat
or horizontal transcoding was put on the manner of teaching/learning a
second language which heavily relied on the vocabulary learning of
word lists in both languages. As was discussed in chapter two this ten-
dency is explained by the developing structure of the learners’ bilingual
lexicon and their low conceptual proficiency and weak language control
mechanism, all being a result of L2 teaching methods which no matter
how hard they try cannot imitate the L1 acquisition process.

in standard non-natural L2 acquisition where a new language is taught

via classroom instruction there is a reversal of the L1 acquisition proc-
ess. When acquiring L1 the intention to communicate meaning that
Paradis (1998: 3) called the “micro genesis of an utterance” comes first
and well before the actual word labels for things are acquired. L2 teach-
ing often starts with the labels (foreign words) which are imposed on
the L1 labels and it is taken for granted that the learners themselves will
make all the necessary conceptual connections to encode not only the
form but also the meaning. Teaching the rules of grammar is often
given priority and a focus on form overrides concern for the content
(Whyatt 2009b: 182).

Having such language acquisition history, as well as having been repeat-

edly penalized for applying wrong word forms, tenses and violating the
128 Chapter 3

rules of English syntax during tests which focus on the formal aspects of
language, it is not surprising that L2 learners themselves attach more at-
tention to form than to meaning also when they translate.
Many studies into the nature of translation performed by L2 learners
confirm this focus on form. Lörscher (1986, 1991, 1992), Krings (1986b)
and Gerloff (1986), among others investigated the translation process us-
ing the method of thinking aloud in which L2 learners with sometimes
only an intermediate level of proficiency were asked to translate a text
orally at the same time verbalizing all their problems and dilemmas (see
Börch 1986 on the history of the method). Verbal data collected in the
form of Think Aloud Protocols (TAPs) showed that L2 learners defined as
‘inexperienced translators’ concentrate on linguistic forms without ana-
lyzing their meaning with respect to the context. Lörscher (1991) pointed
out that the form-oriented approach is favoured among L2 learners (see
chapter 2 on lexical links between L1 and L2) and only if it fails to pro-
duce any kind of equivalent TL form, the sense oriented approach is taken
(see Whyatt 2006b). To quote:

Substitutions of signs mainly occur in the lexical domain and result

from vocabulary equations which the subjects have learned in foreign
language lessons at school or at the university. Above all, decontextual-
ized and purely sign-oriented vocabulary learning, which even today is
rather widespread, forms and provides a large number of purely sur-
face-structure lexeme equations (...) which become available to the sub-
ject through an automatic association process (Lörscher 1992: 408).

Whyatt (2003) pointed out that the focus on form is present even at the
stage of reading an L2 text for translation (see Bajo and Macizo 2004).
During the first encounter with a text in L2 learners at a fairly advanced
level of language proficiency (2BA in English Studies) who were classi-
fied as beginners in terms of translation competence due to their limited
experience in translation (they completed one semester of a translation
course as a part of their practical English module) tended to underline un-
known words (33%) and use a dictionary (9%). In contrast, with growing
experience in translation and only a slightly higher level of language pro-
ficiency (3BA in English Studies), only 9% of the subjects classified as
intermediate in terms of translation competence (they completed 3 semes-
ters of a translation course as a part of their practical English module) un-
derlined unknown words and no subjects used a dictionary during the first
 Translation as an untrained ability … 129

reading. With advanced L2 learners (2MA in English Studies) and ad-

vanced level of translation experience (students just about to finish their
conference interpreting programme) no subjects underlined unknown
words or used a dictionary during the first reading before translating the
text but they admitted to thinking about translation problems during the
first reading (75%). In the group of beginners only 48% reported thinking
about translation problems when reading for translation but the percent-
age was higher (64%) in the intermediate group. This change of approach
to the text to be translated is probably due to the fact that practice in trans-
lation first of all makes students aware of the need to explore the context
for clues about the textual/functional meaning of an expression. It is also
possible that the developmental shift in approaching a translation task is
facilitated by the developmental shift in the way L2 learners access their
bilingual mental lexicons. As discussed in chapter two the focus on form
rather than meaning can be explained by the organization of L2 learners’
bilingual lexicon which at the initial stages of L2 appropriation is subor-
dinate with strong lexical links between L1 and L2 and weak conceptual
links between L2 forms and their mental representations. A translation
from L2 into L1 then relies on the process of association using direct lexi-
cal links often without checking for the conceptual/meaning content of
the lexical items. As can be expected such word-for-word translation can,
and frequently does lead to amusing translation.
Focus on form and lack of confidence in vocabulary use was con-
firmed by the study of dictionary use by L2 learners in the process of
translation. Varantola (1998), Atkins (1998) and Whyatt (2006a) pointed
to the general tendency among L2 learners performing translation tasks to
over rely on bilingual dictionaries in search of TL equivalents rather than
checking monolingual dictionaries for the meaning of L2 lexical items.
Whyatt (2000) reported that during an L2 to L1 written translation from
the 231 words that were checked in dictionaries, 156 cases were a search
for the desired TL (target language) forms with a possible exclusion of the
meaning analysis of the SL forms in their actual context.
To sum up, comparing natural translators to language learners with-
out experience in translation and without awareness of the cognitive
processes translation involves, it is clear that there is a problem with
conceptual proficiency. In a way then the form-focused approach to
translation shows that the complex nature of language as a socio-cultural
construct (see Gardner, 1979: 193) which makes its forms inseparable
130 Chapter 3

from their socio-culturally established meanings, connotations or innu-

endos may initially exceed an L2 learner’s cognitive abilities. However,
as pointed out by Pavlenko (2005: 446) conceptual transformation must
follow second language learning. So far, it has not received sufficient at-
tention from psycholinguistics and second language acquisition theories
alike and it has been neglected in L2 teaching and learning methodolo-
gies. It is possible that translation tasks which by nature involve both at-
tention to form and meaning could be used to speed up the acquisition of
conceptual proficiency. Research into L2 translation shows that the ini-
tial form-focused approach adopted by L2 learners gradually but fairly
quickly gives way to meaning-based translation with more exposure to
translation tasks combined with corrective feedback from their teachers
or peers (Whyatt 2007a, 2009a). It seems that similar to the suggestion
made by the mental lexicon research (discussed in chapter 2) there is a
developmental shift involving a change of attitude towards form –
meaning relationships observable in how the L2 learners’ untrained
translation ability is being transformed into a more skilled translation
performance. Some evidence can be found in TAP studies (Whyatt
2008) and deserves some attention as a potential turn in the process of
the evolution of translation as a human skill.

3.4. Developmental shift in translation as a bilingual ability of L2 learners

– TAP studies

Most, if not all, empirical data on the progression in the range of translation
abilities demonstrated by L2 learners comes from translation process stud-
ies which used the method of Think Aloud Protocols (TAPs). Thinking
aloud is a method taken from introspective psychology (see Börsch 1986
for details) which has been used to study the translation process by such
scholars as Krings (1986a, 1986b), Gerloff (1986), Kiraly (1995, 1997),
Lörscher (1991), Kussmaul (1995), Jääskeläinen and Tirkkonen-Condit
(1991), Whyatt (2000, 2010). It was the first empirical research method to
provide any insight into mental processing when translating. In the experi-
ments subjects were asked to translate a text and at the same time verbalize
all the problems and dilemmas they experience while performing the task.
These verbal reports, called Think Aloud Protocols (TAPs) were tape or
video-tape recorded and analyzed with an aim to reveal what goes on in the
translating mind (‘the black box’). However, as pointed out by Ericsson and
 Translation as an untrained ability … 131

Simon (1984) the subjects could only verbalize what they were consciously
processing and as a result anything that they were able to translate without
conscious problem solving and decision making became inaccessible
through TAPs. This limitation of the method became the reason for criti-
cism received by the researchers who studied TAPs (see Snell-Hornby
2006). Hönig (1991), for example suggested that the term Talk Aloud Pro-
tocols would be more appropriate than Think Aloud Protocols. In his arti-
cle, “Holmes’ ‘Mapping Theory’ and the landscape of mental translation
processes”, he pointed out that in the translation process conscious process-
ing is intertwined with subconscious processing, cognitive and intuitive
processes complement each other and contribute to finding solutions for
specific translation problems. This complexity of the process remains un-
questionable in TS literature. However, no matter how partial an insight
into the translation process was gained thanks to TAP studies it has contrib-
uted to an increased understanding of at least the conscious side of the
process (see Cronin 2005, Piotrowska 2007). Analyzing this, in a way ver-
balized inner speech of a translating person allowed to find some patterns
in the performance which leads from a SL text to its TL translation. It is
through a TAP analysis that translation became to be perceived as a prob-
lem solving, decision making activity (although many scholars had earlier
indicated these aspects, e.g., Levý 1967). It became clear that the process of
meaning transfer proceeds smoothly until a specific problem breaks up the
translation flow, is immediately identified as a problem and to use Hönig’s
term passed over to the ‘conscious workspace’ and it can then be verbalized
in TAPs. Obviously, like in any general action theory, the conscious mind
will make plans (or use strategies) to solve the problem using all the avail-
able cognitive resources.
TAPs showed that translators use a range of strategies and the choice
of strategies will depend on their translation proficiency. Gerloff (1986)
studied the TAPs of foreign language (L2) students with no previous ex-
perience in translating and reported on their overreliance on dictionaries
and their superficial meaning analysis which resulted in word-for-word
translation (note the parallel predictions of the RHM of mental lexicon
discussed in chapter two). The focus on form frequently sacrificed mean-
ing and led to odd incomprehensible translations. Although many scholars
who used L2 learners as subjects (Krings 1986b, Lörscher 1986, 1991)
were criticized for publishing results with general conclusions about the
translation process (see Snell-Hornby 2006), their research stimulated
132 Chapter 3

more interest in the translation process as well as it provided some insight

into translating done by inexperienced translators, L2 learners with a low
level of L2 proficiency.
The Think Aloud Protocol studies of inexperienced translators can be
used to infer information about the initial stages of development of trans-
lation skill. The form oriented horizontal translation (discussed in the pre-
vious section) was characteristic of L2 learners or users for whom:

a) translating is a novel experience

b) L2 proficiency is at a low level (beginners or intermediate)
c) Translating relies on bottom-up processing
d) Interlingual transfer operates on small translation units (words or
phrases)
e) Contextual information is not used for meaning interpretation

TAP research which mostly focused on the range of strategies applied to

creative problem solving in the translation process (Mackenzie 1998,
Kussmaul 1991) reported that with growing experience in translation
tasks L2 learners show signs of transformation from form-oriented to a
sense-oriented approach. Whyatt (2000, 2006b, 2007a, 2007b) studied the
TAPs of advanced EFL students after a yearly course of translation and
reported that there are signs of a transition from horizontal processing to-
wards vertical processing which:

a) is a feature of professional translators

b) involves effort aimed at a detailed analysis of meaning
c) makes use of top-town processing to verify bottom-up processing
d) interlingual transfer operates on larger textual/discourse units
e) contextual information is explored and used for meaning interpre-
tation

What is more, the analyzed verbal reports by students who participated in

a yearly course of translation show that in comparison to absolute begin-
ners they exhibit much higher metalinguistic awareness. A lot of verbali-
zations demonstrate that apart from problem spotting and problem solving
procedures an important metalinguiastic skill becomes detectable in the
translation process. Subjects start to self-monitor their solutions and fre-
quently reflect on language and the socio-cultural context of languages
they use in the process. The growing awareness of being involved in in-
 Translation as an untrained ability … 133

tercultural communication which disclosed gaps in intercultural compe-

tence seem to be the first turn in the development of translation ability
towards the stage of skilled performance. Then the realization that transla-
tion is a purposeful activity which must fulfill its communicative function
brings care for how the emerging TL text is structured and which words
are chosen out of many competing equivalents. As a result metalinguistic
awareness is developed and used to ensure a good quality translation
product. Let us devote some attention to these fundamental developments
in the evolution of translation as a human skill.

3.4.1. Intercultural competence

It is possible that neither the teachers nor the learners themselves are
aware of how little L2 learners know about the everyday practicalities of
the L2 reality for a simple reason that the need to know them has never
arisen (Whyatt 2008). Translating however makes these specific demands
even in a text as simple as an Old English Apple Pie recipe (see Whyatt
2008) which, according to the translation brief received from the teacher,
was requested by a neighbour who wanted to make the cake. The L2
learners when translating the text will inevitably start comparing the two
realities, Polish and English traditions, inherent expectations, available
ingredients, etc. In consequence they will classify the conceptual content
of the lexical items as the same or different to their own mental represen-
tations of reality lexicalized in their own native language. Being aware of
their reader’s expectations and the range of concepts which are within the
limits of his/her comprehensibility they will perform a truly intercultural
task while translating the text. In effect the information included in their
bilingual mental lexicon will be more precise in terms of conceptual con-
tent and the conceptual-lexical links will become clearer.
If one follows Levý (1967) in his view that a text is a picture of reality
it must be seen as standing in some kind of relation to the socio-cultural
heritage it is a part of. Therefore, in the understanding of the text, students
quickly learn that they have to go beyond the understanding of the lin-
guistic means into the understanding of some section of reality (Whyatt
2007b: 139-140). A similar perception of a text, or rather a text reception
process, which has been used in TS (Snell-Hornby 1988) is Fillmore’s
(1977: 61) frames-and-scenes metaphor where the reader in the process of
interpreting the SL text has to come to see the scene that the author of the
134 Chapter 3

text wanted him/her to see. As suggested earlier the way this scene will be
re-created in the mind of the reader/translator will depend on how the
frame is structured, and on all the backstage cognition the reader/
translator brings into the comprehension process. At this point however,
the translator is only half way through the translation process. Next the
scene has to be put in a TL frame so that it can, upon being read by the TL
receiver, be again unraveled into a meaningful scene, which depending on
the function of the translation, can be an authentic reflection of the scene
intended by the SL text author. Both metaphors assume a fluent coopera-
tion between the lexical and conceptual knowledge and there is no doubt
that this specific communicative transaction requires intercultural compe-
tence, defined here as an awareness of conceptual similarities and differ-
ences between the two languages and their respective realities.
In fact there is a twofold difficulty to be faced by an L2 learner/
translator in this respect and depending on the direction of the translation.
First, when translating from L2 into L1 he/she has to get to know the
relevant section of the L2 reality to ensure an adequate understanding of
the SL text for translation. As pointed out by Komissarov (1991: 43),
“understanding can be achieved only if the information contained in lan-
guage units is supplemented by background knowledge of facts referred
to in the message”. In L2 to L1 translation intercultural empathy defined
as the human ability to put oneself in the position of another person will
at least in theory be easier to achieve because of the shared cultural back-
ground with the TL text readers. However, in L1 to L2 translation, inter-
cultural empathy with the L2 addressees is much more difficult to achieve
even for experienced translators as it requires not only the knowledge of
the L2 culture but also the ability to see the translator’s L1 culture with
the eyes of the L2 reader eguipped with a different cultural heritage and
thus different expectations. In both situations intercultural competence is
essential to see the SL text and its TL translation as being a part of a lar-
ger extralinguistic context.
Gud, Kunst and Kim (1992) distinguish two types of contexts that
function in intercultural interaction (including translation): external and
internal. The external context includes the setting and location of the ac-
tual interaction and the meanings ascribed by the society to those two fac-
tors. The internal context, “is the culture that interactants bring to the en-
counters” (Byram and Feng 2005: 912). Byram and Feng point out that,
“in intercultural communication misunderstanding is much more likely to
 Translation as an untrained ability … 135

occur because the internal contexts that is the ways interactants use to per-
ceive the situations and each other and the meanings they associate with
the settings can differ greatly from one culture to another” (Byram and
Feng 2005: 912). Similarly a majority of translation errors committed by
inexperienced translators with a low level of conceptual proficiency do
not only result from poor linguistic skills but they are frequently an out-
come of low conceptual proficiency underlying their use of L2 (see chap-
ter 2) and low intercultural competence. Recent research into the complex
phenomenon of Cross Linguistic Influence (CLI, see Jarvis and Pavlenko
2008) provides many examples of language transfer. By practicing the
skill of translation and analyzing their translated texts in a translation
class, where the L2 learners ideally receive feedback sensitizing them to
conceptual transfer, they become aware of the similarities and differences
at the conceptual level of their mental lexicon. Some concepts are shared,
some are unique to only one of the languages and others overlap only to
some extent (see De Groot 1992, Pavlenko 2009). This intercultural
awareness is what L2 learners gain first of all from having to use both
their languages in translation tasks.
Fostering intercultural competence in L2 learners is a subject of grow-
ing concern among some SLA scholars. The educational challenge of the
Intercultural Approach is immense, especially taking into account the dy-
namics involved in the concept of culture. Witte et al. (2009: 6) quotes
Matsumoto (2000: 24) saying,

Culture is a dynamic system of rules, explicit and implicit, established

by groups in order to ensure their survival, involving attitudes, beliefs,
norms, and behaviours shared by a group but harboured differently by
each specific unit within the group, communicated across generations,
relatively stable but with the potential to change across time (Matsu-
moto 2000: 24).

There is a general awareness among those interested in the intercultural as-

pect of L2 teaching and learning that intercultural competence is very much
left out of L2 teaching methodologies. As pointed out by Byram and Feng
(2005: 925) both the intercultural competence and empirical research on its
acquisition is “still very limited and at any rate far more limited than that of
studies investigating second language acquisition” (Byram and Feng 2005:
925). This does not mean that the awareness of the importance of intercul-
tural competence is not present as confirmed by the recently increasing
136 Chapter 3

number of publications which address the interplay between language and

culture (see Hinkel 1999). There is also a common understanding that inter-
cultural competence is difficult to teach for many reasons. First, L2 teaching
in still very much linguistically oriented despite the fact that communicative
competence has been a key aim in foreign language teaching methodology
for decades. Culture in modern language departments is often taught as a
knowledge subject. Byram and Feng (2005) strongly object to teaching cul-
ture in the ‘fact-oriented approach’ where culture is dissected into smaller
sections which constitute separate topics for teaching. Culture in this fact-
oriented approach is viewed as the ‘big C’, that is culture as civilization, as
well as the ‘small c’, culture in everyday life. Kramsch (1993) cited by
Byram and Feng (2005: 917) views this approach as ‘inappropriate or even
damaging’ as it ignores the fact that culture involves ‘a social construct, a
product of self and other perceptions’ (Kramsch 1993: 205) and might ulti-
mately lead to teaching stereotypes and what is worse promote ethnocen-
trism. Instead, Kramsch (1993) suggests that language should be taught as
‘social practice’ where the focus is on relevant socially situated meanings.
Translation tasks seem to fit very well in this paradigm and the recent re-
vival of the interest in using them to promote intercultural competence in L2
learners seems well justified (Witte, et al. 2009, Cook 2010). By performing
translations L2 learners not only acquire and enhance bilingual knowledge
but they participate in intercultural communication.
To elaborate on the acquisition and participation issue, Pavlenko and
Lantolf (2000) use the distinction between two approaches to L2 learning
made by Sfard (1998), the acquisition and participation metaphors. In the
acquisition metaphor language learning is viewed as acquiring knowledge
like an object or commodity which involves learning and internalizing
rules and specific linguistic entities. In the participation metaphor learn-
ing is a process of becoming a member of a certain community (Sfard
1998: 6). Yet, mature L2 learners, e.g., students of foreign language de-
partments at university level who study a foreign language to be teachers,
translators or to use their expert knowledge in any other ways, are very
much aware that their aim is not to see themselves as members of another
culture. Their aim is to be able to meaningfully participate in the L2
community, but since they will always bring their own culture into this
participation, the participating itself is intercultural and consequently en-
riching for both sides of the communicative interaction.
 Translation as an untrained ability … 137

In contrast to other language tasks in which only L2 is used, transla-

tion by its nature includes the intercultural aspect of communication. L2
learners do start with a focus on forms when translating for the first time
or in the first stage of translation skill development but they very quickly,
especially when given appropriate corrective feedback on their transla-
tions, become to realize that in translation not only two languages but two
worlds meet. Negotiation of meaning like in any other kind of communi-
cation has to account for the background information which the partici-
pants, here the L2 learner, the text to be translated and its potential receiv-
ers bring with themselves (the inner context). Are the L2 learners pre-
pared for considering intercultural issues in their L2 communicative com-
petence? Or are they left on their own in a tacit expectation that they will
anyway in their own minds build some bridges between the two cultures,
comparing them and finding what they share and what is perceived as dif-
ferent, or foreign? Although there is more awareness of the need to pre-
pare the L2 learners for intercultural communication, there is lack of re-
search on the acquisition of intercultural competence (Sercu 2004: 84). It
is possible that translation as a part of the curricula for students of Mod-
ern Language Departments at the tertiary level is a valid source of inter-
cultural education. Indeed, research shows that L2 learners at university
level value and enjoy translation tasks and use them as a gateway to their
intercultural competence (Whyatt 2007b, 2008, Sewell 1993, Sewell
2004). Acquiring intercultural competence is possibly a lifelong process,
but getting it started in the L2 classroom which gathers future language
experts, is essential although difficult due to limited exposure to L2 socio-
culture. Translation tasks seem to provide a virtual reality substitute for
authentic cross-cultural communication.
However, many scholars remain skeptical about whether intercultural
competence can be achieved in institutionalized foreign language teach-
ing. Kordes (1990: 287-288), Byram and Grundy (2002), Swain and Lap-
kin (1998, 2000) report on the disappointing effects of foreign language
teaching with respect to the L2 learners’ intercultural awareness. Block
(2007) as cited by Witte (2009: 93) concluded that intercultural compe-
tence2 cannot be attained in foreign language learning institutions or by a
Year Abroad experience of being immersed in the foreign language com-
munity (Block 2007: 145-185). Some scholars suggest that only “pro-
–––––––––
2
which he sees as developing a ‘second language identity’.
138 Chapter 3

longed and intensive exposure to the foreign language society can facili-
tate a second language identity, or in our context, an interculturally com-
petent translator” (Witte 2009: 94). Although it might be true, it is not an
option to consider for millions of foreign language learners and aspiring
translators. Other means have to be found to foster intercultural compe-
tence (Block 2003). As suggested by Byram and Feng (2005: 925) culture
learning can be done as socialization by the teacher as a mediator into an-
other culture and should be combined with encouraging the students “to
reflect critically and analytically on their own culture”. Through compar-
ing, juxtaposing and analyzing they become in the process informed in-
tercultural mediators.
Indeed, there is another important dimension to learning about L2 real-
ity through practicing translation skill. Translation tasks as an intercultural
endeavour imply an important change in the L2 classroom. Room has to be
granted to the L2 learners’ both languages and cultures and this means de-
mocracy in the L2 classroom. “The fact that students are asked to compare
a foreign reality to their own helps them to feel that they can contribute to
an informed debate, instead of depending on the ‘know-it-all’ teacher” (De
Oliveira Harden 2009: 370). As pointed out by some researchers (Cronin
2005, González Davis 2004, Auerbach 1993, Witte 2009) with their native
language allowed back into their L2 learning process, the L2 learners are
given a voice and cannot be silenced anymore because of not being able to
express themselves fully in the foreign language. Translation does not fa-
vour one language over the other, but as observed by Auerbach (after De
Oliveira Harden 2009: 370) the use of L1 “reduces anxiety and enhances
the affective environment for learning, takes into account socio-cultural
factors, facilitates incorporation of learners life experience” (Auerbach
1993: 20 quoted after De Oliveira Harden 2009: 370). Consequently, in
translation classes the learners at last feel they are bringing something
meaningful to their learning experience – their L1 and their native concep-
tual competence (Kramsch 2009). Resting on this solid ground they are
keen to fill conceptual gaps underlying their L2 knowledge. In a translation
class conducted along the lines of collaborative learning (González Davis
2004) and social constructivist ideals (Kiraly 2000) not only two languages
but two realities and two worldviews meet. The conceptual level is scruti-
nized for differences and similarities and the intercultural translations are
frequently a topic of heated debates. Students translating a recipe for an Old
English Apple Pie into Polish, fiercely argue about the impossibility of
 Translation as an untrained ability … 139

translating ‘cooking apples’ into Polish referring to their native conceptual

competence according to which in Poland we do not formally distinguish
between cooking apples and eating apples (see Whyatt 2008). Other voices
are heard saying that people just know which apples to use for cooking and
which are for eating. There are also suggestions that it might be a good idea
to put the names of varieties of apples that are used for cake making in Po-
land. Then the name of the cake itself is problematic in translation, and all
the possible existing kinds of apple cakes available in Polish culinary tradi-
tion are considered as potential candidates. Again lacking unanimous
agreement students make their individual choices. Even the most quiet and
inhibited learners feel they have something to say in a translation class (see
González Davies 2004). To quote, since translation requires intercultural in-
teraction, “[t]he students regain their voices, as individuals and as a group.
They can not only express their opinions about the foreign text and about
the translated text they eventually produce, but they can also disagree with
the teacher, question his/her interpretative choices and even bring new in-
formation to the classroom” (De Oliveira Harden 2009: 370). This changes
the privileged position the foreign language teacher has been occupying for
decades in the classroom. The role of the teacher is that of a guide, negotia-
tor and advisor (Kiraly 1995) in the learners’ exchange of bilingual knowl-
edge, worldviews and experiences. Such a democratic system removes
anxiety and restores L2 learners’ self-esteem and, what is of utmost interest
in education motivates the learners to work for their very much needed in-
tercultural competence and take personal responsibility for decisions made
to solve intercultural problems in translation (De Oliveira Harden 2009).
This individual responsibility for the communicative quality of emerging
TL texts forces the L2 learners acting as novice translators to weigh up the
words they choose and reject others as not fitting the socio-cultural context
of the TL reality. In other words there is a growing metalinguistic aware-
ness in novice translators who meticulously scrutinize L1 and L2 words,
their forms and meanings. This brings the second major transformation in
the progression from translation as an untrained ability to translate to a
more informed and more skilled translation performance.

3.4.2. Metalinguistic awareness

Metalinguistic meaning about language can be in its simplest defined as a

more technical knowledge of language, its structure, constituting compo-
140 Chapter 3

nents and function. Bialystok (2005: 125) distinguishes metalinguistic

knowledge from linguistic knowledge as something related but qualita-
tively different from the application of linguistic knowledge in compre-
hension or production, as something added, a kind of embellishment,
privilege of the few, the more intelligent, the more multilingual. On the
other hand, she points out that any aspect of one’s linguistic knowledge
can become a part of one’s metalinguistic knowledge if it is focused on
and reflected upon. However, according to Paradis (2009) one’s metalin-
guistic knowledge cannot be converted into linguistic competence, but
remains available and can be used to compensate for gaps in L2 linguistic
competence as it is frequently the case especially with late bilinguals or
adult L2 learners.
(Bialystok 2005: 127) notes that bilinguals are at an advantage over
monolinguals in terms of their metalinguistic awareness. Metalinguistic
awareness can be defined as “an awareness or bringing into explicit con-
sciousness of linguistic form and structure in order to consider how they
relate and produce the underlying meaning of utterances” (Mora 2007).
Metalinguistic ability is defined as the “ability to make language forms
objective and explicit and to attend to them in and for themselves (…) the
ability to view and analyze language as a thing, language as a process,
and language as a system”. As can be expected and as it was suggested in
chapter 1, people differ in the range of their metalinguistic abilities which
can explain the idiosyncrasy in the use of language in general. Some peo-
ple, e.g., fundraisers can write an appeal letter which will make people
open their hearts and wallets, whereas others struggle to apply for the
benefits they are entitled to. While some learners are able to structure
their performance during an oral exam in the way which makes the exam-
iners disregard some gaps in their factual knowledge, others fail to show
off with the knowledge that they truly possess. Research into the transla-
tion performance of L2 learners shows that when two languages and two
cultures meet in one mind (Whyatt 2007b) metalinguistic reflection is a
natural consequence. Gile (2004) considers it a significant ally in educat-
ing reflective practitioners.
Malakoff and Hakuta (1991) observed that

The evaluation of the target-language sentence, both in terms of the

meaning it conveys and the sentence structure in which that meaning is
embedded, requires the ability to recognize language as a tool and as a
rule-governed system. The translator must evaluate his or her use of the
 Translation as an untrained ability … 141

tool (…) it is this necessity to reflect on language and language use

across two languages that makes translation a metalinguistic skill, par
excellence (Malakoff and Hakuta 1991: 150).

When translating novice translators have to focus on form to derive mean-

ing and look for appropriate verbal cladding in another language. In other
words translators are in and out of language/languages. They have to ob-
jectify, gain a distance, focus on minute aspects which are mostly irrele-
vant in every day communication, then they observe their solutions and
perform metalinguistic judgments, and finally they either accept or reject
their initial decisions.
A good source of empirical support comes from TAP research which
studied L2 learners performing translation tasks and which showed that
L2 learners provide a lot of verbalizations in comparison with experi-
enced translators who verbalize very little (Kussmaul and Tirkkonen-
Condit 1995: 190). Whyatt (2007a) analyzed verbal reports by students
who participated in a yearly course of translation and showed that in
comparison to absolute beginners they exhibited a much higher metalin-
guistic awareness. Subjects start to self-monitor their solutions and fre-
quently reflect on language and language use relevant to both linguistic
systems which they use in the process. All of the experiences of objectify-
ing, assessing, judging as appropriate or inappropriate make them more
sensitive to both their native and their foreign language as a tool in cross-
linguistic communication. The Self-monitoring of language production is
a valuable metalinguistic skill which not only helps to control interference
but generally allows to gain some control over the two working lan-
guages. Yet, very little attention has been paid to the role of metalinguistic
abilities3 in the development of translation as a professional skill.
It seems that in translators with some but limited experience translation
is essentially a metalinguistic activity which they try to consciously control
(Whyatt 2010). González Davies (2004) termed this stage conscious in-
competence and conscious competence. Through the practice of translation
they develop their contrastive capacity which intrinsically involves compar-
ing structures and words at the level of form and meaning. This interlingual
experience has many effects on the L2 learner’s view of language as a tool
in cross-cultural communication (Whyatt 2007b). In the briefest possible
–––––––––
3
Cummins (1991: 70) argues that one’s metalinguistic abilities are transferable and
refer to the learner’s both, or all languages just as one’s writing skills.
142 Chapter 3

way it is possible to say that through translation tasks learners experience

linguistic relativity in the neo-Whorfian sense (see Pavlenko 2005, Jarvis
and Pavlenko 2008, Whyatt 2008) when they look at words and structures
in a translation close up they become metalinguisticaly aware of the limits
of their own language and their foreign language. TAPs show the incredible
low self-confidence in making decisions when the time comes for choosing
one of the competing equivalents (to use the most persona non grata word
of TS). Dictionaries are notoriously used as an authority and it is not rare to
hear from L2 learners in reply to ‘why did you decide to choose this word’
a defensive reply, ‘because I found it in a dictionary’. Whyatt (2009a)
pointed out that communicative confidence in L2 learners can be a way be-
hind their communicative competence. Translation as a specific task places
specific demands on the choice of TL vocabulary appealing for accuracy,
contextual appropriateness, and stylistic choices matching SL and TL dis-
course patterns. The frequently favoured avoidance strategy, meaning that
learners do not use words they are uncertain of is out of question in transla-
tion tasks. Instead they have to operate within the limits marked out by the
SL text and the target language system. Their creativity is constrained and
doubts arise, ‘can I use this word in this particular context?’ Krings (1986a)
reported that his subjects not knowing which equivalent L2 word would
correspond to an L1 word tended to choose the first one given in a diction-
ary entry. This is how frequently repeated funny translations arise when, for
example the Polish word rosół [‘broth’] was translated as ‘culin’, because
the first word in the bilingual dictionary entry next to rosół was the abbre-
viated domain label ‘culinary’ (culin.)4.
This attitude shows not only low L2 proficiency but also a lack of
metalinguistic awareness referring to the multilayered relationships be-
tween form and meaning not to mention the contextual impact on mean-
ing interpretation. Anderman (1998: 39), for example, quotes Christo-
pher Hampton, British playwright and translator of Ibsen, who likened
the process of translation to a “gigantic crossword puzzle involving a
huge number of tiny decisions, from the choice of words to striking the
right stylistic note, making the right decision on a number of different
linguistic levels”. These choices are difficult for novice translators who
“produce and select from alternatives” (Pym 2003: 10). Many comments
in TAPs are “an expression of stepping outside one language system to
–––––––––
4
Agnieszka Chmiel (private communication).
 Translation as an untrained ability … 143

objectify languages, rules, structures and functions” (Mora, 2007).

Toury (1995) when rejecting Harris’s proposal of translation as an innate
skill said that translation skill is dependent on one’s interlingual capacity
that is the ability to establish similarities and differences across the two
languages. This ability to make contrastive cross-linguistic judgments
seems to fall within the realm of metalinguistic abilities. Learners, more
than professionals who reap the benefits of their experience, while pro-
ducing a translation constantly scrutinize their choices, assess the com-
municative quality of their TL version, engage in debates with them-
selves irrespective of the direction of their translation process. TAPs are
a good source of the creativity involved in generating and selecting be-
tween equivalents as well as proof of the learners growing awareness of
the increasingly necessary skills of text production (Koller 1992: 20,
Gile 2004). As suggested by Malakoff and Hakuta (1991) the fact that
through translation L2 learners start to reflect on the way they use lan-
guage in the process of conveying meaning across language barriers
suggests that they start to see language as a tool, as a means to an end
and not an end in itself. This understanding is inherent in translation as a
communicative activity. Later on in the development of translation to
reach the level of expertise, the metalinguistic care for the quality of the
TL text becomes integrated with intercultural competence and demon-
strated in the growing ability to self-revise and adjust the translation to
the needs and expectations of the TL text reader.
To summarize there is little doubt that structured exposure to transla-
tion experience (e.g., in a translation course for students of modern lan-
guage departments) leads to some transformations in their view of lan-
guage as a tool in cross-cultural communication (Vienne 1998, Whyatt
2007b). Research has shown that through practice the human ability to
translate evolves from the intuitive naïve rudimentary attempt to mediate
meaning from one language to another to a more conscious ability to ex-
tract sense/meaning expressed in a SL text and transport it into a TL text.
In view of the fact that metalinguistic knowledge has received little atten-
tion from translation scholars, it is possible that its role in the makings of
the translator has been taken for granted as many other issues such as SL
text comprehension or the balanced bilingualism of professional transla-
tors. Analyzing its role in the process of translation skill development
might throw light on some still unresolved issues, like for example why
natural bilinguals do not make good translators (see chapter 2) whereas
144 Chapter 3

translators who are non-native L2 users can produce a “better” translation

than natural bilinguals (Shreve and Koby 1997: xiv). If being exposed to
two language systems encourages metalinguistic awareness, then being
exposed to translation tasks will by analogy further encourage metalin-
guistic awareness.
Another viable question to ask is what effect the experience of trans-
lating might have on the developing bilingual knowledge. Do bilinguals
who become engaged in translation practice become interlingual indi-
viduals? In view of the ample data on the effects of bilingualism on cog-
nitive and linguistic abilities, the question of the effects of translation
practice on the cognitive and linguistic abilities of those who become ac-
tively involved is well justified. Some, although not many, interesting
suggestions have been made and they point to a possible re-organization
of the bilingual memory.

3.4.3. Re-organization of the bilingual memory

The interest in bilingual memory organization (see chapter 2) and lexical

access in the translation process has been very minor if not non-existent
among translation scholars which can be surprising in view of the fact
that in translation, and more urgently in interpreting we are very much in-
terested in fluent access and retrieval of words from memory. Although
the effects of translation experience on the organization of the bilingual
memory and patterns of lexical access have not been empirically investi-
gated some opinions have been voiced which allow to tentatively assume
that the bilingual memory may be reorganized to accommodate memory
traces of interlingual nature (Whyatt 2006c, Paradis 2009, Tymczyńska
2011).
Presas (2000) refers to Harris’s (1977) idea of the third competence of
a bilingual consisting in the ability to perform transfer operations between
both languages saying that,
To carry out these transfer operations, the translator must establish
bridges or linking mechanisms between his or her working languages.
In this specific aspect, therefore, translation competence would depend
on the bridges or linking mechanisms which he or she has established.
I would suggest that the acquisition of translation competence consists
precisely in this reorientation of bilingual competence towards interlin-
gual competence (Presas 2000: 27).
 Translation as an untrained ability … 145

Noticing that studies on novice translators revealed that the bridging

mechanisms are initially very “rudimentary: frequently consisting of
automatic 1:1 associations, almost always restricted to the lexical level”
(ibid.) Presas suggests that the development of translation competence
(TC) consists in improving these bridging mechanisms.
Yet, as put by Snell-Hornby (1992), the L2 learners who enter transla-
tion training programmes have to be weaned away from their associative
thinking in terms of equivalent words (horizontal translation) and taken
towards a more holistic view of the text and consequently of their two
languages. Gile (2004) showed that “it is possible to teach students very
rapidly to unlearn word-for-word, thoughtless translation and turn to
analysis and reformulation, in spite of the many years spent with the lan-
guage-equivalence paradigm”. “This is why the translation scholar has to
be a semanticist over and above everything else. But by semanticist we
mean a semanticist of the text, not just of words, structures and sentences.
The key concept for the semantics of translation is textual meaning”
(quoted after Bell 1991: 79). Seleskovitch (1976) insisted that in the
translation process the meaning of the SL text unit has to be ‘deverbal-
ized’, i.e., processed at the level of conceptualization and then brought
down again through the process of lexical access with formulation and ar-
ticulation in the target language form. This vertical route is viewed as
characteristic of mature professional translators. In view of the complex-
ity of bilingual knowledge discussed in chapter 2, cross linguistic influ-
ence and the language imbalance in most L2 learners and translators, this
transformation is not a matter of a switch but a long-term process which
combines intercultural competence and metalinguistic knowledge.
However, Paradis (1994, 2009) seems to have a different opinion and
while referring to the hierarchical nature of the bilingual lexicon (see
chapter 2) suggests that the horizontal processing does not have to be re-
served for inexperienced translators. To quote,

Speakers have two translation strategies at their disposal (…). In Strat-

egy 1, the naïve strategy, probably adopted by occasional interpreters
who are used to speaking one language to one group of speakers and the
other to another group, translation is accomplished via the conceptual
system, according to the normal process of implicit linguistic decoding
(comprehension) of the source language material followed by encoding
(production) of the target language material (…) (Paradis 2009: 180).
146 Chapter 3

According to Paradis this translation route is automatic in the sense that it

is not consciously controlled but requires the simultaneous activation of
both languages, but the activation is not to the same extent. Referring to
De Groot and Christoffel’s (2006) article on bilingual control in transla-
tion and simultaneous interpreting Paradis predicts that under the condi-
tion of differential activation of the two languages,

translation from stronger into the weaker language may yield the best
outcome since: Comprehension is guaranteed and the target text may be
adapted to the translator’s proficiency by using circumlocutions and
other devices, when necessary. The meaning (of the source language,
correctly understood, may be accurately rendered in the weaker lan-
guage, in spite of its possibly somewhat compromised morphosyntactic
form (Paradis 2009: 181).

The last comment indicates that Paradis does not take into account the
developmental nature of translation as a human skill which has been
documented to exploit lexical links prior to conceptual connections estab-
lished much later with increasing L2 proficiency. What is more, the trans-
lation direction however claimed by Paradis (1994, 2009) to be able to
yield the best outcome is not in line with the generally accepted transla-
tion policy (binding in the EU translation services) according to which to
ensure high quality, the translation should always proceed from the trans-
lator’s weaker language (language B) into his/her stronger language (lan-
guage A). This translation axiom, although still binding in theory, has re-
cently been questioned by Pokorn (2005) who demonstrated that success-
ful translation is possible into the translator’s non-native language.
Strategy 2 called by Paradis conscious is bypassing the conceptual
level and relies on direct transcoding based on “code-to-code links be-
tween the source language and the target language involving at least the
lexical and syntactic level of processing” (Paradis 2009: 180 cf. Ruiz et
al. 2008). Gile (2009: 240) refers to such paired expressions in L1 and L2
using the term ‘Translinguitic Equivalences’ and quotes De Groot and
Christoffels (2006: 198) who stress that memory traces are formed and
strengthen every time a given pair of translation equivalents is used. As
explained by Paradis, professional simultaneous interpreters, rely on ex-
tensive metalinguistic knowledge in the form of the learned translation
equivalents which they gain through their translation experience. To
quote, “In this scenario translation is carried out directly via (learned) as-
 Translation as an untrained ability … 147

sociation links between the lexicons (or any level of linguistic structure,
including syntactic constructions) without going through the encod-
ing/decoding route” (Paradis 2009: 180). Contrary to Strategy 1, Strategy
2 relies on the declarative memory which draws on the metalinguistic
knowledge of learned associations between L1 and L2 forms. This obser-
vation may seem far-fetched and creates the illusion that in the develop-
ment of translation as a skill we in fact first try to discourage the form-
oriented translation in inexperienced translators and then come to the
point where Paradis (2009) calls the conceptually mediated translation
‘naïve’ and needlessly time consuming. On the other hand, it has to be
remembered that Paradis (2009) draws his views on the grounds of simul-
taneous interpreting and a reservation has to be made that his conclusions
do not have to apply to written translation. Research into simultaneous in-
terpreters has shown that to cope with the time pressure and mental effort
management in this very specialized type of translation which can be flu-
ently performed only by especially trained interpreters and not by acci-
dental naïve (to use the adjective chosen by Paradis) translators accuracy
might be sacrificed for fluency (Shlesinger (2000) and the ‘get on with
the show’ slogan). Yet, the extent to which this is indeed the case does not
have to compromise the overall conceptual content of their translation. It
might be possible that the information drawn from the extralinguistic con-
text and the pragmatic situation shared by all participants in which simul-
taneous interpreting is done supports the communicative function even if
accuracy is sometimes secondary to fluency. As in any oral language use
the participants of verbal exchanges are much more lenient in tolerating
inaccurate and poorly phrased content (see Kopczyński 1994). This atti-
tude, however is entirely unacceptable in written translation which cannot
rely on the transience of the translated message and lenient attitude of
those who will read the translation.
To safeguard adequate interpretation of the SL text author’s intended
meaning accuracy at the conceptual level is a primary requirement for a
written translator. What is more, contrary to simultaneous interpreting
unless the actual translation performance is recorded for analysis written
translation always results in a permanent record in the form of translated
texts which can undergo critical evaluation at some point. It is therefore
possible that Paradis (2009) by distinguishing the two strategies by which
the mental lexicon can be accessed may imply that interpreters and trans-
lators re-organize their mental lexicons in a different way and access them
148 Chapter 3

by different routes depending on the actual translation task. This possibil-

ity was also reported by Presas (2000: 25) who pointed out that “We can-
not rule out the possibility that more than one of these four types [subor-
dinate, associative, coordinate and compound organizations – BW] could
be operative in any given translator, nor that one type or another could be
dominant at different stages of a translator’s training process”.
As this chapter has shown natural translators process information at the
conceptual level sacrificing the structural aspects, L2 learners exploit lexical
links at the expense of conceptual content. A developmental shift has to oc-
cur so that in professional translation where the skill is trained neither con-
ceptual content nor formal structural aspects are compromised. This kind of
balance requires efficient cognitive control. It is likely that the progression
from translation as an untrained ability to translation as a trained skill results
in the functional reorganization of the translator’s bilingual memory. This
assumption however, as so far lacks empirical support. Analyzing the trans-
formations within the bilingual knowledge of a developing translator has not
attracted as much interest as the search for added capacities in the process of
translation skill development. Nevertheless, it is possible that if more atten-
tion is paid to the transformation within the bilingual knowledge structures
as a result of translation experience many problems faced by translation
trainers and trainees will be easier to understand if not overcome. Finally,
this attempt to see the human ability to translate in its developmental conti-
nuity will hopefully encourage more cooperation between language learning
and translator training pedagogies and practitioners.

3.5. Conclusions

The purpose of this chapter was to demonstrate that translation skill is built
on bilingual foundations which allow bilinguals to translate without any
special training. The so called natural translation is characterized by focus
on content and is used as a communicative strategy which allows for inter-
action in multilingual communities. However, as observed by Malakoff and
Hakuta (1991) natural translators (bilinguals) cannot be expected to pro-
duce a flawless translation of written texts. Natural translators who assist in
intercultural communication frequently perform oral translation when for-
mal correctness in not expected and there is a high tolerance of language in-
terference. Functionally they are excellent communicators and there is no
doubt that bilinguals are able to translate when by translation we under-
 Translation as an untrained ability … 149

stand being able to communicate information expressed in one language in

the other language of the bilingual. Written translation is more problematic
and there is some evidence that the innateness hypothesis (Harris and
Sherwood 1978: 155) is not enough to guarantee efficient flawless transla-
tion. Beatens Beardsmore (1982) reported cases of perfectly balanced bi-
linguals (see chapter 2) who while able to “function extremely well in two
languages in clearly demarcated situational contexts often find it difficult to
spontaneously translate between their languages without heavy interfer-
ence” (Beatens Beardsmore 1982: 88).
L2 learners acting as translators exhibit another manifestation of the
natural untrained ability to translate. Research focusing on the effects of
translation practice, where translation is used in the modern sense of in-
tercultural communication, shows that translation tasks have a lot to offer
to L2 learners. With the changing views on language as a social construct,
with the increasing number of multilingual learners who bring their own
culture and their own language to their foreign language learning experi-
ence, and with recent advances in understanding bilingual knowledge
which clearly point out that it is impossible to silence the L2 learner’s na-
tive language, “the time is certainly ripe for a re-evaluation of the benefits
translation can bring to the process of learning a second language and its
cultural context” (Witte et al. 2009: 4). Tracing the benefits L2 learners
can draw from authentic translation tasks will consequently point to the
major stages in the evolution of translation as a human skill, an integra-
tive skill which can be used to enhance L2 attainment in the context of
teaching and learning a foreign language. Discussing the most significant
developments the experience of translation can stimulate in L2 learners
and comparing L2 learners as translators to bilinguals acting as natural
translators has prepared the grounds to suggest that if the translation skill
is further developed, e.g., in a translation training programme or through
other exposure and involvement in translation tasks there must be a quali-
tative shift in the development of translation as a human skill.
In this respect the authors who have investigated translation as an un-
trained ability of bilinguals suggest that the evolution of translation skill re-
quires something more than bilingual knowledge and some other capacities
have to be added. The discussion of translation as an untrained ability of bi-
linguals presented in this chapter points to some directions where these
added capacities may reside. Natural translators could become profession-
als if they increased their metalinguistic abilities and became more meticu-
150 Chapter 3

lous about the form. L2 learners to start with take the translation process
the other way round and focus on form at the expense of meaning. Later
through increased understanding of the nature of translation they gradually
develop a more holistic sense-oriented approach to translation.
Malakoff and Hakuta (1991) suggest that translation proficiency can
be viewed as “the product of an interplay between meatlinguistic maturity
and bilingual proficiency” (Malakoff and Hakuta 1991: 149) while trans-
lation strategies are the third factor which has the power to “enhance per-
formance, but not beyond an optimal-level limit determined by the two
linguistic factors” (see Fisher and Pipp 1984). Strategies however, arise
only with repeated experience of translation followed by feedback which
aims at better translation performance. It is when L2 learners follow the
pattern of action which includes not only planning the translation task,
executing it but also rehearsing the results that their ability to translate
becomes transformed into a skill. Ideally, it should take place “under op-
timal conditions of support, experience, and feedback” (Malakoff and
Hakuta 1991: 149). It is possible that the repeated experience of transla-
tion leads to a reorganization of the bilingual memory of the translator
and that the bilingual person who chooses to pursue a career in translation
is not only a bilingual but also an interlingual individual with bridges
built between L1 and L2 which allow for efficient cross-linguistic lexical
access. The scant research into natural translation provided a valuable in-
sight into translation as a natural ability in bilinguals although as admitted
by Malakoff and Hakuta (1991) it only “begun to scratch the surface of
empirical relationships between the different abilities involved in transla-
tion”. Many questions arise. What are the cognitive consequences of be-
ing involved in a repeated experience of translation as an untrained ability
for natural translators (language brokers) and for L2 learners? What are
the additional capacities which develop in the course of translation skill
development and how do they interact with the bilingual knowledge
structures? Which aspects critically affect translation proficiency both in
terms of speed and the communicative quality of translated texts?
To sum up, even the most competent and the most wanted translator of
today was once an L2 learner (or a natural bilingual) and it is likely that
the most successful translator in fact never ceases to see himself/herself as
a language learner (Robinson 1997). However, this is not a popular view
among translation scholars who prefer to view translation as a special
skill reserved for professionals (Newson in Malmkjaer 1998). In this re-
 Translation as an untrained ability … 151

spect Translation Studies as a discipline are not without blame for the lack
of interest at the early stages of development of translation as a human
skill. With the first call for papers from the University of Bologna, the or-
ganizers of The First International Conference on Non-professional Inter-
preting and Translation scheduled for 2012 the tide may be changing. As
expressed by the organizers,

the aim of the conference is to provide a forum for discussion in a rela-

tively recent and often neglected field of language and cultural media-
tion. Despite being a hugely spread and submerged practice, non-
professional interpreting and translation has always been the poor rela-
tive of both interpreting and translation studies and, as such, neglected
and under-researched by academia and condemned by professional
categories5

Perhaps finally hopes might be raised that TS will become an all-inclusive

discipline to study all forms and facets of translation, not exclusive to pro-
fessional translation and ready to investigate the continuity of translation
expertise development. It is possible that it is at this non-professional
stage that an important shift in the skill occurs which includes a transition
from the lexical to textual level in the learner’s understanding of meaning.
This is also a stage when major changes occur in the learner’s view of
language as a tool in intercultural communication which makes it neces-
sary to include the cultural information encoded in languages. This pro-
gression on the developmental continuum of translation as a skill is be-
lieved to be a part and parcel of every translator. Depending on the indi-
vidual language acquisition history, L2 learners who choose to pursue a
translation career can be already equipped with some additional capacities
including intercultutal awareness and metalinguistic abilities which will
strengthen the bilingual foundations and prepare the ground for their fu-
ture translation competence and translation expertise. Others, who choose
to pursue a career in translation following their intuition rather than past
educational experience of translation will have to acquire these abilities in
a translator training programme or by other means (e.g., self-coaching,
mentoring, internship in translation agencies, etc.). Those who will not in-
tentionally and consciously seek to further their translation ability past
their natural untrained capacity will remain natural translators.
–––––––––
5
(http://npit1.sitlec.unibo.it/).
152 Chapter 3

Can one then say after Klaudy’s (2003: 133) that “translator training
starts where foreign language teaching ends”? In view of the analysis of
natural translation abilities presented in this chapter as well as with re-
spect to the current views on bilingual knowledge, there seems to be no
clear cut boundary between translation proficiency and language profi-
ciency. Quite on the contrary, both proficiencies seem to complement
each other and feed on each other (Whyatt 2008). Consequently the rela-
tionship holding between them is that of mutual benefit. Chapter 4 will
look at translation as a trained skill of those who want to provide transla-
tion services as professional translators and therefore intentionally and
consciously seek to acquire translation competence.
Chapter 4

It has been by translating professionally that I myself

have learned (and continue to learn) how to translate
Kiraly (2000: 8)

Developing translation competence

The purpose of this chapter is to look at translation as a trained skill in

those bilingual language users who aspire to take up translation as their
professional career. It is assumed that the development of translation as
a trained skill is a conscious choice of career which like in any other
profession requires professional competence with all the necessary skills
and knowledge needed to provide professional translation services.
Questions which arise especially from the perspective of the L2 learner
and an aspiring translator include: What do I need to add to my bilingual
knowledge and the intercultural and metalinguistic awareness that I have
already started to acquire to become a professional translator? It turns
out however that, “[t]he literature on Translation Studies lacks a consis-
tent description of the abilities and skills required from a professional
translator” (Alves et al 2001: 46-47, see also Shreve 1997: 121, Orozco
2000: 199). In the sections below an attempt is made to review different
approaches to translation competence (TC) in order to gather some
guidelines on how to best structure the learning environment in which
the natural ability to translate can evolve into a sophisticated repertoire
of translation as a professional expert skill. Finally, some attention is
paid to translation as a profession with its requirements and market de-
mands in order to elicit the background for the course of evolution of the
translator’s professional self.

4.1. Translation competence

Despite Bell’s (1991) appeal that translation theory should account for
the process which is required to produce a translation as well as the
abilities and skills necessary to perform the task, scholars are divided
154 Chapter 4

on their view of translation competence1 (TC). As expressed by the

PACTE (Process in the Acquisition of Translation Competence and
Evaluation) research group, “[u]nlike other disciplines in which numer-
ous studies have been carried out to determine what constitutes expert
knowledge in the field and how this knowledge is acquired, no gener-
ally accepted model of translation competence exists in the field of
Translation Studies” (PACTE 2003: 46). Pym (2003) provides a review
of different attempts to define translation competence with a conclusion
that the changing perspective on how to define the term runs parallel to
the evolution of Translation Studies as a scholarly discipline. The atti-
tudes displayed by various scholars can be divided into three perspec-
tives: the additive perspective, the componential perspective and the
holistic perspective. Let us follow Pym’s review from these three stands
looking at each of them with the eyes of a novice translator. A novice
according to Hoffman (1997: 200) is no longer a ‘naïve’ (i.e., somebody
completely ignorant of the domain), but an apprentice (meaning the one
who is learning preferably under the eye of a master or a trainer) self-
motivated to acquire all the necessary skills and knowledge.

4.1.1. The additive perspective on TC

The point that the human skill to translate must include something more
than linguistic competence in the two languages was made in chapter
three. Toury (1995: 246) considered it essential that bilingual knowledge
has to be supplemented by the ability to establish similarities and differ-
ences across languages which he branded interlingualism and which
probably is a part of one’s metalinguistic and metacognitive abilities
rather than a function of their linguistic competence. These “added ca-
pacities are inherently different in different people, part of different
mental structures” (Toury 1995: 248) and they are probably influenced
by one’s language acquisition history. Toury (1995) maintained that they
are “trainable at least up to a point” through practice in translating in
context combined with ‘environmental feedback’, that is translation
product analysis and assessment in the educational translator training
–––––––––
1
Shreve and Angelone (2010: 4) note that the notion of translation competence is
in fact an adaptation from Chomskyan linguistics and the many attempts to define it
did not refer to cognitive concepts of long term memory, schemas or declarative
knowledge.
 Developing translation competence 155

context. This added capacity was branded as transfer competence and

suggested that a kind of bridge, an easy passage between the two lan-
guages has to be built in the mind of the translator. Presas (2000) sug-
gests a more evolutionary perspective which takes into account the ex-
isting bilingual knowledge and says that the acquisition of translation
competence consists in “reorientation of bilingual competence towards
interlingual competence” (Presas 2000: 27). A bilingual speaker to carry
out translations viewed as ‘transfer operations’ has to establish “bridges
or linking mechanisms” between both working languages. However,
since the translation process research showed that the transfer mecha-
nisms of novice translators are highly rudimentary (see chapter 3) rely-
ing on automatic lexical associations between L1 and L2, Presas sees the
development of TC consisting in “the improvement of these bridging
mechanisms”. Although the idea as such is very interesting, it remains a
theoretical notion and awaits empirical support. The most important
question is the level at which these interlingual bridges are being cre-
ated. Are they built at the conceptual level or at the lexical level and
how do they respond to the dynamic nature of meaning. Is there any
empirical evidence that as a consequence of building these linkages be-
tween the two working languages there is a qualitative and quantitative
difference in lexical access as compared to a bilingual who does not
seek to master the skill of translation? If this is the case then, a third
language mode (see Grosjean 2001), the interlingual mode in language
use would have to be recognized. This would imply that translation is
not an extreme form of bilingualism (Paradis 2005: 412) but that we are
in fact talking about the emergence of an interlingual individual (Whyatt
2009c) and interlingualism in Toury’s (1995) sense is the added capac-
ity, a result of transformation within the bilingual foundations of transla-
tion competence.
Another doubt arises, however, stemming from the fact that every
translation task is in a way a novel experience, as it involves a new SL
text, new audience, perhaps new text type, and a translator who, al-
though is physically the same person, is different because of being en-
riched by his/her most recent translation experience. In this situation the
bridges and the linking mechanisms become re-negotiated and therefore
cannot be viewed as static linking mechanisms but have to be re-set with
every new translation task. This is an issue that I will come back to in
156 Chapter 4

more detail in chapter five. Nevertheless, the term transfer2 competence

as added capacity to enable translation has had many followers who saw
it as a feature which is distinctive from bilingual competence but which
involves the ability to code-switch and the metaphor of building bridges
while changing codes seemed appealing. Still in the late 1980s transla-
tion competence was defined as “the ability to put together [verbinden]
the linguistic competences gained in two languages” (Koller 1979: 40),
or “as a summation of tongues” (Pym 2003). A view which was criti-
cized (Pym 2003) and its rejection was confirmed by TAP research
pointing to the fact that even bilingual knowledge at a high level of L2
proficiency does not guarantee translation competence.
Wilss (1976) for example suggested that translation competence
should be “clearly marked off the four monolingual skills: listening,
speaking, reading and writing” and viewed as ‘supercompetence’ (Wilss
1976: 120). Disregarding what would now be considered as a non- PC
term, many TS scholars started to add other sub-competences to the ‘su-
percompetence’ of translation. In effect it was assumed that translation
competence is more than a summation of language knowledge.
Koller (1992) claimed that TC is “not simply more language compe-
tence […] but also the creativity involved in finding and selecting between
equivalents and in the increasingly necessary [skill of] text production”
(Koller 1992: 20 quoted after Pym 2003). Snell-Hornby (1992) pointed to
the need to develop a holistic approach to texts and incorporate cultural
contexts which novice translators find very problematic. Skopostheorie
(Reiss and Vermeer 1984) stressed the awareness of function and text type
as an important part of translation competence. In effect the list of added
capacities which should be included in TC seemed to endlessly grow.
Bell (1991) offers a relevant illustration in the form of a long list of
knowledge components where TC is viewed as the sum of the following
items: “Target-language knowledge, text-type knowledge, source-language
knowledge, subject area (‘real world’) knowledge, contrastive knowledge,
then decoding and encoding skills summarized as “communicative compe-
tence” (covering grammar, sociolinguistics and discourse)” (for more pro-
posals of what TC should include see Pym 2003). One can hardly imagine

–––––––––
2
Note the disharmony between the use of the word transfer in the term transfer
competence in TS and the use of the word transfer (i.e. negative transfer) in SLA theo-
ries and in Cross Linguistic research.
 Developing translation competence 157

something more intimidating to a novice translator than being provided with

such an extensive list of requirements (see Korzeniowska and Kuhiwczak
1994). This is not everything, though and TC should also include, apart from
the sum of all possible kinds of declarative and procedural knowledge, the
“knowledge of translation processes” as pointed out by Wilss (1988).
Some scholars, especially those involved in TAP research claimed that
the term ‘TC’ should be abandoned as non-definable (Lörscher 1991).
Schäffner and Adab (2000) in their introduction to Developing Transla-
tion Competence explicitly surrendered saying that, “competence” in-
volves any number of other terms and thus can be used as “a cover term
and summative concept of the overall performance ability which seems so
difficult to define” (Schäffner and Adab 2000: x). Pym (2003) responds to
this suggestion saying that it involves a paradoxical reduction of TC to
performance ability (compare Chomsky’s (1965) competence-
performance divide). In terms of pedagogical implications such a defini-
tion is far more deficient than the componential view of TC which runs in
parallel to the additive perspective and which most probably was an at-
tempt to organize all the suggested components into what should ideally
be included under the heading TC that makes it unique, different from
language competence and makes its acquisition possible.

4.1.2. The Componential perspective on TC

According to Pym (2003) the perception of TC as a multi-componential

cluster concept was a reaction to the break of TS from linguistics and its
orientation as an interdisciplinary field of research. One could also sug-
gest that it is a result of the increasing awareness of translation as a proc-
ess and translation as a complex skill. Researchers working in the PACTE
group noted that the additive perspective also shows the bias of individual
researchers (House 1986, Bell 1991, Nord 1991, Pym 1993, Kiraly 1997,
Hatim and Mason 1997, Risku 1998, Presas 2000) and as such is “limited
in scope as they deal only with limited aspects of translation compe-
tence”. What is more and far worse though, is that most of the additive
definitions of TC (as something more than language competence) are pre-
scriptive rather than descriptive, based on what is expected of a translator
as a professional rather than on the basis of structured research projects
with clear methodology and empirical rigour. In this respect all the defini-
tions of translation competence and models lack empirical support and
158 Chapter 4

ecological validity not to mention statistical significance. As the PACTE

group observes, “As far as we know, only two studies have attempted an
empirical approach to research into translation competence as a whole”:
Lowe (1987) and Stansfield, Scott and Kenyon (1992) and even these as
pointed out by Orozco (2000: 113) cannot be generalized because of the
small number of subjects they studied. As a consequence, listing the com-
ponents of TC is nothing more that chasing the unknown ideal which
given the structure of our mental lexicon can look quite different in dif-
ferent people. What is needed is empirical research to gather ecologically
valid data in order to describe the components of TC and to establish the
connections and hierarchies between them. This is precisely what the
PACTE group aims at (PACTE 2003: 47).
TC is viewed as expert knowledge needed to solve problems and
make decisions to achieve the communicative aim of translation which is
perceived as a communicative activity. The holistic model of TC decom-
posed into several sub-competences has been developed and re-defined
since 1998 by the PACTE group. The researchers draw insights from sev-
eral sources which include: research in other disciplines such as psychol-
ogy, pedagogy, language teaching which define notions such as “expert
knowledge”, “competence” and “learning processes” as well as the mod-
els used to define TC in TS (see Orozco and Hurtado Albir 2002).The ba-
sic premises of the 1998 model include the following points:

1. Translation competence is the underlying system of knowledge needed

to translate;
2. Translation competence is qualitatively different from bilingual com-
petence;
3. Translation competence is an expert knowledge and, like all expert
knowledge, comprises declarative and procedural knowledge; the lat-
ter is predominant;
4. Translation competence is made up of a system of sub-competencies;
5. The sub-competencies of translation competence are considered to be:
a language sub-competence in two languages; an extra-linguistic sub-
competence; an instrumental or professional sub-competence; a psy-
cho-physiological sub-competence; a transfer sub-competence, and a
strategic sub-competence;
6. All the sub-competences are inter-related, hierarchical and their rela-
tionships are subject to variations (PACTE 2003: 47-48).
 Developing translation competence 159

Although the division into the 6 sub-competences seems plausible their

description offered by the authors suggests that it is difficult to single
out all the pockets of knowledge included in each of them. For example,
language sub-competence was defined as “the underlying system of
knowledge and abilities necessary for linguistic communication in both
languages”. There is no mention however of metalinguistic abilities
unless they are simply taken for granted. The extra-linguistic sub-
competence includes “implicit or explicit knowledge about the world in
general and specific areas of knowledge” including: knowledge about
translation (its ruling premises: types of translation unit, the processes
required, etc.), bicultural knowledge, encyclopedic knowledge and sub-
ject knowledge. The one question which immediately arises is why
‘knowledge about translation’ is not included in the third instrumen-
tal/professional component which includes: knowledge and abilities as-
sociated with the practice of professional translation such as the skill
and knowledge of documentation, use of tools and modern technologies,
knowledge of the professional market and the profession (prices, types
of briefs, etc). The transfer sub-competence is, to quote, “the central
competence that integrates all the others. It was defined as the ability to
complete the transfer process from the source text to the target text, that
is to understand the source text and re-express it in the target language,
taking into account the purpose of the translation and the characteristics
of the receptor” (PACTE 2003: 48). Although the transfer competence
(which is nothing new in TS) is considered central, it is the strategic
sub-competence which “plays an essential role in relation to all the oth-
ers, because it is used to detect problems, take decisions, and make up
for errors or weaknesses in the other sub-competences” (ibid.). This im-
plies that it is in fact the central sub-competence although this role has
already been ascribed to the transfer competence (as quoted above). Yet,
the most ill-defined sub-competence is the psycho-physiological sub-
competence defined briefly and vaguely as, “the ability to use psycho-
motor, cognitive and attitudinal resources”.
The researchers involved in the PACTE research group realized them-
selves that there were some shortcomings and confusions of the compo-
nential view of translation competence published in 2000 and re-assessed
them in 2003. Although the model of TC is still open to re-
conceptualization, its major outline was presented in the 2003 publication
entitled, “Building a translation competence model”.
160 Chapter 4

Most notable changes consisted of:

1. getting rid of the transfer sub-competence but instead including it in

the definition of translation competence which now is defined as, “the
underlying system of knowledge needed to translate. It includes de-
clarative and procedural knowledge, but the procedural knowledge is
predominant. It consists of the ability to carry out the transfer process
from the comprehension of the source text to the re-expression of the
target text, taking into account the purpose of the translation and the
characteristics of the target text readers” (PACTE 2003: 58);
2. replacing language sub-competence by bilingual sub-competence and
defining it as “predominantly procedural knowledge needed to commu-
nicate in two languages”. This requirement is already questioned by the
existing knowledge about bilingualism (see chapter 2) as it would only
refer to natural bilinguals who however as pointed out by Grosjean
2002, Nida 2000, and others, are not the best material for professional
translators since translation involves the need to keep the two languages
of a bilingual consciously apart. The further comment that the bilingual
sub-competence should include the “specific feature of interference
control when alternating between the two languages” is very much wel-
come but it has to be also admitted that the conscious control of inter-
ference means that one has to consciously monitor ones linguistic out-
put employing metalinguistic knowledge which is declarative and not
procedural (see Paradis 2009). A similar welcome expansion of the bi-
lingual sub-competence is made in the 2003 publication, to include
communicative rather than language competence marked by reference
to pragmatic, socio-cultural aspects of language use as well as gram-
matical and lexical knowledge of the two languages;
3. Other changes introduced to the 1998 model of TC included shifting
the knowledge about the translation processes component to the in-
strumental/professional rather than extra-linguistic sub-competence;
4. The strategic sub-competence after getting rid of the transfer sub-
competence was granted a crucial role in TC since it performs central
executive roles and is used to: “plan the translation project; activate,
monitor and compensate for shortcomings in other translation sub-
competencies; detect translation problems; apply translation strategies;
monitor and evaluate both the translation process and the partial results
obtained in relation to the intended target text, etc.” (PACTE 2003: 57);
 Developing translation competence 161

5. Finally, the previously introduced psycho-physiological sub-

competence was redefined as psycho-physiological ‘components’ on
account that “it forms an integral part of all expert knowledge”
(PACTE 2003: 57). The change though is rather controversial. First,
there is not much room left for development if something is called
components. Secondly, the psycho-physiological sub-competence car-
ried some hope that translation competence will finally be humanized
and see the translator as a developing human expert in the skill of
translation who does not only have to possess the five sub-
competences but is actively involved in the integrated use of all the
knowledge needed for a specific task.

To sum up, as compared with the other sub-competences the largest expan-
sion of the elements took place in the bilingual sub-competence which re-
sulted in explicit importance granted to the previously left out issues of
language interference and language control involved in language use for the
purpose of translation. Since the changes to the 1998 model were achieved
in the process of empirical studies which used a number of research tools
such as questionnaires, observation sheets, PROXY3, and retrospective
think aloud protocols, the refinement introduced in the bilingual sub-
competence allows to assume that the bilingual competence is the first to
undergo major transformations as the result of translation experience aimed
at acquiring translation competence. This aspect was raised by Toury
(1995) in his interlingualism postulate and by Presas (2000) in her interlin-
gual bridges metaphor and seems relevant in view of the dynamic nature of
bilingual knowledge discussed in detail in chapter 2.
The revised model of translation competence can be graphically repre-
sented in the following way:

–––––––––
3
A computer programme similar to Camtasia used to record the data flow between
the computer user and the Internet pages which are consulted during the translation
process.
162 Chapter 4

Fig. 1. Model of translation competence revisited.4

The authors (PACTE 2003: 60) themselves admit that the model is open
to modification and the major purpose of the extensive research project is
to expand our knowledge “about how translation competence functions
and how it is acquired” (PACTE 2003: 61). This will inform the translator
training programmes and will hopefully lead to better curricula design
(see Kearns 2006). However, the attempt to decompose translation com-
petence is not approved by some translation scholars who prefer to see TC
as just one thing (Pym 2003).

–––––––––
4
The EMT (European Master’s in Translation) expert group set up in April 2007 by
the EU to make specific proposals with a view to implementing a European reference
framework for a Master's in translation also took a compositional approach and sug-
gested 6 areas of competence: translation service provision competence, language com-
petence, intercultural competence, information mining competence, thematic compe-
tence and technological competence (http://ec.europa.eu/emt).
 Developing translation competence 163

4.1.3. The holistic perspective on TC

A list of shortcomings of the componential perspective on TC according

to Waddington (2000: 135) includes the following factors:

1. it is hard to know how many components should be a part of transla-

tion competence;
2. the definitions tend to concern ideal competence, and are thus incom-
plete without a model of the learning process (cf. Toury 1995: 238);
3. there is a dearth of empirical evidence for most of the available mod-
els (quoted after Pym 2003: 487).

Pym adds another shortcoming of the componential models of TC which

includes heavy assumptions not just about what translation is and how it
should be taught, but about the level at which specific teaching is needed,
and for how many years. These assumptions according to Pym reflect the
researchers’ bias exerted by their institutions for which they work and
which offer curricula design according to their own perceptions of transla-
tion and an ideal translator who should know “A, B and C”. To quote,
“Multicomponentiality has undoubtedly followed the fragmentary devel-
opment of the profession; it is obviously a response to interdisciplinarity
and the break with linguistics; but institutionally it operates as a political
defense of a certain model of translator training. And that model is not the
only one, nor necessarily the best” (Pym 2003: 487). Pym (2003) refers
here to the Spanish system of translator training which divided language
teaching from translator training following the trend in the 1980s and 1990s
urging for the discontinuity of translator training from language learning at
tertiary level in some countries. Advocated by renown scholars including
Snell-Hornby (1994) it did lead to “cutting the umbilical cord” (Snell-
Hornby 1994: 433) with Modern Language Departments in Spain, the
country where the PACTE group is based. Pym explains their professional
bias in the following words: “Now, once that battle was won (in Spain in
1991), many of the independent translation schools then had to fill up their
four- or five-year programs. As we have seen, those institutions had a direct
interest in multicomponentiality” (Pym 2003: 492).
Rejecting the multicomponential perspective on TC, Pym suggests re-
turning to the concept of transfer competence initially suggested by Toury
(1984, 1995), Krings (1986b), Koller (1992) which in line with Wilss
164 Chapter 4

(1976: 120) becomes a kind of “supercompetence” that explains the “sin-

gular specificity of translation” (Pym 2003: 488). To quote, “the minimal-
ist “supercompetence” approach means accepting that there is no neat
definition of all the things that translators need to know and will be called
upon to do. Nor is there any reason to suppose that competence is at all
systematic, like the grammatical and phonological rules that once pro-
vided the term with its archetypal content. What we need, beyond lists
and systems, is a concept that might define translating and nothing but
translating. Only then could we orient the rest” (Pym 2003: 488).
In the minimalist definition of TC Pym (2003) refers to his earlier writ-
ings which located the development of two-fold functional competence5 in
translator trainees described as “The ability to generate a series of more than
one viable target text (TTI, TT2 … TTn) for a pertinent source text (ST);
The ability to select only one viable TT from this series, quickly and with
justified confidence”. As put by Pym (2003) the union of these two skills
concerns translation and nothing but translation although he admits “there
can be no doubt that translators need to know a fair amount of grammar,
rhetoric, terminology, computer skills, Internet savvy, world knowledge,
teamwork cooperation, strategies for getting paid correctly, and the rest, but
the specifically translational part of their practice is strictly neither linguistic
nor solely commercial. It is a process of generation and selection, a prob-
lem-solving process that often occurs with apparent automatism” (Pym
2003: 489). Looking at these text generating and text selecting abilities
however, its exclusiveness to translation is mistaken. In creative writing one
also generates and selects from many versions. Another drawback of the
minimalist definition of the translation competence offered by Pym is the
lack of reference to the position from which TC is viewed. Traditionally TC
defined as the underlying knowledge needed to translate assumes the ideal
in line with the use of the term in linguistics (Chomsky 1965), that is the
state of the knowledge ascribed to the idealized professional translator
which would provide a point of reference for translator training methodolo-
gies. It seems that Pym does not see TC as an idealization but blurs the line
with the translation performance of a professional translator. Yet, as Pym
observes himself “competence cannot be confused with questions of profes-

–––––––––
5
Note the term ‘functional competence’ which seems is opposition to the traditional
account of competence (the underlying knowledge) vs. performance (manifestation of
competence in action).
 Developing translation competence 165

sional qualifications, no matter how teachers like myself might worry about
training students for the workplace” (Pym 2003: 482)6. Still, it is unlikely
that the view of TC as the ability (see the discussion of confusing the terms:
ability, skill, competence in chapter 1) to generate more than one viable tar-
get text and the ability to select just one with confidence is that taken by the
professional translator at work. The only perspective in which this view of
TC could apply is the initial stages of translator training when these two
abilities are stressed and practiced. What is more, translation as any other in-
stance of language use is a process of choosing between viable alternatives,
this need to choose is not exclusive to translation at all and neither is ‘what
drives generation and selection of optional expressions referred by Pym as:
“that kind of reductive reasoning, impeccably pragmatic, [which] is perhaps
the hardest part of translation competence to convey” (Pym 2003: 490). Al-
though the justification offered by Pym (that the major virtues of the will-
fully minimalist definition lie in things it does not say) might be intellectu-
ally interesting, it is vastly unhelpful in informing aspiring translators about
the knowledge base they need to acquire and which will serve them in their
future professional life as a translator.
However, in his defense of the minimalist definition of translation
competence Pym (2003) points to the fact that although the minimalist
view of TC does not indicate that there is an ideal translator profile it al-
lows for other competencies to join in to the effect that the translation
competence might become a “minor component in the range of skills re-
quired of intercultural professionals” (Pym 2003: 491). In recognizing
that translators in fact require multicompetence (Gouadec 2007) Pym ac-
tually echoes the PACTE’s view of expert knowledge which implies mul-
ticompetence. In view of the minimalist understated complexity both atti-
tudes to TC the multicomponential one and the minimalist one do not
have to be seen as mutually exclusive but they might just look at transla-
tion competence from a different perspective. PACTE is looking at it from
the perspective of the translator trainer and the minimalist approach is
taken from the perspective of the professional translator. As it can be ex-
pected the views taken from such different vantage points will differ. The
translator trainer is aware of the fact that the beginning, novice translators

–––––––––
6
The EMT expert group define competence as “the combination of aptitudes,
knowledge, behaviours and know-how necessary to carry out a given task under given
conditions” (http://ec.europa.eu/emt).
166 Chapter 4

cannot see the forest for the trees when with possibly limited bilingual
knowledge they embark on a translation course. To be weaned away
(Snell-Hornby’s term 1992) from their natural translation strategies (de-
scribed in chapter 3), it is best to show them all the knowledge compo-
nents that participate in the translation process. The minimalist approach
would apply to the professional experienced translator who having gone
through the forest himself/herself maybe, quite a long time ago, can no
longer see the trees, which here are a metaphor for the hurdles every indi-
vidual novice translator has to surpass to integrate all the kinds of knowl-
edge which participate in the process of producing a translation. The de-
velopmental perspective in which the translator is viewed in this work
points to an obvious fact that it is only after getting through the forest and
negotiating the passage through the trees that it is finally possible to ap-
preciate the view of the forest, i.e., TC in its orchestrated entirety7.
To sum up, translation scholars disagree as to what should be included
in the definition of translation competence to the point that some might
suggest that in fact such a definition is not needed. Since the tasks per-
formed by translators change with the market demands, the definition of
translation competence would have to be anyway constantly revised to
keep up with the growing demands placed on the work expected of transla-
tors (see Pym 2003). Fraser (2000: 53) explicitly admits that TC is used “as
shorthand for the skills, expertise and judgment that a professional transla-
tor develops from a combination of theoretical training and practical ex-
perience”. Needless to say TC used in this sense remains intangible in
terms of the actual contents. Having shown the lack of unanimous agree-
ment as to what constitutes TC it is equally regrettable that despite the long
history of translation practice and the constantly growing demand for trans-
lation services the process of translation competence acquisition has not
been a frequent object of empirically validated research (see Cronin 2005).
“If few studies of translation competence exist, there are even fewer of
translation competence acquisition” (PACTE 2003: 49). Let us now exam-
ine what is available in the literature that could be used to provide guidance
to novice translators starting their developmental route towards translation
as a trained skill performed by a professional translator.

–––––––––
7
Compare the progression from unconscious incompetence to conscious incompe-
tence to conscious competence and finally to unconscious competence (González Davies
2004).
 Developing translation competence 167

4.2. Acquiring translation competence

Although there have been some empirical studies comparing the transla-
tion performance of students (from modern language departments and
from translation training programmes) and professional translators with
the conclusion that both groups differ in terms of how they approach the
task (Jääskeläinen 1989, Tirkkonen-Condit 1990, Jääskeläinen and
Tirkkonen-Condit 1991), no attempt (at least to my knowledge) has been
made to follow up the development of translation competence in a longi-
tudinal study (however some research projects have been set up, e.g.,
Orozco and Hurtado Albir 2002, PACTE 2003). “It is generally assumed
that students develop translation competence through their studies on
translation. However, nobody has yet defined the process through which
students acquire translation competence” (Orozco and Hurtado Albir
2002). Several researchers especially those who are actively involved in
translator training (Kussmaul 1995, Kiraly 1995, 2004, 2005, Cronin
2005, Seguinot 1997, Gile 1995, Hansen 2006) have made some propos-
als which although fragmentary can be used to stake out the route along
which the untrained ability to translate (discussed in chapter 3) develops
into a trained skill with its underlying translation competence.

4.2.1. Some suggestion of how TC can be acquired

The PACTE group defines translation competence acquisition as:

1. A dynamic, spiral process that, like all learning processes viewed from
a cognitive perspective evolves from novice knowledge (pre-
translation competence, i.e., natural ability to translate discussed in
chapter 3) to expert knowledge (translation competence); it requires
learning competence (learning strategies) and during the process both
declarative and procedural types of knowledge are integrated, devel-
oped and restructured;
2. A process in which the development of procedural knowledge and,
consequently, of the strategic sub-competence are essential;
3. A process in which the translation competence sub-competencies are
developed and restructured (PACTE 2003: 49-50).
168 Chapter 4

Several reservations are made which allow for relations, hierarchies and
variations in the process of acquiring sub-competencies. They include the
following points: 1) the sub-competences are inter-related and compen-
sate for each other; 2) do not have to develop in parallel; 3) are organized
hierarchically; 4) variations occur in relation to translation direction (di-
rect or inverse), language combinations (languages which do not share the
same script, e.g., English and Chinese), specialization (legal translation)
and the learning context (formal training, self-learning). A reservation is
made that the learning context and the teaching methodologies will di-
rectly influence the acquisition process, something that was acknowl-
edged a long time ago in SLA research (see chapter 2 on the relations be-
tween the teaching methods and the mental lexicon structure).
All the points seem relevant, but what is missing is the description of
novice knowledge, explained here as pre-translation competence. If we
assume that learning new skills and knowledge, e.g., in the process of TC
acquisition “is a dynamic process of building new knowledge on the basis
of the old. What may be most important in the process is the restructuring
of existing knowledge, rather than acquisition of new information”
(PACTE 2000: 103). Therefore, defining what is the assumed nature of
pre-translation competence is essential. A complete novice to translation
will have only bilingual knowledge to rely upon and extralinguistic sub-
competence. As I pointed out in chapter two the bilingual foundations of
translation competence are complex. Bearing in mind the difficulties in
describing TC as expert knowledge in the previous section, it might be
easier to make some assumptions which can always be questioned but
which nevertheless will provide some points of reference.
Let us then assume that the novice translator is in fact an L2 learner.
Since I described the way L2 learners translate in chapter three, let me
here refer to Presas (2000: 29) who views the development of translation
competence as consisting of three basic learning processes: “(1) the ac-
quisition of previously non-existent competences; (2) the restructuring of
already existing competences in order to facilitate transfer competence;
(3) the acquisition of strategic competence” (Presas 2000: 29). Within the
transfer competence Presas points to the following psycholinguistic proc-
esses which have to take place when the reorientation of bilingual compe-
tence towards interlingual competence occurs: “(1) specializing in com-
municative competence in two languages (oral or written, reception or
production); (2) restructuring, reorienting and broadening mechanisms of
 Developing translation competence 169

code switching and bilingual memory; (3) integrating the mechanism to

control interference” (Presas 2000: 29). Suggestions made by Presas
(2000) are very much in line with the PACTE proposal, and although nei-
ther Presas nor the PACTE group refer to one another in an explicit way,
they both imply that the fundamental restructuring in the process of TCA
(translation competence acquisition) takes place within the bilingual sub-
competence. It is possible that the experience of translation combined
with feedback received from translation trainers has the power to trans-
form the bilingual competence into an interlingual competence (see chap-
ter 3). It is something that was suggested by Toury (1995) in his chapter
entitled “A bilingual speaker becomes a translator” and ascribed to the so-
cial motivation behind the development of translation skill and the social
function of translation as a final product (Toury 1995: 148). As I sug-
gested in chapter 3 this transformation is marked by awakening or aug-
menting the novice translator’s awareness of intercultural and metalin-
guistic aspects involved in the act of translating. Similarly, Kiraly (1995:
101) in his Pathways to Translation talks about the need to emphasize the
acquisition of interlingual, intercultural and intertextual associations in
translation trainees.
This reorientation of bilingual competence to include the interlingual
route of access to both languages in the mind of a beginning translator in-
stigated a debate on whether the acquisition of translation competence fa-
cilitated by formal translator training should be altogether separated from
L2 teaching and learning (Snell-Hornby1992: 433). Some countries, e.g.,
Spain, employed the postulate of total separation between translator train-
ing and L2 teaching and learning as if denying the continuity of transla-
tion skill development. The outcome experiences have shown that the
separation goes against the evolutionary nature of translation as a human
skill which has to be built on bilingual foundations. Translation compe-
tence and bilingual knowledge remain in an active mutually beneficial
symbiosis. Indeed, to quote Pym (2003),

Suddenly there was no pressing need to separate translation from bilin-

gualism, for example, just as there was little interest in the idea of sepa-
rate professional communities. It could all be brought under the one
roof; it would all be part of “translation competence.” In this sense stra-
tegic importance should be placed on attempts to define the special
ways in which not only translation should be taught, but also the way
languages should be taught for translators (cf. the “pre-translation com-
170 Chapter 4

petence” outlined in Presas 1998). The discontinuity that was once

sought by virtually all translation scholars has now become a rather thin
and very debatable line (Pym 2003: 492).

It is precisely then that the importance of bilingual foundations was in fact

recognized as encompassing what Presas (2000) called ‘pre-translation
competence’. The above view points to the shared interest between modern
language departments and translation training programmes which lies in
educating L2 language experts who will use their L2 expertise for various
purposes, and translation is then to be seen as a specialization of L2 exper-
tise. As pointed out in chapter 1 and 2 both areas, bilingual knowledge and
translation competence benefit from translation experience.
Focusing on the specialization of bilingual skills in the process of TC
acquisition it seems plausible to suggest that the beginning translators have
to become interlingual language users and as such they have to develop cer-
tain mechanisms which will allow them to control their language use (Green
1993). In other words they have to restructure their bilingual knowledge to
become conscious language users of their L1 and L2. Presas (2000: 29)
mentioned restructuring, reorienting and broadening mechanisms of code
switching and bilingual memory and integrating the mechanism to control
interference. This implies that an additional competence (i.e., the ‘super-
competence’ suggested by Wilss 1976) needs to be developed in the process
of re-orientation from bilingual language use to interlingual use in transla-
tion. Kiraly (1995: 101), relying on cognitive and psycholinguistic ap-
proaches to translation, suggests that attention should be paid to cognitive
resources that translation students need to acquire. As observed by Shreve
(1997: 124), “The cognitive basis of professional translation may derive
from cognitive skills shared with bilinguals, but we must make the case that
other cognitive structures have been added and that a variety of specific
transformations of the knowledge of translation and of how to translate oc-
cur if and when an individual embarks on a course of acquiring language
experiences of a certain type, for instance, translation experience”. Similar,
Martin (2006) pointed out that the beginning of translation competence ac-
quisition starts with enrolment into translator training programmes where
primarily students develop a deeper understanding of translation as an activ-
ity and as a mental process. Research involving L2 learners with wider
translation experience (Whyatt 2007b) or translation trainees (Kussmaul
1995) showed that increased understanding of the translation process gained
 Developing translation competence 171

through repeated translation practice (professional sub-competence) had a

positive effect towards translation performance. In a similar vein Orozco
(2000: 201) who defines the aim of her research as “to shed new light on the
whole process of acquisition of translation-specific skills and abilities, as a
necessary step in finding an effective translation teaching methodology that
will guide trainee translators in their learning process” suggests that the
changing notions about translation in the translation students are a index of
their progress in terms of translation competence.
In PACTE’s view students who start to specialize in translation come
to acquire a range of sub-competences (discussed above) in the sense that
they are instructed or guided to discover what aspects of the translation
process are also important and have to be incorporated in their use of bi-
lingual knowledge for the purpose of translation. In a way their meta-
cognitive awareness of the translation process and its social function
starts to develop. In PACTE’s estimation (2000) however, the novice
stage of TC acquisition “could be defined as the stage when the sub-
competences have been acquired, at least partially, but they do not interact
with each other” (PACTE 2000: 103), that is they are not integrated into
an efficient cooperating network of knowledge and skills. González Da-
vies would probably suggest that the novice stage can be described as the
sate of conscious incompetence as the students first become aware of
what they still do not know. Whyatt (2007b) quoted some comments ex-
pressed by novice translators showing their disappointment that although
they knew both languages they felt let down by their knowledge when
translating. The comments included the following:

“I had tremendous problems recalling words from memory”

“my knowledge of the FL is not sufficient to translate”
“I spent so much time on using dictionaries”
“dictionaries do not always help”
“I thought I knew my own language but I had so many doubts”
“I thought I knew this word but it meant something else”
“translation is such a slow tedious process I could never make a living out
of it”

From this depressive start when the students’ linguistic confidence is a bit
shattered there is a gradual process of learning to awaken their self-
awareness and eventually build their self-confidence as translators. In a
172 Chapter 4

way their initial disappointment with the fact that having the knowledge
of the two languages they still experience difficulties when using them for
the purpose of translation reflects the common naïve view that everybody
who knows two languages can translate (see Hejwowski 2004). This
widespread opinion clashes with the students’ translation performance and
as a result they start to develop a new set of skills (Whyatt 2007b). As I
pointed out in chapter 3 the experience of translation leads to an increased
understanding of its nature and purpose and as a result it brings about a
transformation within the bilingual competence of advanced L2 learners
at university level. The transformation (discussed in chapter 3) includes
intercultural awareness underlying languages and metalinguistic aware-
ness which might prove to be a pre-cursor of strategic competence which
in competent translators encompasses the entire translatorial action (see
Holz Mänttäri 1984). Thus, the assumption that in novice translators the
progression on the translation competence scale entails a dynamic interac-
tion between the sub-competences until the hierarchy suggested by
PACTE is established with strategic sub-competence holding responsibil-
ity for the entire process and product of translation seems plausible. What
is still missing is a change of perspective to see the progression in transla-
tion as a trained skill from the point of view of the actual trainee who
comes to understand the nature of the translation process perceived as a
communicative social action and who gradually experiences the transfor-
mation within himself/herself.
Although scholars differ on how TC is best defined they all agree that
translation competence is experience-derived knowledge. Examining the
links between TC and the experience of translating might shed some more
light on the evolution of translation as a human skill to reach a profes-
sional level.

4.2.2. The correlation between TC and translation experience

(Shreve 1997: 127) points to the fact that “individuals who practice trans-
lation professionally have developed their ability to do so deliberately and
in the process of the deliberately sought language experience of translat-
ing”. This observation points to two factors involved in the process of
learning to translate: the learning person and the learning environment.
Shreve (1997: 127) also points to the fact that the experience of translat-
ing can be acquired in several ways including translation schools, or “by
 Developing translation competence 173

receiving mentoring from another translator”, “or by independently seek-

ing out and completing translation tasks”. Providing the trainee with the
experience of translation becomes the starting point in the process of de-
veloping the human skill of translation. Some scholars have suggested
certain aspects of translation experience which play an essential role in
the making of a translator as a trained professional.
Toury (1995) came up with the concept of socialization into the profes-
sional community. He identified the novice translator as primarily lacking
self-confidence in the steps taken to solve translation problems. Like a nov-
ice in any other complex skill, the beginning translator needs feedback and
guidance. The process described by Toury as socialization allows the be-
ginning translators to acquire and assimilate norms, to “develop routines
which enable them to cope in an effective way not with some abstract no-
tion of ‘translation’, but with a concrete task, that of translating in compli-
ance with what conception of that mode of text production is pertinent to a
particular societal group” (Toury 1995: 250). Toury points to the fact that
the norms which are internalized by the novice translator may differ with
respect to many factors including place and time and therefore the process
of initiation into the translating profession may have to be repeated. How-
ever, there are some issues which remain very vague, including the most
fundamental questions like: ‘When does the process of initiation in the pro-
fession start? Is it with entering the translation training programmes, or is it
with starting to make a living as a translator? How the productive routines
are developed in self-taught translators who have never received any formal
training and yet they are successful respected professionals?
Toury (1995) suggests that a bilingual person who becomes a transla-
tor-in-the-making has to embark on a journey of self-discovery. To quote,

What trainees really need is the opportunity to abstract their own guid-
ing principles and routines from actual instances of behaviour, with the
help of responses to their performance which are as variegated as possi-
ble. Thus during the training period, the pedagogically most appropriate
concepts are those associated with experiencing, exploration and dis-
covery, involving as they do a considerable element of trail and error
(Toury 1995: 256).

The experience, however and the process of self-discovery has to be at the

same time responsive to the environmental feedback which according to
Toury (1995) is normative in nature.
174 Chapter 4

In the initial stages of a translator’s development, the feedback directed at

him/her is exclusively external: a novice simply has no means of assess-
ing the appropriateness of various possible products and/or of the alterna-
tive strategies that would yield them. (…) The only way is to have the
culturally acknowledged criteria supplied from without, namely by those
who already have, or believe they have them, and who are powerful
enough to impose their will. At the beginning, the main environmental
feedback one receives is thus overt responses to one’s verbal products, fi-
nal or interim. In the latter case the responses may act as a kind of an ex-
ternal monitoring device, to the extent that the translator immediately re-
sponds and revises his/her production accordingly. (…) Little by little,
however, translators might start taking potential responses into account,
too, i.e., during the production itself. They thus develop an internal kind
of monitoring mechanism, (…) (Toury 1995: 250).

Toury’s suggestions of the internal monitoring device are in line with the
observation of growing metalinguistic awareness in L2 learners who are
involved in translation tasks.
The role of feedback in the development of translation as a trained
skill becomes an essential part of the normative effect of translation ex-
perience past the stage of translation as a natural untrained ability of bi-
linguals. Kiraly’s (1995) observed that to progress further in the process
of evolution of translation as a human skill the translator in the making
needs guidance and feedback to develop what he called the translator’s
self concept, that is the awareness of his/her role in the translation proc-
ess. Feedback both external (from trainers, peers, mentors or clients) and
internal (the translator’s self-reflection) is thus an essential part of the
translation experience from which translation competence derives.
Kiraly (1995, 1997, 2005) in his social-constructivist approach8 com-
bined the social and the individual perspective in the context of translator
training. To quote,

The approach to translation skills instruction that began to evolve in

Kiraly (1995) involves a doubly articulated view of translation process:
from the translator’s perspective, looking outward toward the social
situation in which professional translation occurs, and looking inward
toward the mental processes going on in the individual translator’s
mind during the production of a translation (Kiraly 1997: 139).

–––––––––
8
According to which humans learn from experience and from being actively in-
volved in action.
 Developing translation competence 175

This approach that combines the social and the individual psycholinguis-
tic/cognitive perspective of the translator-in-the-making draws attention
to the need to structure the learning environment in a way which will pro-
vide the most appropriate translation experience and feedback for transla-
tion trainees (Schäffner and Adab 2002). Toury (1995: 256) admitted that
the environmental feedback which trainee translators receive is at risk of
being limited and one-sided. To counterbalance this deficiency of external
feedback Toury (1995) suggests exposing students to additional kinds of
feedback more representative of the real world, including “less profes-
sional parties”, “non-conformist tutors” and peers acting as “assessors in
all seriousness” (Toury 1995: 256-257). All the parties mentioned by
Toury can act as potential receivers of the translation product, or editors
responsible for the quality of translated texts.
Responding to external feedback is a crucial factor in shaping the
translator and allowing him/her to progress from translation as an un-
trained ability to translation as a trained skill. The range of experiences
and the fact whether they were or were not combined with constructive
feedback will sensitize the novice to the cues which have to be picked
up in the translation process and which are usually unnoticed by bilin-
guals acting as natural translators. Such cues include a multitude of fac-
tors such as style, cohesion, coherence and other socio-cultural factors.
These cues as pointed out by Shreve (1997) are noticed when there is a
change of focus from the micro to the macro context or when in the
process of text comprehension a text is processed as a whole, in a more
vertical sense and not in a horizontal string-like fashion which is symp-
tomatic of those translators who rely on their bilingual abilities only
(Jonasson 1998). As pointed out by Shreve (1997) these macro-textual
clues become noticeable by novice translators only when a certain
amount of translation experience has been gained which led to a change
in the translators’ understanding of translation as an activity (Orozco and
Hurtado Albir 2002, Presas 2000, González Davies 2004). The novice
translator has a different perception of translation than the professional
translator and experiences different translation tasks. Similarly, “the
novice translator has different expectations of the process. This may
mean that there is a selective attention on certain kinds of translation
cues, for example lexical cues, at the expense of others” (Shreve 1997:
135). As observed by Shreve (1997: 136) “The movement within the
translator ability space is related primarily to the nature, range, and fre-
176 Chapter 4

quency of translation tasks over the course of a translator’s acquisition

history”. Shreve (1997) probably refers to the fact that the intensity of
translation experience can stimulate the development of translation
competence whereas infrequent translation experience will not have a
bearing on the progression on the translation skill continuum, more so if
it is not combined with external guidance or appropriate corrective feed-
back. Similarly, the kinds of translation tasks aimed at translators at dif-
ferent points on the developmental continuum differ in their level of
complexity and the experience of translating them results in different
cognitive gains for those who experience the task. To illustrate the point,
L2 learners are not likely to be expected to practice translation on a 10-
page financial report or on an academic paper concerning geology since
they struggle with translating a recipe for an Old English Apple Pie
(Whyatt 2008). The most appropriate selection of translation tasks and
adjusting them to the stage of translation skill development of the train-
ees is a matter of discussion for curricula design (see Kearns 2006,
Kelly 2005).
Kiraly (1995) observes that with increasing experience in translating
“skills are less likely to be acquired by repeated practice; less likely to
develop naturally without specific training and pedagogical intervention,
and more likely to involve translation quality at levels beyond that of
mere semantic and syntactic correctness” ( quoted after Kelly 2005: 15).
In his proposal of the “translator’s self-concept” defined as the growing
awareness of the role the translator plays in the process, he points to the
development of “a functioning translation monitor” (cf. Toury’s (1995:
250), internal kind of monitoring mechanism’).
In view of the accumulating knowledge about the translation process
(see Krings 1986a, Gerloff 1986, Lörsher 1986, 1991, 1992; Tirkkonen-
Condit and Jääskeläinen 2000, Hansen 2002) the recognition of the need
to replace the traditional teacher-centred transmissionist approach to
teaching translation (still somehow present in Toury 1995) has become
widely recognized. As observed by Cronin (2005) although translation
schools have been in operation for more than half a century now at least
in western Europe (see Cronin 2005: 250), it is only in the 1990s that, the
importance of providing translation trainees with genuine translation ex-
perience which would reflect the real life requirements of the translator’s
work was recognized (Kiraly 1995, Cronin 2005). The gradation of the
translation experience in a formally structured environment has become
 Developing translation competence 177

an issue of concern for translator trainers. In pedagogical practice, how-

ever it turned out that despite the fact that translation as a profession has a
long history there is a vast pedagogical gap when it comes to the institu-
tionalized training of future translators.

4.3. TC acquisition and the “pedagogical gap”

In the late 1980s and early 1990s rapid changes in the political and eco-
nomic life in Europe, together with growing globalization brought about an
equally rapid transformation of the translation market (Kiraly 2005). The
demands for translators led to the establishment of translator training pro-
grammes throughout Europe from east to west. The “pedagogical gap” was
severely felt by many trainers. The shortage of guidance led to few ground-
breaking publications which at least offered a different more promising
way of teaching translation so that the skill developed during training
would prepare the students for their professional challenges of the market
which needs them. It was in 1995 that Gile published his Basic Concepts
and Models for Interpreter and Translator Training, Kussmaul his Training
the Translator and Kiraly his Pathways to Translation. Pedagogy and
Process. The three publications contributed major insights into how the
learning environment should be structured to provide the best conditions
for the trainees who are motivated to develop their translation skill into pro-
fessional expertise. Let us briefly discuss their major contributions.

4.3.1. Focus on the translation process

The major source of the need to revise the learning environment in which
translation skill is developed to reach professional excellence comes from
the growing understanding of the translation process. Within the practi-
cally oriented school in the new discipline of Translation Studies in the
mid 1980s growing interest in the translation process brought an empiri-
cal turn in studying translation. Translation scholars from Germany
(Krings 1986b, Lörsher 1986, 1991, 1992, Kussmaul 1995) and Finland
(Tirkkonen-Condit and Jääskeläinen 2000) started to use TAPs to investi-
gate how language processing during translation is handled by the human
mind. As expressed by Cronin (2005: 251) with all the imperfections TAP
research provides “some way of getting to know what goes on in the
translator’s head and therefore allows us to formulate, however tenta-
178 Chapter 4

tively, a psychologically valid description of the act of translation”. The

1980s interest in translation as an intellectual process of problem solving
and decision making provided a lot of insight into the complex nature of
the mental operations that are carried out in the translator’s mind. Until
then most teaching methodology was product oriented, focusing on com-
parison and contrast between the SL text and its translation, or between
the translations performed by inexperienced translators and the
model/ideal translation produced up to professional standards. The mental
journey which started with reading the SL text and finished with present-
ing the TL text was believed to be inaccessible and confined to the ‘black
box’ of the translator’s mind. This long-lasting belief was possibly very
flattering for experienced translators as it created reverence and a belief
that translation is a special gift which is granted to few individuals. Espe-
cially in the area of literary translation, the skill was, and to some extent
still is, believed to be a matter of ‘talent’. Nida’s (1981) title “Translators
are born not made” serves here as a sufficient piece of evidence. Baker
(1992) in a much revealing publication firmly stated that translators as
any other professionals should be able to explain the source and principles
of their expertise instead of saying that they are translating well because
they have a flair for it. Piotrowska (2007) points out that the debate
whether the skill of translation is art or craft is from today’s point of view,
when translation is first and foremost a business, rather dated. To meet the
market demands more and more translators need to be trained and the
training has to prepare them for their future career in translation. As in
any other profession marrying individual predispositions and inclinations
(their talents) with professional training (see Gile (2009: 7) quoting
Healey’s (1978) view that “Translators are made not born”) is an ideal
match most likely to reach the level of expertise.
The increased understanding of the problem solving and decision
making that goes on in the mind of a novice inexperienced translator
brought about suggestions that process-based approaches to translation
training are more likely to succeed than previously applied product-based
approaches. These suggestions seem to echo an appeal made by James
Holmes who said “It seems to me that before we can know how to train
translators, we have to know what takes place in the translation process”
(1988: 95-96). Holmes’ conviction is still shared by many scholars
(Shreve 1997, Kiraly 1995, Hansen 2003, Cronin 2005) with keen interest
in relating the findings of the translation process research to translator
 Developing translation competence 179

training. Although the effort to unravel all the intricacies of the translation
process continues and remains an open area of investigation Holmes’s
conviction holds valid and allows to hope that the more we know about
the translation process the more we will know about the process of TC
acquisition. Current research allows one to assume that depending on the
stage on the developmental continuum of translation as a human skill that
a given translator has reached, the translation process is handled differ-
ently and therefore it results in different products in terms of quality (see
chapter 3). Consequently, making translation trainees aware of the mental
processes which need to take place for a SL text to be translated into a TL
text in order to bring the most desirable translation products thus becomes
a vital part of their professional training.
Gile (1995) is emphatic that translator training should adopt a process-
oriented approach. Since a lot of his suggestions refer not exclusively to
interpreting but to Translating, where the capital ‘T’ denotes both transla-
tion and interpreting, the quote below is valid also for teaching written
translation:

The idea is to focus in the classroom not on results, that is not on the
end product of the Translation process, but on the process itself. More
specifically, rather than simply giving students texts to translate, com-
menting on them by saying what is “right” and what is “wrong” in the
target-language versions produced, and counting on the accumulation of
such experience and indications to lead trainees up the learning curve,
the process-oriented approach indicates to the student good Translation
principles, methods, and procedures (Gile 1995: 10).

The verb, ‘indicates’ points to implicit guidance rather than to explicit in-
struction. The importance of raising awareness in beginning translators of
the process which accompanies the intellectual activity of translating has
been acknowledged by many scholars. The process of translation has been
divided into different stages, for example Ljudskanov (1969) saw only two
phases: SL text comprehension and TL text production (also Gile 1995),
Seleskovitch (1976) insisted that the translation process also includes the
stage of ‘deverabalization’ in between the comprehension and production
phase, when the translator has interpreted the message but has not yet
found the verbal means in the TL to express it. Other scholars termed the
middle stage differently, e.g., the transfer meaning stage. Wilss (1976: 118)
suggested a more detailed view of the translation process which includes
180 Chapter 4

SL text analysis, assessment of translation problems, solving translation

problems, assessment of equivalence between the SL text and the TL text
and re-expression. Bell (1991) in his psycholinguistic model offered a more
detailed description of the translation process which includes: SL text →
visual word recognition system → syntactic analyzer → semantic analyzer
→ pragmatic analyzer →TRANSLATE (planner, idea organizer, concep-
tual and semantic representation, general world knowledge) → pragmatic
synthesizer → semantic synthesizer → syntactic synthesizer → TL text.
Hönig (1991) proposed an idealized model of the translation process
stressing the fact that it takes place in two workspaces, the cogni-
tive/conscious space and the intuitive/subconscious space. While TAP re-
search is able to capture the cognitive conscious problem solving and de-
cision making, it is unable to access the intuitive processing which takes
place in the translator’s mind. The fact that professional and novice trans-
lators significantly vary in terms of the amount of verbalizations produced
during thinking aloud allows one to deduce that most language processing
performed by translating novices is cognitive rather than intuitive in na-
ture. The two modes of processing are generally acknowledged by theo-
rists and professionals with ample support from translation process stud-
ies (Jääskeläinen 1989). The problem-free spontaneous smooth transfer is
nonstrategic in nature and referred to as ‘automatic’ (Séguinot 1997). The
strategic problem-solving and decision making phase is non-automatic
and marked by a disruption in the progression of the translation process.
This disruption is caused by a psychological or cognitive barrier in the
form of a translation problem (see Piotrowska 2007: 62) which needs to
be solved. In the process-oriented approach it is assumed that drawing
students’ attention to how they process information, spot and solve prob-
lems and make final decisions will help them to understand the process
which frequently will involve non-linguistic knowledge, or will require
the use of non-linguistic knowledge to make linguistic decisions. Kiraly
(1997: 139 based on Boekaerts 1981) postulates that the two processes,
subconscious and controlled “do not necessarily form a dichotomy”
(Kiraly 1997: 151) in the process of translation. Quite on the contrary,
both subconscious processing and cognitively controlled problem solving
interact in the mental space of the translator’s mind (see Hönig 1991). The
uncontrolled (not consciously controlled) workspace is where spontane-
ous, automatic or ‘tentative translation elements’ occur and where transla-
tion problems surface “when automatic processing is incapable of produc-
 Developing translation competence 181

ing tentative translation elements. These problems are brought into the fo-
cus of attention in the controlled processing centre, and a strategy is cho-
sen and implemented in an attempt to deal with them” (Kiraly 1997: 151).
Kiraly stresses that the intuitive processing should be encouraged as it
plays an important role in the translation process. Piotrowska (2007) dif-
fers from Kiraly and postulates that in order to develop professional com-
petence in translation students their training should foster the strategic
(conscious/cognitive) approach.
Piotrowska (2007) in her pioneering book set in the Polish context, in
which she tries to lay the foundations of methodology for teaching written
translation suggests that translator training should focus on the conscious
strategic part of the translation process. Translation students should be
taught to recognize translation problems and apply strategies and proce-
dures which will help them to make decisions on how to solve the prob-
lem. Piotrowska (2007: 62) points out that the translation process can be
automatized in certain areas but nevertheless there will always remain a
vast area where conscious decision making will have to take place.
To sum up the vast body of knowledge on the translation process in-
cludes both prescriptive assumptions and empirical data and although the
process-oriented approach to translator training is a welcome change to
the product-oriented approach it has to be operationalized to provide
pedagogical guidance. In view of the multitude of sometimes quite subtle
knowledge needed to translate implicit guidance rather than explicit in-
struction becomes a favoured approach which in consequence leads to a
revolution in the translation classroom, which has led to the change in the
power relations between the trainer/teacher and the trainee/student. These
changes similar to the process-oriented approach are aimed at providing
the best learning environment for the development of translation compe-
tence in future professionals.

4.3.2. Learner-centred approaches to translation training

The growing body of knowledge about the nature of the translation process
has led those interested in searching for more effective teaching methods to
realize that the old fashioned transmissionist models have to be replaced. Al-
ternatives needed to be found to the methodology which relied on the ‘read
and translate directive’ in which future translators would be given a text to
translate at home and the class following their homework involved compar-
182 Chapter 4

ing the translation produced by individual students to the model translation

in the possession of the teacher. The methodology based on what Kiraly
(2005) calls WTNS (Who will take the next sentence) technique was not in
line with translation process research and very much behind the approaches
which revolutionized foreign language teaching methodology (González
Davies 2004: 11). The Communicative approach shifted the entire focus of
SLA from linguistic accuracy to pragmatics with its social rules of language
behaviour and complex discourse grammar. The Humanistic Teaching Prin-
ciples gradually replaced authoritative “chalk and talk” (González Davies
2004: 12) teacher-oriented methodology. The learner-centred approaches to
language teaching/learning meant that “for teaching to be effective, a learner
must be able to create meaningful and personally relevant patterns” (Caine
and Caine 1994: 90, quoted after González Davies 2004: 12). For the teach-
ing/learning to bring desired effects, the learner has to become involved in-
tellectually and emotionally into the learning process. New innovative ap-
proaches and concepts aimed at activating the learners appeared including
learner autonomy, reflective teaching, meaningful learning, and motivation.
These notions became to be advocated by pedagogically oriented translation
scholars such as Kiraly (1995, 2004), Cronin (2005), Colina (2003a), Gon-
zález Davies (2004) to name the most outstanding authors of publications
which have tried to bridge the pedagogical gap in teaching translation as a
trained skill. The delay in the emergence of the new approach in translator
training is significant. In 1971 Dell Hymes introduced the concept of com-
municative competence which became a ‘key concept’ in second language
methodology. Canale and Swain (1980) published a definition of communi-
cative competence which became the bedrock of the Communicative Ap-
proach. The first publications on translation teaching which included Hu-
manistic Teaching Princlples appeared in the mid 1990s (Kiraly 1995, to
some extent also Gile 1995 and Toury 1995) and early 2000 (González Da-
vies 2004, Cronin 2005, Colina 2003b). The new approaches combined the
advances in understanding the nature and process of translation with the
Communicative Approach, Humanistic Teaching Principles, Cooperative
Learning and Social Constructivism. The learner is no longer a receiver of
knowledge transmitted by the teacher but a subject in the very centre of the
learning environment, the teacher becomes a guide, an advisor, a partner in
the mutually shared learning process.
In Kiraly’s (2000) view the primary aim of translation training is to
empower trainees and enable them to develop their ‘professional selves’.
 Developing translation competence 183

The focus is shifted from teaching to learning. Teachers become assis-

tants, consultants and facilitators. Learning to develop translation as a
trained skill can only be achieved via ‘situated experience’ which on the
one hand is related to the student’s prior knowledge but at the same time
to the professional behaviour it aims for. With the focus on autonomy and
learning through authentic action based on the ideals of Social Construc-
tivism and Collaborative Learning, the learners are involved in real world
translation projects. In Kiraly’s (1997) own words,

Constructivism provides an educational framework that is particularly

compatible with the contemporary emphasis on process in translator
education. Rather than teaching students how to translate, this approach
sees the primary task of the instructor as that of putting students into re-
alistic professional translation situations where they will collaborate
with each other in constructing their own understandings of “appropri-
ate translation behaviour”. Collective and individual consciousness rais-
ing will be the teacher’s main goal rather than the unidirectional trans-
mission of information (Kiraly 1997: 157).

Creating this kind of learning environment provides according to Kiraly

conditions for acquiring translation competence. A classroom is turned into
a “forum for the development of expert, personalized and language-
mediating skills” (Kiraly 1997: 159-160). Translation students do not have
to go out of the classroom for real life work experience, but instead the pro-
fessional reality should enter the classroom and provide conditions for what
Kiraly calls ‘cognitive apprenticeship’ and ‘situated cognition’. Students
working in groups according to the principles of Cooperative Learning
where team work, on the one hand provides opportunities for generating
ideas, negotiating solutions and reflecting on the process, and on the other
hand contributes to more effective learning. When cooperating with others
students do not feel as inhibited as they might do in a student teacher inter-
action, they work as a team supporting each other rather than competing
with each other. The fact that they are united in a collective effort to solve
translation-problems indirectly works towards the improvement of their
communicative and social skills (see González Davies 2004: 13).
Most of the ideals of the educational philosophy are convincing, some
might have cause for concern about how well they will practically fit in
with a curriculum design (Kearns 2006) which has a limited number of
teaching units per course.
184 Chapter 4

The question of situated cognition (Risku 2002) and a ‘translation

project workshop’ where students deal with professional authentic texts
is very attractive and shared by many other translation scholars. Nord
(1988, 1991) through her scholarly work has earned the name of a
strong advocate of the profession-based learner-centred approach in
which training should simulate professional practice with the awareness
of its function and ‘meaningful realistic purpose’ (after Kelly 2005: 12).
Pym talks of “the need to use “authentic” or badly written texts as bases
for discussion and debate, and the desirability of involving students in a
wide range of professional or semi-professional activities” (Pym 2003:
493). Other scholars favour a task-based approach pointing to the fact
that the texts used in the translation classroom have to be well adjusted
to the actual abilities of the trainees. Appeals are made to revise the cur-
ricula to create training programmes which will utilize the process-based
approach and learner-centered approaches (Massey 2005, Kiraly 2005,
Cronin 2005) while some scholars raise an important issue of training
the translator trainers and consider the fact that translation students of
today may have different abilities (Kelly 2005: 43-44). Other debated is-
sues include the question whether translation trainees should be edu-
cated or trained (Kearns 20089), whether they should specialize in par-
ticular ESP areas of expertise or whether the focus should be on flexibil-
ity and adaptability to the demands of the translation market. Kiraly
(2005b) hypothesizes that:

translator competence emerges as the result of the collaborative com-

pletion of authentic translation work, and by observing translators, both
non-professional and professional, in the socially-situated praxis of au-
thentic translation work, we can acquire a privileged view of the nature
of the translation process and glean readily applicable insights into how
to best help students develop their capability to function as professional
translators in the real world outside of the academic ivory tower (Kiraly
2005b).

–––––––––
9
Kearns (2008: 209) discusses the vocational/academic dichotomy in translator
training and concludes that the “reality of curricular renewal in translation education is
messier than is often acknowledged”. While providing educational arguments that uni-
versities rather than vocational training colleges are the right place to educate future
translators Kearns (2008) also draws attention to consider the market demands and pro-
vide translation trainees with transferable skills.
 Developing translation competence 185

Yet, the questions asked by trainees are still the same, what do I need to
learn to be able to translate with facility? How long does it take to de-
velop self-confidence which is needed for fluent competent translating?
Can I provide translation services without formal training? These ques-
tions are still not easy to answer satisfactorily for the trainees who are
frequently extremely self-motivated to work on acquiring their profes-
sional competence, committed and devoted to their self-development as
translators.

4.4. Different routes to translation competence

It remains to be recognized that the human skill to translate may develop

along different evolutionary lines. The importance of training is today un-
disputable, but historically it is a fairly novel development. Institutionalized
translator training programmes do not have a long history and as observed
by Cronin (2005) have not successfully developed solid pedagogical foun-
dations. Compared with the history of translation as a service to humanity,
institutionalized translator training is a modern invention full of frequently
debated challenges (Kelly 2005, Cronin 2005, Płusa 2000, 2007). Respect to
history and countless expert translators who have collectively played a key
role in the transmission of knowledge in what Donald (2001) calls mind-
sharing cultures requires the recognition of the fact that there are many dif-
ferent paths which can lead to the development of expert translators. It fre-
quently comes as a bit of a puzzle that throughout history there have always
been many excellent practicing translators who have never been trained in
translation training institutions. The argument is frequently raised by oppo-
nents of institutionalized translator training who believe that, “Translators
cannot be ‘trained’: they emerge by a natural process” (Gouadec 2007: 253).
On the other hand, complaints are being voiced that translation training in-
stitutions are not really successful in preparing their trainees for the open
market. Opponents of the rigid requirement of formal translator training
claim that the human skill to translate will develop in some individuals who
deliberately engage in the experience of translation and practice translation
without any institutionalized support. To my knowledge there are no clear
data on the proportions between the number of translators with formal quali-
fications and practicing translators who are self-taught at least in those coun-
tries, including Poland and the US, where formal accreditation (licensing) is
not required to provide translation services. To reconcile the different paths
186 Chapter 4

to translation we can only assume that in both cases there is a learning proc-
ess involved, a broad concept of education, or self-education perceived in its
broad anthropological sense as the up-bringing of the mind. In this sense a
translator in the making is then best perceived as an expert learner placed ei-
ther in the context of formal institutionalized training, or reliant on mentor-
ing and self-learning. Robinson (1997: 51) suggests that translation involves
“complex processes of conscious and unconscious learning” and one should
add that it is so provided that one is indeed devoted to learning and develop-
ing the human skill to translate.
In the spirit of Giddens10 one could then say that the human skill to
translate is never static if it is supported by continuing practice coupled
with self-reflection. As such it is then best perceived as a product of a ho-
listic endeavour. Those who want to progress their skills from their natu-
ral ability continuously work towards expertise acquisition and reflect
upon the way they do it. The need for enhanced reflexivity (to use Gid-
dens’ (1991) term) comes first of all from having access to two linguistic
systems which results in natural metalinguistic awareness that language is
purely a communicative tool which can be effectively used in communi-
cation (also in intercultural coomunication via translation). Clearly self-
reflectivity is a feature of expert learners who use it in order to build up
their self-confidence necessary for making decisions on how to solve
translation problems. Self-confidence in turn is a feature of entrepreneu-
rial personalities (Koh 1996) which apart from self-confidence are charac-
terized by self-management, internal locus of control and thus self-
direction, need to achieve (expressed in creative thinking, flexibility,
adaptability) and the propensity for moderate risk taking based on self-
knowledge and the awareness of one’s abilities (Korunka et al. 2003). For
translators as well as for L2 users (see Whyatt 2009a) self-confidence is a
fluid quality and the right amount has to be well balanced so that one’s
self-confidence is truly productive and far from being under- and over-
confident it is based on a sound assessment of one’s capabilities. It is only
then that self-confidence will assist a translator in his or her work in
which informed guesses frequently have to be made relying on one’s in-
tuitive proceduralized skills of language use. In effect translators with

–––––––––
10
Giddens, Anthony is a British sociologist who pointed out to the role of self-
reflectivity both on an individual and on the institutional level in the creation of identity
in post-modern societies.
 Developing translation competence 187

considerable experience become a collection of seemingly contradictory

but in fact complementary attitudes of being confident and humble expert
learners.
An expert learner reflects on the process of learning to evaluate the re-
sults, uses the knowledge of oneself as a learner to become more aware of
effective strategies, and is able to control and monitor the process in order
to achieve a desired aim (Ertmer and Newby 1996: 1). The ideas so much
discussed in educational contexts under the term of learner autonomy might
be essential to self-taught translators who develop their skills by working
out appropriate procedures. In the process of self-critical creative learning
they use reflective thinking, or as put by Schön (1983) reflection-in-action
which relies on a self-questioning approach via which they gain a metacog-
nitive control of their learning process. In institutionalized translator train-
ing programmes this aim is achieved differently, e.g., by guidance, correc-
tive feedback which aims at installing appropriate working habits. In both
cases the interplay between personality features, existing abilities and feed-
back (or self-feedback) is essential for the development of translation skill
to reach the level of expertise. Undoubtedly translation expertise relies con-
tinuously on the reflective professional practice in which translators have to
handle complex and unpredictable translation problems. At the same time
their experience is used to revise their skills and procedures in the process
of endless self-development and self-mastery. Needless to say, there is no
ultimate level of attainment in translation expertise development (Shreve
1997) although there is a tacit agreement among TS scholars that it takes
about 10 years of practice in providing translation services to be considered
an expert professional.
In terms of expertise development research, Hoffman (1997) drawing
from the area of cognitive psychology suggested that it is only after 10
years of professional practice that one becomes a true professional, an ex-
pert in any area of expertise. With reference to interpreters he defines an
expert as “one whose judgments are uncommonly accurate and reliable,
whose performance shows consummate skill and economy of effort, and
who can deal effectively with rare or tough cases, and who has special
skills or knowledge derived from extensive experience with sub-domains”
(Hoffman 1997: 199-200). As Hoffman stresses the act of becoming a
professional is best seen in the developmental progression, as a route (or a
journey) which stretches from a ‘naïve’ (i.e., somebody completely igno-
rant of the domain) who only by making a conscious choice to pursue a
188 Chapter 4

career in translation becomes a novice motivated to acquire all the neces-

sary skills and knowledge. The novice then becomes an apprentice (mean-
ing the one who is learning preferably under the eye of a master or a
trainer, i.e., in a translation training programme). This can take up to 12
years in the traditional craft guilds but, as suggested by Hoffman, can be
equivalent to 8 months to 3 years of translation training programmes.
Then the apprentice becomes a ‘journeyman’ (defined by Hoffman (1997:
199) as a person who can do a day’s work unsupervised but under orders).
Then usually after 5 years as a journeyman one becomes a professional
and an expert. Undoubtedly, it is not only the length of the translation ex-
perience but its quality and intensity that matters.
Another range of factors which are rarely discussed in the context of
professional translation skills development includes affective factors
which most likely stimulate the process of translation expertise develop-
ment. Most translators in the making as well as those with considerable
work experience are extremely self-motivated and dedicated to the devel-
opment and maintenance of their knowledge and skills. They enjoy the
activity and find it self-rewarding (Sofer 1997/2006). I shall return to a
detailed discussion of expertise development and the role of affective fac-
tors in chapter five. For the time being, however let us have a look into
the world of professional translators in order to see the present discussion
on translation competence and expertise against what is expected and re-
quired of translators in today’s world.

4.5. The professional self and translation as a profession

In recent years and possibly parallel with the need to educate translators
in a response to growing market demands, more and more attention has
been devoted to the development of translation expertise. Yet, some
scholars involved in translator training become clear that the schools of
translation and training programmes available are only a starting point, an
initial stage on the route of professional development as translators. As
observed by Płusa (2007: 86) the diplomas are only a kind of a pass to
their future career as professional translators.
A lot that we know about translation as a profession is mostly derived
from theoretical assumptions and from idealized expectations. As ob-
served by Walters (2005), the human factor of the professional self with
all cognitive limitations and psychological constraints is not taken into
 Developing translation competence 189

account and there is a dearth of empirical research into the translation per-
formance of practicing translators for whom translation is a way of mak-
ing a living or supplementing their income earned elsewhere. To quote,

The profession of the translator, with its problems and practical consid-
erations, is sporadically, though seldom systematically, treated in journals
and newsletters. The practice of the profession under pressure, working to
deadlines, satisfying different types of client, using dictionaries of all
kinds, whether monolingual or multilingual, general or technical, hard-
copy or computer-based, by in-house or freelance practitioners, making
use of expert or other informants (often the clients themselves) is occa-
sionally discussed, usually on an anecdotic level (Hickey 1998: 3).

It seems that interpreters have recently attracted more attention as re-

search subjects than written translators (Pöechhacker and Shlesinger
2002). Due to some interest from cognitive psychologists (De Groot
2000) we have come to realize that the human factor means that even pro-
fessional experienced interpreters make mistakes and tend to sacrifice ac-
curacy for the sake of fluency. In written translation the performance of
professional translators has not been extensively studied. Rarely, profes-
sional translators share their ideas about their translation experience in
scholarly accounts with reference to predominant theoretical considera-
tions (see Biguenet and Schulte 1984 for a collection of essays in which
literary translators attempt to reconstruct the translation process). At the
time when Think Aloud Protocols were used to investigate the translation
process professional translators were considered poor subjects for the
method as they were not verbalizing a sufficient amount of verbal data fit
for analyzing their thought processes. Investigating the performance of
written professional translators by inference from the finished product,
that is mostly by the method of error analysis was rare and problematic as
it required assessment against a well defined norm, but the norm was dif-
ficult to define as it depended on other dynamic factors such as the func-
tion and recipients of the TL text (see Toury 1995, Hermans 1991, Pym
1998). More recently any product oriented analyses have been considered
old-fashioned with the focus laid on the process as well as being prob-
lematic without a clearly defined concept of translation quality (see
House 2001 for a review of different approaches to translation quality and
translation quality assessment). Although most translators and translation
scholars know immediately whether the translation of a given text is ade-
190 Chapter 4

quate (not to use the subjective term ‘good’) scholars are unequivocal
about what makes a good translation (see Korzeniowska and Kuhiwczak
1994). As observed by Duff (1981: 1) most discussions of quality in trans-
lation result in criticising faults. To quote, “It is unfortunate that transla-
tion is perhaps more often criticized for its defects than praised for its
merits. The faults of a bad translation are immediately apparent, the vir-
tues of a good one may easily pass unnoticed” (Duff 1981: 1). House
(2001) suggests that the quality of a translation should be considered on
two levels: the primary level of linguistic description based on the knowl-
edge of language and the norms of language use and on the second level
referring to “value judgements, social, interpersonal, ethical questions of
socio-political and socio-psychological relevance, ideological stance or
individual persuasion. Without the first one, the second one is useless, in
other words, to judge is easy, to understand less so” (House 2001: 257).
Yet, an intersubjectively verifiable set of procedures on which translation
quality assessment could be judged is not available and remains another
puzzle in the area of human translation.
Still it is assumed that professional translators are expected to produce
translations of high quality both linguistically flawless and communica-
tively fitting in with socio-pragmatic requirements. As suggested by some
scholars (Venutti 2000) it is best if the translator is in fact invisible in the
finished product (see Jakobsen 1994). The translator is to give voice to the
author of the SL text but should not be heard himself/herself. The question
whether it is at all possible has been raised and attempts have been made to
empower translators (Tymoczko 2007). As it was mentioned in chapter 1,
the demands placed on professional translators are high. The EU directive
specifies the profile of an EMT (European Master’s in Translation) holder
listing the following requirements:

a professional translator (“a high-calibre graduate”) in addition to the

basic professional requirements such as initiative, intellectual curiosity
and motivation should have sufficient language skills (perfect command
of the first language and thorough knowledge of two or more other lan-
guages), thematic knowledge that is familiarity with economics, finan-
cial affairs, legal matters, technical or scientific fields (including exper-
tise in the acquisition of knowledge)

The applicants for the MT should additionally have: good powers of

concentration, an ability to grasp varied and often complex issues
swiftly, a high degree of motivation and intellectual curiosity, an incli-
 Developing translation competence 191

nation to show initiative and imagination, an ability to work consis-

tently and under pressure, both independently and as a member of a
team, a readiness to accept advice and an ability to give feedback, basic
computer skills11.

As it can be gathered from the above list of requirements the demands are
extremely high, not for “a mere mortal” (Walters 2005: 212), not for eve-
rybody (Hejwowski 2004). It can be expected that those who are able to
meet such requirements are already self-selected and not only equipped
with excellent language skills but also with adequate cognitive make-up
to accommodate all the requirements specified in the quotes above.
Yet, professional translation in the sense of translation services avail-
able on the market has many facets and it is not reserved for degree hold-
ers in translation. In many countries, including Poland and the US for ex-
ample, formal accreditation is not needed to provide translation services
on an open market with the exception of sworn translators. In effect all
translations which are not legally binding documents whose translation
has to be certified by a stamp and the signature of a sworn translator can
be translated by anybody who undertakes the job of translating them,
anybody who considers himself/herself a translator (see chapter 1). In
consequence, there is a problem of quality and there are many translations
publicly available which should not see the light of day (see Korzen-
iowska and Kuhiwczak 1994) and which most likely were produced by
natural translators who have not developed translation competence in a
professional sense discussed in this chapter but who use their untrained
ability to translate.
As pointed out by Shreve (1997), professional translation is not
equivalent with advertising translation services or with being a graduate
of a translation training programme but with the ability to produce what
Shreve (1997) calls a ‘constructed translation’ most likely meaning high
quality communicatively and formally flawless translation. The underly-
ing knowledge needed to produce such a translation (TC) cannot be as-
similated in the form of step by step instruction but develops over time
as a result of intensive practice and feedback combined with growing
awareness of the ethics involved in the profession (Gile 2004). The pro-
fessional associations of translators try to safeguard the interest of pro-
–––––––––
11
(http://ec.europa.eu/dgs/translation/programmes/emt/key_documents/emt_sample_c
urriculum_en.pdf)
192 Chapter 4

fessional translators and the trust of clients seeking translation services

by issuing recommendations and setting up standards to ensure high
quality translation.
In the Translator’s charter published by FIT (The International Federa-
tion of Translators) in Section I, General obligations of the translator, it
states:

1. The translator shall possess a sound knowledge of the language from

which he translates and should, in particular, be a master of that into
which he translates;
2. He must likewise have a broad general knowledge and know suffi-
ciently well the subject matter of the translation and refrain from un-
dertaking a translation in a field beyond his competence12.

Without the support of legislation, these recommendations as well as

those issued by associations of professional translators in individual coun-
tries13 which provide care over the ‘right level of translation competence’
(see Płusa 2007: 85) are not always followed. The translation market is a
fast growing business and some suggest (Gouadec 2007) that survival on
the market is the best measure of professionalism. The question ‘who is a
professional translator?’ is then equivalent to the question, ‘How success-
ful is a person in providing translation services?’ The measure of profes-
sionalism is then equivalent to how well can a translator function on the
translation market open to huge translation agencies networking their jobs
internationally and to free-lance translators.
Who is a professional translator? Chriss (2006), the author of Transla-
tion as a profession offers the following definition:

Professional translators are applied linguists whose ability to work with

language, write well, and for freelancers to operate a business, repre-
sents their source of income. Professional translators are people dedi-
cated to their languages and the nations, societies, and cultures which
come with them. They are devoted to their ability to understand their
–––––––––
12
http://entomologia.rediris.es/pub/nj_bscw.cgi/d526459/Carta%20UNESCO.doc
13
In Poland there are two associations which issue recommendations and safeguard
the quality of translation services provided by the members they associate: STP, Sto-
warzyszenie Tłumaczy Polskich [Association of Polish Translators] and TEPIS, Polskie
Towarzystwo Tłumaczy Ekonomicznych, Prawniczych i Sądowych (Polish Society of
Business, Legal and Court Translators/Interpreters).
 Developing translation competence 193

source language and write in their target language. They recognize that
translation is both an art and a craft, and so are committed to deepening
their knowledge of the fields they translate in, and to cultivating greater
facility for writing about such matters (Chriss 2006: 140).

The above quote is very capacious and underlines the combination of the
social and individual aspects involved in the profession understood as a
service to a community in the global and in the local sense. The profes-
sional translator is perceived as an expert fully committed to achieving
and maintaining his/her expertise. This perception fits in well with the de-
velopmental perspective on translation as a human skill discussed in this
work. Yet, some scholars argue that not every professional translator is an
expert translator (Jääskeläinen 2010: 215) and the length of experience
has to be considered together with their job requirements and the quality
of their work (Sirén and Hakkarainen 2002: 75). As observed by
Jääskeläinen (2010: 215) the fact that one works as a professional transla-
tor does not guarantee “consistently superior performance in the domain”,
a key feature of expertise (Ericsson et al. 2006: 3). Consistently as sug-
gested by Sirén and Hakkarainen (2002: 75) a further distinction should
be made between expertise in translation and professionalism.
To sum up, becoming a professional translator means joining the profes-
sional market. More and more books are written to advice those who want
to pursue a career in translation telling them how to set themselves up as
professional translators, how much to charge, how to deal with clients.
Some of the practical guides are written by experienced professionals and
provide invaluable information about the profession which complements
academic training or self-coaching for those who see their career in lan-
guage translation. Sofer’s (1997/2006) The Translator’s Handbook (sixth
edition, US based) Samuelsson-Brown’s (2010) A Practical Guide for
Translators (fifth revised edition, UK based), Chriss’s (2006) Translation as
a Profession (based on a series of previously published articles and US
based) or McKay’s (2006) How to Succeed as a Freelance Translator (also
US based) are a minefield of practical information about the profession, its
demands and benefits, hardships and delights. Sadly, there are also pseudo-
guides to the business of translating, like for example Férailleur-Dumoulin’s
(2009) A Career in Language Translation where professional translation is
devoid of ideology and work ethics and portrayed as a dictionary-based
computerized commodity to be sold on the open market (see a review by
Whyatt 2011). Undoubtedly, the experience of practitioners described by
194 Chapter 4

true masters and experts in the profession is not only precious for novices in
the profession but also for TS scholars and translator trainers. In the spirit of
James Holmes who pointed to this rich source of information and stressed
the need “to “translate” and integrate the everyday speech statements of
practising translators about their translation activity” (Holmes 1988/1994:
96) such guides have empirical value for the study of translation as a human
skill to reach the level of expertise. The message sent by experienced practi-
tioners, who have gone along the evolutionary lines of translation as a hu-
man skill from novices to experts, is not only informative but also inspira-
tional for future practitioners. As put by Sofer
Translation has put me in touch with more areas of human knowledge
and endeavour than almost any other career is capable of doing. At one
time or another, a translator becomes part of almost everything (...). But
best of all a translator never stops learning. Language keeps changing,
knowledge keeps increasing, and the professional translator stays on top
of it all. Once you have developed good translation habits, you will en-
joy the activity of continuous learning of new words and terms, and be-
ing part of the latest advances in many areas of human knowledge. It is,
indeed, a privileged position (Sofer 2006: 151).

Undoubtedly the value of experience allows long serving practitioners to

clearly see the most essential aspects of translation work: its creativity, in-
tellectual benefits of learning new things, its privileges of being involved
in intercultural communication and exchange of knowledge across lan-
guage barriers. Other authors share their experience of practicalities in-
volved in providing translation services. Chriss (2006), for example pro-
vides solid matter-of fact information on taxation, rates, workload, tech-
nology and equipment, accreditation and professional organizations as
well as well-organized information on translator training schools and cer-
tificate programs, and a list of recommended reading, brief but informa-
tive. Real experience-based advice is shared on skill development and
language skills maintenance, on the scarcity of narrow subject dictionar-
ies and the need to rely on parallel reading to find adequate terminology.
To quote from a quick question and answer section:

Q: How do I find good dictionaries?

A: I wish I knew. Most translators, particularly those working in techni-
cal fields, struggle with this problem. Years can pass before new terms
in computers, finance, or what have you appear in print, so translators
 Developing translation competence 195

often rely on parallel reading to find good translation for new terms
(parallel reading is the processing of reading two versions of one text),
sharing their own term list via the Web, and checking with clients and
experts in the subject fields they work in (Chriss 2006: 170).

Despite the great value of practical guides for aspiring translators they more
or less assume that one already is a professional translator and as such miss
a vital link in the developmental continuum of translation as a human skill.
Indeed, not many books are written to shed light on how one becomes a
professional translator and how one can achieve the professional readiness
and the personal self-confidence to enter the translation market.

4.6. Conclusions

The purpose of this chapter was to show the current understanding of

translation as a trained skill in the developmental progression from the
natural untrained ability to translate discussed in chapter 3. Three per-
spectives on professional translation competence defined as the underly-
ing system of knowledge needed to translate were presented and dis-
cussed. The additive perspective views TC as a set of added capacities to
natural translation ability resulting from the bilingual foundations. These
added capacities include metalinguistic awareness, intercultural sensitiv-
ity as well as a range of still unspecified cognitive factors. The compo-
nential perspective sees TC as a set of sub-competencies which include
strategic sub-competence which governs the other four sub-competencies
including bilingual sub-competence, instrumental sub-competence, extra-
linguistic sub-competence and psycho-physiological components. The
third holistic perception of TC sees it as a holistic system of knowledge
both declarative and procedural which allows for fluent translation. The
lack of agreement among TS scholars on how TC should be most ade-
quately defined is counterbalanced by the general agreement that TC is
experience-derived knowledge. From the educational perspective, how-
ever the lack of a generally accepted definition of TC is a major draw-
back. The scant attempts to describe the process of acquiring TC point to
some directions. There is a tacit agreement that TC is developed from the
pre-existing knowledge which according to some scholars (Toury 1995,
Presas 2000) undergoes some transformations as a result of the translation
experience. The changes affect primarily the bilingual knowledge which
has to be re-organized to allow for fluent interlingual lexical access and
196 Chapter 4

retrieval as well as for the efficient control of cross-linguistic interference

especially in the form of negative transfer at the lexical and conceptual
level. There is also a general consensus of opinion that in the process of
learning to translate according to professionally accepted principles nov-
ices develop a set of habits parallel to their growing understanding of
translation as a purposeful socially required activity. Unfortunately, there
is a dearth of empirical data on what working habits are actually acquired,
in what succession and what effects they bear on the quality of transla-
tions produced by trainees.
The PACTE group sees the process of learning to translate to reach
professional expertise as a dynamic learning process in which the pre-
existing knowledge is integrated and re-structured and onto which other
capacities are added. Shreve (1997) sees the development of translation
expertise as an evolutionary space which is in fact limitless as there is al-
ways room for improvement. These perceptions although convincing do
not seem to provide sufficient guidelines for translation pedagogy and
some scholars point to the pedagogical gap in translation teaching meth-
odologies (Cronin 2005, Piotrowska 2007).
Starting in the late 1990s and continuing into the new millennium
publications appeared which pointed to the need to restructure the learn-
ing environment in a way which would provide trainees with the best
translation experience from which they could acquire their professional
translation competence. Process-oriented approaches placed the learner
in the centre of the learning environment putting an end to the old
transmissionist teaching methods. The process-oriented approach in op-
position to the traditional product-oriented approach, which focused on
studying the resemblance the TL text bore to its SL original, viewed the
product as an outcome of the translation process taking place in the
translator’s mind. It became clear that it is the process with its creative
problem solving and decision making performed by the translator that
holds the key to the quality of the final visible product needed and
commissioned by a social need to communicate across language barri-
ers. Scholars interested in translator training (Gile (2004), Kiraly (2005),
Cronin (2005), González Davies (2004), among others) more or less
agree that translation trainees should first become aware of the creative
thought processes (Kussmaul 1991) which are required to translate, and
which are able to accommodate the multitude of factors that need to be
considered in the translation process.
 Developing translation competence 197

Yet, institutionalized translator training is a relatively new concept.

Considering the fact that translators have provided translation services
throughout history it has to be acknowledged that translation as a human
skill can develop from a natural untrained ability and reach the level of
expertise also without institutionalized educational support. The phe-
nomenon of the so-called self-trained translators, who develop their trans-
lation competence and expertise by self-learning, points to the role played
by other factors in the process of translation expertise development.
To sum up, the view of translation as a trained skill which emerges
from this chapter is still far from clear. It suggests that the system of
knowledge needed to translate is difficult to generalize, abstract and con-
vert into an operative set of principles. Translation competence includes
declarative and procedural knowledge and some of the knowledge stores
very subtle data about how languages encode information and how they
are used in social interaction. As suggested by Gile (2004) “translation is
first and foremost a set of cognitive skills, though it also requires knowl-
edge. Such skills, like other cognitive skills, are acquired and mature over
a long period, generally far longer than the few months or years that
translation students spend in the classroom”. Gonzales Davies (2004)
pointed out that in the process of acquiring translation competence the
developmental continuum starts with unconscious incompetence through
conscious incompetence via conscious competence to the stage of uncon-
scious competence14.
The transition from one stage to another is still waiting to be described
and translation pedagogy could benefit from the findings. For the time be-
ing we can only assume that as a result of experiencing the entire cycle of
events and processes involved in producing a translation and supported by
educational feedback or critical self-reflection a new identity is gradually
being granted to the novice translator (c.f. Kiraly’s (1995) self-concept).
The changes are not a result of learning a handful of strategies advising
how to solve translation problems but they are much more holistic and af-
fect knowledge structures at the global cognitive level. Translation as a
profession is a difficult concept to discuss due to the lack of legal regula-
–––––––––
14
The same route can be distinguished in the process of acquiring any other complex
skill including the skill to use another language. Kirsner and Speelman (1998: 6) point out
that during early stages of skill acquisition people are aware of the basis of their expertise
but as they become fluent performers their expertise becomes implicit and they themselves
have lost contact with the computational information underlying their performance.
198 Chapter 4

tions as to who can be regarded as a professional translator. On the one

hand, there are extremely high expectations imposed on professional
translators by scholars, educational institutions and professional associa-
tions, but on the other hand society at large remains unaware of the qual-
ity of translation services for which the market is open to all who claim to
be able to provide them. The development of the professional self in a
true sense of having not only translation competence but also the profes-
sional ethics is still under-researched. Trainees are frequently left without
guidance and stick to the common knowledge that one learns from ex-
perience. But what one precisely learns, in what stages and to what bene-
fits is still subtle knowledge.
In chapter 5 I will present my proposal of viewing the process of
translation skill development to reach the level of expertise as a process
of learning to integrate knowledge into what I termed a Knowledge Inte-
gration Network (KIN) not in a general sense but as a response to a trans-
lation task.
Chapter 5

The key feature of expertise in the performance of

skills is the ability to restructure knowledge
Séguinot (1989b: 80)

Becoming a professional translator:

A proposal of a Knowledge Integration
Network in the development of translation
as a professional skill

In view of the fact that very little is known about the actual process of TC
acquisition I would like to hypothesize that the development of translation
competence to reach the level of expertise is anchored in the translator’s
ability to integrate knowledge for the purpose of translation. Assuming
that creativity is the most distinctive feature of human translation and
since creativity subsumes the ability to integrate knowledge for a specific
purpose it is plausible to expect that translation to reach the level of ex-
pertise requires the activation and integration of all the relevant knowl-
edge. Therefore I would like to propose that expertise in translation relies
on the translator’s ability to temporarily create what I call a Knowledge
Integration Network (KIN) which enables the translator to draw on all the
necessary cognitive resources which are kin to the translation task at
hand. These cognitive resources include all kinds of knowledge that the
translator has at his/her disposal as well as the knowledge which if needed
can become available through external factual research or the use of dic-
tionaries. The ability to build such networks and benefit from them
throughout the translation process develops parallel to the translator’s ex-
perience. I shall expound the operating principles behind a Knowledge In-
tegration Network as well as the assumptions it draws from cognitive and
psycholinguistic theories. Next, I will use the proposal of KIN to account
for the differences in which novice translators and professional translators
approach the task of translating. And finally, I will present the developing
translation skill from the perspective of the translator as an expert learner.
200 Chapter 5

5.1. Creativity in translation

Robinson (1997: 51) suggests that “translation is an intelligent activity, re-

quiring creative problem-solving in novel, textual, social and cultural con-
ditions”. Creativity is the most distinctive feature of human translation (De-
lisle 1988: 37). It was initially more readily associated with the translation
of highly expressive literary texts (Pisarska 1989, Piotrowska 2007) be-
cause its product was a priori creative being a translation of a poem, drama
or any other literary work. However, a close investigation of the translation
process through TAPs which accompanied both individual and group trans-
lation (Kussmaul 1991, 1995) showed that creative thought processes are
ingrained in the translation process in general. This is probably a conse-
quence of the fact that the use of language relies on creative processes.
Creativity in the translation process received a significant amount of at-
tention (Kusmaul 1991, Kussmaul 1995, Mackenzie 1998, Pisarska 1989)
and provides room for combining intuitive and cognitive processing. “Crea-
tivity seems to involve both subconscious production and an awareness of
purpose and aim (Schottländer 1972: 160), both convergent and divergent
thinking, both passion and objectivity (Joerges 1977: 171), both random-
ness and order (Bergström 1988: 30)” (Kussmaul 1991: 93). Usually there
are four phases distinguished in the creative process (which go back to
Poincare (1913): preparation, incubation, illumination and evaluation. De-
pending on the phase the ratio of cognitive to intuitive processing may dif-
fer. In the preparation phase, a problem referring either to SL text compre-
hension, transfer of meaning or re-expression in the TL is spotted and rec-
ognized. Immediately relevant information is gathered and possibly initial
solutions are hypothesized. According to Kussmaul (1991) at this stage
thinking takes place on a cognitive level. In the incubation phase using the
technique of brainstorming thinking is ‘mainly associative and subcon-
scious’ and intuition seems to replace cognition and the process is driven by
“emotion and passion rather than objectivity” (Kussmaul 1991: 94). At this
stage a kind of physical and psychological relaxation seems to be a favour-
able condition. Kussmaul (1991) noted that his protocol subjects sometimes
needed to interrupt the translation process by performing a divergent action
like going to the toilet, getting something to drink or eat or putting a new
piece of music on. Also laughter and a certain amount of fooling around
creates relaxation. When however, the subjects returned to the translation
task they frequently produced “a bright idea” (Kussmaul 1991: 94). It
 Becoming a professional translator … 201

seems that “[a]t a conscious level the mind is at rest, but at the same time
mental activity goes on subconsciously” (Kussmaul 1991: 95). It is impor-
tant that creative processing is fluent and while searching for a solution to a
translation problem, the translator is able “to produce a large number of
thoughts, associations or ideas” in a relatively short space of time. The sub-
conscious processing is demonstrated by pauses in the concurrent TAPs
produced by the subjects. The pause is frequently followed by a ‘bright
idea’ or a fluent change of strategy involving the reorganization of knowl-
edge until the solution is found and the phase of illumination sets in. At this
stage emotions play an important part as if the translator was involved in a
game of skill and was in fact claiming a victory over what posed itself as a
problem, a psychological barrier to move forward in the translation process
and closer towards the intended aim of finishing the task. The phase of il-
lumination is closely related to the next phase of the creative process, the
evaluation when the new ideas are immediately subject to evaluation ac-
cording to how appropriate they are as solutions to the problem recognized
in the preparatory stage. If they are accepted the translator will move on, if
however they are rejected the creative process will start again. Kussmaul
(1991) observes that it is important that all the phases of creative translation
do not follow in a sequence, but are rather interrelated and should be per-
ceived as “moves backward and forward, in loops, as it were, and some
phases are gone through repeatedly” (Kussmaul 1991: 97). Evaluation is
very closely related to illumination and to the preparatory stage where the
specifications for evaluating the creative product of incubation and illumi-
nation are formulated. The assessment of the form accepted as an initial so-
lution with respect to its contextual and situational appropriateness (the
specifications) is in fact a metalinguistic act (see chapter 3) carried out by
the internal monitoring device operating to ensure adequate translation
quality. Kussmaul (1991) observed that skipping the phase of evaluation
may lead to a loss of good ideas produced during the stage of incubation
and illumination. Kussmaul (1991) concludes by way of answering the
question of how to teach creative translation, “The most obvious observa-
tion is that we must take care that our students always preserve a critical
and evaluative attitude toward their ideas that come to their minds. At every
moment of the incubation and illumination phase they must be able to step
back, as it were, and observe what they have been doing” (Kussmaul 1991:
94). There are, however, certain conditions that need to be provided to en-
courage creative translation, such as a positive attitude towards the task,
202 Chapter 5

appropriate level of difficulty, encouraging atmosphere in the class, lack of

harsh criticism, techniques for removing psychological blockages by diver-
gent thinking or providing relaxation, encouraging fluency of thinking and
flexibility as the mental abilities vital for creative thinking. Goleman in his
most recent digital book The brain and emotional intelligence (2011) does
say, “the neural dynamics of creativity tell us why putting an innovative
puzzle aside can be part of the solution”. Kussmaul (1991, 1995) admits
that training creative skills is an ambitious aim but improving such basic
requirements for translation as “the ability to find synonyms or semanti-
cally related words for a given expression” or “the ability of abstracting
meaning from its linguistic form” should have positive effects on the over-
all translation skill. It is also possible that fluency and flexibility might be
trained by other than linguistic tasks. Although Kussmaul (1991) suggests
that in this respect psychologists might have some suggestions, the issue
has rarely been raised in TS literature. Even making students aware of the
creative processes which go on in their translating minds is some achieve-
ment which “may at least help them to guide their own mental activities to
some extent” (Kussmaul 1991: 99).
Creativity as an undisputable aspect of human translation however
should go together with the awareness of the constraints which mark out its
limits. These include the SL text, the linguistic limitations of the TL system,
the communicative expectations of the TL text receivers, and the limita-
tions of the processing mind (Whyatt 2000). Neubert (1997) called this
kind of constrained creativity ‘derived creativity’. To quote:

A translation is a new creation, even though it is guided by its source

text and the requirements of the translation situation. Any experienced
teacher of translation can verify this claim – the same source text as-
signed to a dozen students can yield a dozen, perhaps all equally good,
translations. Each is an expression of the differing experiences, knowl-
edge and creative talents of their respective translators. Yet, they are all,
recognizably, derivable from the source text (Neubert 1997: 17).

It is possible that the awareness of the constraints plays an important role

at the evaluation phase when the potential solutions are assessed, accepted
or rejected1 The evaluation phase of the creative problem-solving is when

–––––––––
1
Neubert (1997: 18) notes that very little is known about the differences between
original creative language use and the derived creativity of translators or interpreters.
 Becoming a professional translator … 203

the translation process meets its product. It is possibly this particular

meeting point that is a vast area where translation pedagogy has a lot to
do to encourage future translators to self-assess, self-reflect on how well
they guided their creative-problem solving, how well they maneuvered
within the space marked out by the constraints. Vital questions have to be
answered here: How well is the translation produced meeting the reader’s
expectations? How well is it respecting the author’s intentions, and what
does it say about the translator himself/herself? As observed by Neubert
(1997), “A translation is not created from nothing; it is woven from a se-
mantic pattern taken from another text, but the threads – the linguistic
forms, structures, syntactic sequences – are new” (Neubert 1997: 17).
Maintaining a proper balance requires a variety of creative strategies in-
cluding transposition (syntactic reordering), modulations (lexical recast-
ing), compensation and a wide range of semantic shifts when a metaphor
is demetaphorized or a generalization is specified (see Neubert 1997: 19).
The creativity ingrained in the human mind and required in language use
in general and in translation in particular is unique and highly individual.
Creativity is sometimes demanding and sometimes frustrating when we
look for the one and only word in our mental lexicons and in all the avail-
able resources. Creativity is a state of mind which is closely linked to
emotions and feelings (see Kussmaul 1991) brought about by the ups and
downs of creative problem solving. In other words, it is the creative think-
ing which combines the physical, intellectual and emotional energies and
capacities of the translator. There is a more or less general consent that
translation is first and foremost an intellectual activity (Gile 1995/2009).
Yet, the other two dimensions, the physical execution of the translation
process either in writing or in speech and the emotional affective factors
which must play a huge role in driving the translation process to the stage
of the actual physical production have not been given just attention. Still,
when translating all three aspects, the physical, the intellectual and the af-
fective processes are involved in a subtle interaction to achieve an in-
tended aim of producing a translation. As suggested by Kussmaul (1991)
they all play a part in enabling creative thinking which is required in hu-
man translation. Although some people might rightly argue that different
texts recquire different amounts of creative thinking, it would be danger-
ous to claim that an instruction manual does not require creative thinking.
Setting off from creativity as the most distinctive feature of human trans-
lation, it is my intention to show that the human ability to translate in or-
204 Chapter 5

der to reach the level of expertise has to involve the ability to integrate
knowledge needed to perform each and every translation task. In brief I
call it the ability to build an adequate Knowledge Integration Network
(KIN) which provides all the necessary cognitive resources to allow for
creative problem solving which then combines the physical, the intellec-
tual and the affective processes.

5.2. Theoretical foundations of KIN

Generally the proposal of KIN is embedded in cognitive studies which al-

low to view one’s linguistic knowledge (here one’s bilingual knowledge) as
a part of general knowledge structures. Some of the knowledge that we
have has been consciously learned and some has been incidentally acquired
(i.e., the knowledge of our native language) throughout our life. Both con-
scious or explicit knowledge such as general knowledge of the world, spe-
cialist knowledge of certain domains, e.g., biology, economy or history and
incidentally acquired knowledge is a collective product of our experience
as physical, intellectual, emotional and social beings. Altogether it forms
our intellectual human capital that we utilize in the form of abilities or
skills with a different degree of success in our everyday life and in our pro-
fessional life. Since the knowledge that we have (including procedural
knowledge needed to execute certain actions) is stored in our memory the
question whether and how we use it is in a way equivalent to the question
of how well we can navigate our long term memory, retrieve and activate
what is needed2. As pointed out by Donald (2007a: 220) what we have
stored in our memory is inactive, just like a DVD or a book in a library. It is
“a dead record” which can be activated by priming, and it will start to live
again only when it is brought back into our awareness and becomes acti-
vated. It is only following this unique resuscitation procedure carried out
within our neural networks and done as neurologists would insist thanks to
the chemical molecules and the brain’s electric circuit (see Dudai 1989, Ei-
chenbaum 2002, Kalat 2007, 2010, Donald 2007a, Squire and Kandel
2009) that we can make use of our knowledge when it is needed to perform
a task at hand. This ability to integrate knowledge and being able to use it

–––––––––
2
As advocated by Hebb (1949) memory storage most likely takes place in distrib-
uted networks, i.e “assemblies of cells, distributed over large areas of cortex, work to-
gether to represent information” (Squire and Kandel 2009: 10).
 Becoming a professional translator … 205

when it is needed according to many scholars is what constitutes human in-

telligence. Needless to say, the concept of intelligence is very complex and
psychologists argue which is the most appropriate framework. Some view
it as general intelligence (the ‘g’ theory) while others view it as a cluster
concept of multiple intelligences (see Waterhouse 2006: 207, Gardner and
Connel 2000: 292). Furthermore, the concept of intelligence has itself re-
cently undergone some significant changes and as a result our understand-
ing of the concept has been expanded to include emotional and social intel-
ligence (Goleman 1995, 2006) which when linked with personality re-
search provides vast grounds for explaining how we perceive reality, inter-
act with others and process incoming information.
As pointed out by Bell (1991: 230), the translation process crucially
depends on our ability to process information from our senses (the physi-
cal domain) when we read a SL text with our eyes or fingers, in case of
the blind, or when we read the gestures of a signing person, in case of the
deaf, and convert what is made available through our senses into mean-
ingful information in our mind. Then when translating we have to carry
the informative content over into a TL system where again with the help
of our senses we type a translation constantly checking with our eyes to
see how well we are doing. In this respect all the issues referring to how
humans process information, organize, store and retrieve knowledge are
essential to the human skill to translate. Wilss (1996) insisted that transla-
tion is a knowledge-based activity. It therefore seems justifiable to sug-
gest that the development of translation as a human skill depends on the
ability to activate and integrate knowledge which is needed for a particu-
lar translation task. In translation however, in contrast to monolingual
tasks, this process of integration will necessarily involve the integration of
two language systems, followed by the conscious and controlled inhibi-
tion of the SL system and activation of the TL system (see the discussion
on language control in chapter 2). It is my intention to suggest that the
proper balance between sufficient inhibition of the SL system and sufficient
activation of the TL system is facilitated by an adequate KIN temporarily
created in the translator’s mind in response to a given translation task.
Although KIN reaches beyond linguistic (or bilingual) knowledge into
our complex cognitive database stored in LTM (Long Term Memory) and
accumulated throughout our lifetime, it seems plausible to expect that it
will produce a priming effect needed for the actual formulation and ar-
ticulation in the form of a TL text. In other words an experienced transla-
206 Chapter 5

tor when reading a SL text for translation will automatically place the text
within his/her knowledge structures and through a ripple effect, or spread-
ing cascaded activation as it is viewed in connectionist models3, some
lexical items will become activated and ready for access when the produc-
tion of the target text starts. A similar kind of activation but spreading bot-
tom up rather than top down was suggested by Seleskovitch and Lederer
(1995) in interpreting practice and termed ‘cognitive complements’. The
authors suggested that on hearing a sound and identifying it as meaningful
words or phrases not only linguistic knowledge and experience with the
item are activated but also other past experiences which have been indi-
vidually associated with the sound. All the activated and integrated
knowledge will participate in making sense by the interpreter (Se-
leskovitch – Lederer 1995: 225). Similar Kiraly (1997) talks about “an in-
tuitive – or relatively uncontrolled – workspace (…) in which various
types of information from a more permanent knowledge store are synthe-
sized with information from source text input and external resources”
(Kiraly 1997: 151). Following this line of reasoning the proposal of KIN
encompasses the entire translation process not only the stage of compre-
hension and stresses the priming effect on the activation of the TL system.
Since, however not all the knowledge that might be needed is avail-
able in our internal memory store, KIN might signal the need to reach into
the external memory storage (books, dictionaries, encyclopedias, libraries,
data bases, the internet). This option is unavailable to interpreters for ob-
vious reasons, for the translators of written texts researching is a part of
everyday practice. Finding out facts that we need to know we simultane-
ously refresh in our LTM or get hold of the words and structures which
are used to express these facts. For example, a translator working on a
translation of a scientific paper entitled, ‘Clogging of deposits in the va-
dose zone under infiltration conditions (Dębina water intake, Poznań)’
without prior factual knowledge of the processes involved will become
aware of the lack of knowledge stored in his/her internal memory which
would suffice to create an adequate KIN and will first search for informa-
tion about the processes described in geology creating in this way a KIN
which will consist of facts and a stock of previously unknown vocabulary.
This process of building KIN most likely takes place every time a transla-

–––––––––
3
Connectionist models view behavioural or mental phenomena as processes emerg-
ing from interconnected networks of simple units (cf. Marcus 2001).
 Becoming a professional translator … 207

tor is faced with a new text. When translating highly specialized texts the
process of creating an adequate KIN will involve intensive research irre-
spective of the direction in which we translate, especially if our factual
knowledge of the field is limited. In simple words the translator will ask
and answer some essential questions: ‘How much do I know about the
subject matter?’, ‘What kind of knowledge is at my disposal?’, ‘Do I have
the TL vocabulary to translate the text or do I need to do some research to
access and integrate all that I need to know to produce an adequate trans-
lation?’ (Stolze 2004: 44). Various scholars implied the knowledge inte-
gration networks which occur even during the first encounter with the SL
text. For example, Holmes (1988) suggested his mind-map theory, Lör-
scher (1992) wrote about expectation structure and Hönig (1991) dis-
cussed the importance of macrostrategy which allows to see the text not
only as a whole but also as a part of a larger background (e.g., field or
domain). The need to integrate all the necessary knowledge for the pur-
pose of translating has been a tacit assumption but not much attention has
been devoted to its effect on the actual linguistic formulation and execu-
tion of the translation process.
The ability to build an adequate KIN which I hypothesize as a crucial
part of the human ability to translate at the level of professional expertise
apart from activating all the knowledge which is kin to the task at hand is
assumed to prime access to the linguistic means needed to produce a par-
ticular translation. In other words, by activating and integrating all that is
needed to understand the SL text, it is expected that the activation and in-
tegration achieved at the translator’s ‘Common Underlying Conceptual
Base’ (Kesckes and Papp 2000, Paradis 2007) will spread down from the
level of conceptualization to the level of the target language infrastructure
activating the links in the bilingual mental lexicon. The process of the TL
text production will then run fluently with lexical access and language
control making sure that only the most adequate matches are chosen and
articulated, i.e., produced. Since every lexical item carries detailed speci-
fications about its syntactic behavior and pragmatic bearing, the process
of writing down a translation can resemble a very sophisticated jigsaw
puzzle. Some pieces, e.g., words which are chosen do not match the se-
mantic pattern and are replaced by others, a piece which is chosen, e.g., a
verb to express the meaning of a noun dictates the route and further lexi-
cal selection and grammatical decisions. Some pieces do not have any
matches, for example in the case of conceptual gaps in the TL system
208 Chapter 5

(Whyatt 2006b) and require compensation strategies. As put by Levý

there is a constant chain of forks in the road and if the translator decides
to choose one direction some choices are limited by the limits of the TL.
The TL text is then in Neubert’s view (1997: 17) woven from the seman-
tic patterns of the TL but with different threads. It is important, however
that the threads are ready for use, within the translator’s reach. It is my in-
tention to hypothesize that this readiness is achieved when an adequate
KIN is created upon reading a SL text for translation and taking in all the
available information about the TL text which has been commissioned by
a client. Fillmore’s (1977) conception of the text as a frame which has to
be unraveled into a meaningful scene is an adequate metaphor for the pro-
posed KIN which goes a bit further and postulates the mental readiness
for re-framing its content in the form of a translation. It is then closer to
Holmes (1988) and his mind-map theory.
By proposing KIN I hypothesize that with experience the human skill
to translate relies on the ability to activate and integrate all the relevant
kinds of knowledge for a particular translation task. The ability to activate
and integrate relevant knowledge develops as a product of translation
practice. The proposal of KIN in the process of translation skill develop-
ment is based on a number of assumptions derived from cognitive psy-
chology, learning theories and memory research which explain the devel-
opmental nature of the translator’s ability to create an adequate KIN for a
task at hand. In translation like in any other action that we perform there is
a multitude of factors which are involved and which will have to be ac-
counted for by the person who performs the action, in our case the transla-
tor. Complex skills such as translation recquire an integration of declarative
and procedural knowledge in a synchronized performance involving our
body (senses) and our mind (the intellect and emotive factors). In other
words, translating requires an efficient working network which will reach
into our long term memory archives while at the same time keeping in con-
scious touch with what is going on in our working memory. To understand
how translators learn to cope with these cognitive demands it is essential to
make some assumptions drawing from various relevant disciplines.

5.2.1. The human mind as a synchronized system

The first assumption is that the human mind works as a synchronized uni-
fied system in order to solve what poses itself as a problem on the way to
 Becoming a professional translator … 209

achieving an intended goal, in our case producing a translation of a SL

text. As pointed out by many TS scholars and proved by the lack of suc-
cess in Machine Translation, translation tasks abound in open ended prob-
lems, and the translator in order to solve them has to draw on all relevant
cognitive resources, and by being creative and flexible has to find a rele-
vant solution. By producing a unique Knowledge Integration Network
(KIN) for each and every translation task the cognitive resources are
ready for use. The literature on translator’s cognitive resources, i.e., on
what translators should know frequently generates extensive lists (Ko-
rzeniowska and Kuhiwczak 1994), but it rarely grants to the translator the
decisive power over which knowledge will need to become activated and
integrated for a particular task. Similarly, there is little research into how
the ability to integrate knowledge changes with the repeated experience of
translating. Cognitive psychology is of help in this respect. As observed
by Donald (2007b: 74), “any learned behaviour patterns involve func-
tional rewiring of the brain, in the sense that functional circuits are cre-
ated that did not exist before”. Consequently, these new functional circuits
can become stored in our LTM and recalled to aid our performance in
similar activities. Whether or not the translator will learn from experience
depends on his/her ability to store and withdraw relevant information, pat-
tern or procedure when it is needed. The synchronized effort which is
needed and the strategic approach to the task of translating might be the
reason why translation has been frequently compared to a game of appro-
priate decisions (Levý 1967, Gorlee 1994).
Comparing translation to a game is by no means a novel idea going
back to the mathematical models of game theory described by Neumann
and Morgenstern (1944). Dinda Gorlee (1994: 67) in her book included a
chapter entitled “Translation and the semiotics of games and decisions”
where she compared translation to games such as jigsaw puzzles and chess.
Other analogies of translation as a game often include a crossword puzzle
(Newmark 1999) and more recently translation as a game of billiards (Ri-
himaki 2005). This comparison is useful when looking at translation as a
purposeful task which requires the translating mind to work as a synchro-
nized unified system to reach the aim of producing a translation. In transla-
tion like in a game of skill (not luck) the translator’s mind is focused on the
task and his/her actions are synchronized and goal oriented. In this respect,
a translator fits into the description of any rational player, who, “…takes
into account the possible consequences of each of the courses of action
210 Chapter 5

open to him [and] is aware of a certain preference order among the conse-
quences and accordingly chooses the course of action which, in his estima-
tion, is likely to lead to the most preferred consequence” (Rapoport 1961:
107-108 quoted after Gorlee 1994: 72). Since however many problems are
novel and unique like in the case of metaphors or neologisms which need to
be translated problem solving will involve also an element of risk-taking
just like in a game of skill. The success of the game [translation] depends
on the knowledge and skills which are needed to perform the task and made
available in the translator’s mind via the process of building a relevant
KIN, which also allows to calculate risk-taking (i.e. translating Polish
pierogi as ravioli or leaving it as proud Polish pierogi on an English version
of a restaurant menu). How much knowledge can be made available in the
translating mind is not only a product of the translator’s experience (less
experienced translators might translate pierogi as ravioli focusing on find-
ing any equivalent similarity known in the TL conceptual repertoire while
more experienced translators will leave the domestic word and inform
about its unique conceptual value). The amount of knowledge which can be
made available for the translating mind working as a synchronized unified
system is also subject to the cognitive constraints of the human mind, our
second assumption behind the proposal of KIN.

5.2.2. Limited capacity of the human mind

“The human mind is a limited capacity processor” (McLaughlin and Here-

dia 1996: 213) which means that our conscious cognitive capacity is in a
limited supply and once exhausted it will have a negative consequence on
our performance. This principle was used by Gile (1995: 152) who divided
the translation process into four components (efforts): comprehension, pro-
duction, memory and coordination and concluded that an imbalance in the
distribution of mental energy between the four components will tax the
translator’s performance. While agreeing that this is likely the case, espe-
cially with inexperienced translators, it has to be acknowledged that each of
Gile’s efforts is in itself a complex process which relies on the retrieval and
integration of information stored in the human memory. As Whyatt (2010:
89) suggested the study of mental effort management in the process of
translation performed by inexperienced translators points to their major
problems with cognitive control over the translation process itself. Due to
their limited experience in translating, inexperienced translators expense a
 Becoming a professional translator … 211

substantial amount of their cognitive attentional resources for conscious

control and coordination of the translation process. This is quite typical of
anybody learning to perform a new complex skill. With practice, however,
including all the stages of the performed action such as planning, execution
and rehearsal (Donald 2006) certain parts of the complex skill due to the
mind’s tendency to divide labour become automatized, do not require con-
scious control and free some attentional resources. By way of analogy, let
us think of reading. “When we learn to read our native language, for exam-
ple we initially move haltingly from word to word, but after practice we
read quickly, moving the eyes to a new location about four times a second
and taking the meaning from more than 300 words in one minute” (Squire
and Kandel 2009: 202). Like in other cognitive skills a learning individual
develops “a feel for how to do the task that is there are actually no facts
memorized about the task but a general sense or intuition about how to pro-
ceed” (Squire and Kandel 2009: 203).
However, Donald (2007b) stresses that for certain actions to became
automatic practice has to be combined with conscious rehearsal, awareness
and integration of what in the skill is new to the learner with what is known
to the learner. It is only then that automaticity will occur as a ‘by-product’
of practice. The interplay between controlled, conscious, cognitive and
automatic, intuitive processing has frequently been acknowledged in trans-
lation studies (Hönig 1991, Kussmaul 1991, Mackenzie 1998, Cronin 2005,
Kiraly 1997). For example, Toury (1995: 251-252) suggested that

[i]n extreme cases, they [translators – BW] may in fact develop automa-
tized ways of handling specific problems, even a series of fixed solu-
tions which are mobilized whenever a certain problem occurs. To be
sure, this is often an efficient way of evading a problem rather than
plunging headlong into it, thus leaving more time and reserving more
energy for less proceduralized parts of the task. Such shortcuts seem to
form an important part of a translator’s acquired ability to cope with
problems in real life situations, involving, e.g., time pressure, growing
fatigue, incomplete knowledge, and much more (Toury 1995: 251-252).

These ‘mental shortcuts’ suggested by Toury (1995) allow the translator

to work within the limited capacity for conscious control (Jääskeläinen &
Tirkkonen-Condit 1991: 89, Jääskeläinen 1999). The ratio between con-
sciously controlled processing and automatized procedures is described
and explained by cognitive learning theories.
212 Chapter 5

Anderson (1985) views the learning process as “the movement from

controlled to automatic processing via practice (repeated activation)”
(quoted after Mitchell and Myles 1998: 86). Robinson (1997) admits that
“experienced translators are fast because they have translated so much that
it often seems as if their brain isn’t doing the translating – their fingers are...
the target language equivalent terms come to them automatically, without
conscious thought or logical analysis” (Robinson 1997: 49). This view
would perhaps explain Paradis’s (2009) claim that experienced interpreters
bypass the conceptual level whereas the less experienced ones interpret via
conceptual mediation. A point needs to be raised, however whether the
speed of interpreting is in fact synonymous with automaticity and whether
the fact that equivalents can be recalled as-if automatically excludes the op-
tion that a conceptual match is nevertheless checked and approved (see
chapter 2 on lexical access).
It seems that the ratio between consciously controlled processing and
automatic processing performed within the limited capacity for conscious
control in the human translating mind is determined by the translator’s
ability to create an adequate KIN which will decide about the mental ef-
fort management throughout the translation process (see Séguinot 2000,
Jakobsen 2005). Some processes will be carried out by automatized pro-
cedures (i.e., when translating into L1) and others will be carried out at
the level of consciousness, e.g., self-monitoring for adequate meaning
transfer). It is only through automatizing certain subtasks that a complex
skill such as translating or interpreting (Chmiel 2006: 49) can be devel-
oped to the level of expertise (Wilss 1996: 102). As observed by Donald
(2007b: 75) when we operate in the conscious mode we cannot tolerate
distractions and we usually trade off accuracy and speed in tasks which
exceed our capacity especially when we have to multi-task like in transla-
tion. This is true of novice translators, speed and accuracy suffer and the
task is time consuming and tiring. When the same task is performed in an
automatic or even semi-automatic mode we can tolerate disturbances and
distractions. Yet, as noted by some scholars automaticity does not neces-
sarily mean that experts always perform the same task faster than non-
experts. Jääskeläinen (2010: 220) notes that some researchers (Sirén and
Hakkarainen 2002) have made the observation that translation in fact does
not become easier with experience which might seem to contradict com-
mon expectations. Similar results were reported in a creative writing task
study by Bereiter and Scardamalia (1993: x) in which “given the same as-
 Becoming a professional translator … 213

signment, experts would work harder and do a great deal of thinking”

(quoted after Jääskeläinen (2010: 220). It is possible that the greater effort
in ‘thinking’ (building a KIN) has not resulted in a faster performance but
most likely in a different performance and most probably more likely to
result in a higher quality product.
Many translation scholars who talk about the ratio between cognitive
control and automaticity in the translation performance of expert translators
mention the role of intuition. Hönig (1990: 157) calls it knowledge that we
are not aware of and Robinson (1997) talks about gut feeling. Squire and
Kandel (2009: 203) say, “Much of what we call “intuition” is probably
learned and is based on nondeclarative memory which stores procedural-
ized automatic action programmes”4. It has to be added, however that even
automated actions are goal oriented and therefore they have to be moni-
tored in awareness to make sure that they serve the common purpose, in our
case translation. In this respect a broad concept of consciousness is needed.
Different action programmes or subtasks will require a different amount of
attention. Donald (2007b) talks about the attentional hierarchies of different
semi-automatic actions. Consciousness itself according to Donald (2007b)
is “not a unitary phenomenon, a matter of all or none. Consciousness is a
graded hierarchy of up-and-running routines. There is a “vivid core” of
sensory awareness and various other elements and routines, running con-
currently under meta-cognitive supervision. At any given moment the
meta-cognitive observer can suddenly broaden the reach of awareness or
focus it on some particular detail” (Donald 2007b: 79). This brings us to the
third assumption underlying the proposal of KIN, conscious and deliberate
meta-cognitive control.

5.2.3. Meta-cognitive supervision and goal orientedness

Translation is a conscious purposeful activity, a service provided for

somebody else than the translator, being it an individual client, an institu-
tion, or a community and as such it is generally expected to be a product
of skilled performance. In complex skills, such as translation certain ac-
–––––––––
4
Chmiel (2006: 49) mentions that interpreters sometimes report on ‘great revela-
tions’, brilliant ideas on how to render a pun and explains after Mac Leod (1998: 59) that
even seemingly unattended stimuli may undergo extensive perceptual processing and
trigger activation at the conceptual level possibly priming access to a “revelation-like
rendition of a pun”.
214 Chapter 5

tions can be performed in consciously controlled and in automatic modes

(Toury 1995, Hönig 1991) but all the goal oriented effort is constantly
monitored in awareness5 to assess its contribution, spot errors and correct
them with respect to their compatibility with the KIN performed in the
translator’s mind who has read the SL text. Neubert (1997) explains this
need for monitoring the process comparing it to monolingual language use:

Translators and interpreters, like editors, are quite aware of not express-
ing themselves when they practice their craft. When searching for TL
means of expression, it is true, they exploit their own repertoire as well
as a variety of resources, both bi- and monolingual. But instead of just
monitoring what they are saying or writing as straightforward L1 or L2
speakers would do, they are continually matching what they are about
to say or write or actually do say or write with what they have heard or
read before in the source text. There is a continual return to the source
text in actuality, and in memory (the trace in memory is all the inter-
preter may have) (Neubert (1997: 18-19).

The proposed KIN suggests that also in written translation the memory has
been activated and the resources integrated into an active KIN to provide
means for an adequate goal oriented monitoring and meta-cognitive supervi-
sion of the translation process. The SL text in its physical presence is avail-
able to refresh the network and keep it active as well as to supply with new
previously unnoticed information which will revise the dynamic KIN and
readjust its weightings. In the entire process complex control mechanisms
operate to ensure first and foremost that the two languages are activated but
at the same time kept clearly apart to avoid interference. As stressed by
Green (see chapter two) language control uses cognitive resources.
Drawing on the ideas of Donald (2001, 2007a, 2007b) on complex
skills, the proposal of KIN assumes that conscious and deliberate meta-
cognitive control is an essential part of the human ability to translate at the
level of expertise. It includes anticipatory planning when the translator is
preparing to translate, meta-cognitive monitoring of self-action when the
translation is in the making and finally, rehearsing when the translator is
improving the final product (TL text) and performing self-critical transla-
–––––––––
5
Wilss (1997: 89) talks about translator awareness, “If we want to encompass ade-
quately the wide-ranging field of human translation, it is necessary to include in transla-
tion studies (TS) the concept of translator awareness (or translator consciousness, for
that matter)”.
 Becoming a professional translator … 215

tion quality assessment, revising and editing the final product. In other
words, whether we plan, execute or rehearse conscious self-regulation is
constantly switched on “high up in the system from which it can intervene
and spot problems at a lower level” (Donald 2007b: 75). Donald (2007b)
gives an example of the piano player who while giving a concert will con-
centrate on his/her interpretation, not on the sequence of melodies or finger
movements but he/she will nevertheless notice if suddenly something hap-
pens to one of the fingers which will affect the overall performance. Need-
less to say, the same piano player years ago had to acquire the complex skill
by practicing the hands and consciously controlling the movements on the
keyboard. The novice and the expert translator share a similar history be-
fore they can make the most of their ability to integrate knowledge and
produce a professional excellent translation. Their expertise, however
would not have developed without their conscious effort to integrate differ-
ent elements of their actions, some of which will through practice and ex-
tensive rehearsal become “automated action-programmes and eventually
make up a coherent and seamless patterns of complex behaviour, such as
driving a car or dancing” (Donald 2007b: 75), or translating.
It is possible that in the development of translation as a human skill it
is precisely the notion of the internal ‘meta-cognitive observer’ which
needs more attention from TS. The awareness of this meta-cognitve fac-
ulty has been voiced by some scholars including Bell (1991) Toury
(1995)6, Gile (1995), Hansen (2003), Kiraly (1995). The meta-cognitive
supervision over the translation activity or as put by Holz-Mäntäri (1984)
the whole translatorial action will rely on KIN produced for a particular
translation task and as such remains an essential part of the human ability
to translate at the level of expertise. This ability is not given but develops
over time with repeated experience of translation and a sizable amount of
reflective thinking.
To sum up, the proposal of KIN, that is the ability to integrate knowl-
edge for translation assumes that the human skill to translate relies heavily
on meta-cognitive capacities to self-organize the translator’s knowledge
base in response to a novel translation task. It takes a holistic approach to
stress the synchronized human effort needed to translate, it respects the at-

–––––––––
6
Toury (1995: 248) talks about ‘an internal control mechanism which develops in-
side the translator first in response to the environmental feedback and later through the
translator’s effort to ‘consider potential responses’ to the translation as a final product.
216 Chapter 5

tentional limitations of the human mind with the subsequent need to auto-
matize certain actions and make use of external memory resources to ac-
cess and acquire knowledge that is unavailable to the translator at the time
when it is needed. KIN stresses the meta-cognitive supervision conducted
within a broad concept of consciousness. It is the translator’s ability to in-
tegrate knowledge for every specific translation task which establishes the
essential link between the translator’s LTM and WM (Working Memory)
which “takes in” the SL text to be translated. KIN relies on a common prin-
ciple in human social cognition according to which what we encounter as
new is in our effort to make sense integrated into a whole range of familiar
old scenarios (Donald 2007a: 220, Bartlett 1932). To perform complex
tasks humans have the capacity for the temporal integration of knowledge
both declarative and procedural including all the required sub-skills learned
or acquired so far (see Bell 1991: 249). This temporary integration “allows
the mind to oversee short-term events and episodes from a deeper back-
ground vantage point, while bracketing the fast moving events in the fore-
ground, and placing them in an accurate context”7 (Donald 2007a: 220).
For the translator the fast moving new event is the SL text which is
opening gradually as the eyes run through the text, simultaneously and pos-
sibly without conscious control various experiences are awakened in the
translator’s LTM8. The ‘awakening’ takes place in both episodic and se-
mantic memory with old schemas, scripts being activated like past stories
we have heard or experienced ourselves (Fillmore 1977, Sperber and Wil-
son 1986, Schank 1986, Squire and Kandel 2009). In this respect the trans-
lator is just a reader of the SL text but as studies suggest (Whyatt 2003,
Stolze 2004) the awareness of the purpose to translate the text bears conse-
quences on which knowledge domains will be activated and integrated into
a dynamically created KIN for translation. In translation a task-specific
KIN offers the means for the conscious meta-cognitive control which will
provide global supervision over the translation process executed to result in
a TL text as a product. The question which knowledge domains will be-
come activated and integrated for the translation to run smoothly with ade-
quate division between consciously controlled actions and automatic or
–––––––––
7
in psycholinguistic terms this means top-down processing (see Bell 1991: 249).
8
Wilss (1999: 220) spoke of lack of memory research in Translation Studies and
noted, “Our main intellectual asset is our long term memory: it is a link between percep-
tion and reflection, and makes translation/interpreting processes of varying complexity,
under varying conditions (text type, client’s instructions, etc.) possible”.
 Becoming a professional translator … 217

semi-automatic routines is an indicator of the stage on the developmental

continuum of translation as a human skill.

5.3. Developmental nature of the ability to integrate knowledge

What exactly has to happen for a novice who struggles to transfer mean-
ing across language barriers to become a confident self-fulfilled expert
translator still remains in many aspects unclear. Are there any stages in
the evolution of the professional skill to translate? What is the average
time that is needed to develop self-confidence as a translator? Which as-
pects of the job are the most difficult to understand for trainee translators?
Which aspects of the expertise are the most difficult to share with nov-
ices? Does translation get easier with experience9? Translation process
studies despite the criticism they frequently have received about their
methods (see Tirkkonen-Condit and Jääskeläinen 2000, Jääskeläinen
2010) have revealed some aspects of how novice and professional transla-
tors translate. It therefore can be assumed that comparing what we know
about the similarities and differences between novices and professionals
will inform us, or at least will provide some insight into, what stretches
between the two stages on the developmental continuum. In other words
the aim is to see whether hard work will bear ‘beautiful fruit’
(Jääskeläinen 1996).
By proposing KIN in the development of translation expertise I hypothe-
size that what divides the novices from the professionals is their ability to
integrate knowledge. The ability to produce a KIN for a specific translation
task is responsive to experience in the sense that the translator’s ability to
create KINs and benefit from the integrated knowledge network develops
over time and in correlation with practice and self-reflection. It is a slow
process. As a result it can be hypothesized that novice translators have seri-
ous problems with building an adequate KIN. They frequently integrate only
the most obvious knowledge structures, the physically visible and the physi-
cally needed bilingual knowledge, acting on the common misconception
that it is sufficient to translate. Alternatively, it is possible that learning to in-
tegrate one’s bilingual knowledge for the purpose of using it for translation

–––––––––
9
Although some scholars suggest that “translation does not become easier with
growing experience and expertise (Sirén and Hakkarainen 2002: 71) it is not specified
what easier means, does it mean faster or less complex.
218 Chapter 5

is the first stage in the process of the evolution of translation as a human

skill. In the very early stages (see chapter 2 on L2 learners as translators) of
experiencing translation L2 learners in fact integrate only the lexical knowl-
edge of their L2 and look for word-for-word semantic correspondences, fre-
quently disregarding pragmatic functional aspects. It is possible that a con-
tributing factor is the lack of conceptual proficiency in the Common Under-
lying Conceptual Base (Kesckes and Papp 2000, 2007) which has to account
for cross-cultural relationships of similarity, difference or partial resem-
blance. Yet with exposure to translation tasks, L2 learners and users fairly
quickly learn a holistic approach to a SL text and their TL counterpart which
has to function in a TL communicative reality (see Gile 2004). With experi-
ence and appropriate corrective feedback (or self-critical reflection), e.g., as
translation trainees they learn to build complete networks of knowledge,
first under the external guidance of their teachers/trainers (or under their
own supervision if they are themselves expert learners). Later when they
start to provide translation services in their professional life they rely on
their own meta-cognitive supervision. To explain in detail the developmen-
tal nature of the ability to build and use KINs in translation let us look at
how inexperienced and professional translators differ in their approach to
the task of translating a text and how differently they handle the translation
process leading to the physically present product (Jääskeläinen 2010, Nor-
berg 2003, Schmidt 2005).

5.3.1. Approaching the SL text

Translation process studies have provided substantial evidence that stu-

dent translators approach the SL text which they are about to translate in a
different way than professional experienced translators (Kussmaul 1995,
Asadi and Séguinot 2005). First of all, novices focus on lexical processing
taking ‘word’ as a translation unit (Lörscher 1991, Hansen, 2002) and
they tend to rely on semantic meaning disregarding the role of context
and pragmatic situational sense. Whyatt (2003) showed that some inexpe-
rienced translators tend to read aloud or whisper to themselves adding in
this way another modality to enhance comprehension. Beginning transla-
tors sometimes start translating without reading the entire text prior to the
onset of translation. Some like to underline unknown words or immedi-
ately use a dictionary. Their focus is at the micro-level of the text and they
rarely cross sentence boundaries when interpreting the meaning of the SL
 Becoming a professional translator … 219

text items (Whyatt 2000). They initially show very low text awareness,
disregard pragmatic markers to identify the type of discourse the SL text
belongs to. Jonasson (1998) showed that novice translators ignore the
macro-level information referring to cohesion, coherence, purpose and
function of the SL text (and the TL text) as well as its text type and dis-
course markers to indicate the level of formality (style). In other words
they process a SL text in a horizontal fashion (see De Groot 1997) with
excessive bottom up processing rather than integrating it with top-down
processing (Kussmaul 1995). The integration of top-down processing
with bottom-up processing is assumed in the proposal of KIN and in
many metaphors capturing the essence of interpretative processes in trans-
lation including the mind-map theory of Holmes (1988), Levy’s text-as-a-
picture-of-reality (1967) or Pym’s (1992) text-as-an-object-to-be-
interpreted constructs. The integrative principle is also present in different
terms ascribed to approaching the SL text for translation such as ‘posi-
tioning the text’ (Stolze 2004) or ‘the stage of orientation’ (Jakobsen
2002, Buchweitz and Alves 2006).
In contrast, when reading for translation professional translators focus
on the macro-level first. As pointed out by the TAP research (Jonasson
1998, Bernardini 1999) their approach is functional and in their text
analysis prior to the onset of translation they first pick up the macro-level
information. Nord (1988/1991) suggested that it includes the following
chain of questions: Who is to transmit to whom, what for, by what me-
dium, where, when, why, a text with what function? On what subject mat-
ter is he/she to say what (what not) in which order, using which non-
verbal elements, in which words, in what kind of sentences, in which
tone, to what effect? (Nord 1991: 144). In Holmes’ (1988) words the pro-
fessional translator is from the very first encounter with the SL text creat-
ing a mental map, a macro-strategy (Hönig 1991) or an expectation struc-
ture (Lörscher 1991) which will guide him/her on the way to the TL text.
Taking the above arguments into account it can be tentatively con-
cluded that the difference in the way inexperienced translators and experi-
enced professionals approach the SL text shows the underlying difference
in their ability to build and use an adequate KIN which will affect the dis-
tribution of attentional resources during the first reading. The novices
seem to integrate only their bilingual knowledge and more precisely their
linguistic knowledge of both languages which is indicated by the fact that
they spot unfamiliar words, underline them or look them up in a diction-
220 Chapter 5

ary. They proceed in a linear form oriented way and their focus on form
overrides making sense (see Lörscher 1992). The professionals activate
their memory and integrate the new SL text within their entire knowledge
base (see Sperber and Wilson 1986, Whyatt 2003). In other words they
see it as a whole. To quote: “The holistic approach to texts for a translator
means that when positioning the text the translator instead of or prior to
“text analysis” will ask: What are the knowledge elements at my dis-
posal? Am I able to understand the message? From which country, which
social group does it come? What do I know about that culture?” (Stolze
2004: 44). One could add more questions that a professional translator
asks and answers. The reading is goal-oriented and the translator with ex-
perience establishes how the SL text fits in with what he/she knows (can
access from his/her LTM) and what knowledge will have to be acquired to
complete the KIN needed to translate. All the macro-level questions
(quoted above from Nord 1991) come in a kind of dialogue with the trans-
lator’s knowledge base (see Whyatt 2003). This interaction shows that a
temporary KIN is in the making which integrates the new SL text with the
translator’s cognitive repertoire. Seleskovitch (1978) and her idea of
deverbalization10 or the use of visualizations described by some scholars
(Kussmaul 200511) can be taken as metaphors describing the integrating
principle of a KIN activated in the translating mind.
The novice for whom translation is a fairly novel experience will lack
the ability to activate all the relevant knowledge domains and will focus on
the most obvious integration of linguistic knowledge of the SL and the TL.
This is also the case possibly due to the fact that the bilingual knowledge in
itself needs all the cognitive attention from novice translators who in most
cases are L2 learners rather than proficient L2 users (see Cook 1992).
When reading a SL text in their L2 all or most of their cognitive capacity
will be taken by working out the meaning. If the SL text is in their native
language the comprehension effort (see Gile 1995) is reduced (at least in
texts not requiring any domain specific knowledge) but problems occur at
the stage of meaning transfer which will still remain horizontal (see Kuss-
–––––––––
10
However, apart from hypothesizing deverbalization the idea how the linguistic
messages are processed into deverbalized carriers of meaning and what the meaning car-
riers are was not explained by Seleskovitch (1978).
11
Kussmaul (2005) having analyzed dialogue TAPs and making use of the cognitive
notions of scenes and frames and prototypes arrived at the conclusion that “visualizing de-
tails of a scene helps translators to arrive at creative translations” (Kussmaul 2005: 378).
 Becoming a professional translator … 221

maul 1995 on how inexperienced translators suppress their creative solu-

tions and opt for the application of systemic rules). It seems that language
control (see Green 1993) engages all the cognitive resources with no capac-
ity left for self-monitoring at the conceptual level.
With practice and corrective feedback, adequate exercises in text
analysis translation trainees gradually acquire the ability to process a SL
text not only as a whole but also as a part of the SL reality. To quote
Stolze (2004: 44), “[i]nitially one will only have an individual under-
standing of the text and will need further research, before being able to
present the message responsibly and comprehensibly. Later on, the trans-
lator will have acquired more experience and will be able to infer the
relevant knowledge right from the beginning”. This statement suggests
that translators with a modest amount of experience will consciously try
to build a relevant KIN (cf. the stage of conscious competence by Gon-
zález Davies 2004) and the explicit process will frequently be time-
consuming. With growing experience there is a transformation from ex-
plicitly created KINs to more implicit knowledge integration processes
which are faster and run under meta-cognitive supervision which does not
require huge attentional resources (cf. Donald 2007b and his hierarchical
perception of consciousness). However, in unknown knowledge domains
even the most experienced professionals will have to invest a substantial
amount of time to create an adequate KIN for the task at hand. Some pro-
fessional translators tend to complain that those who seek translation ser-
vices do not take this part of the translation work into account as they pay
per translated page or per hour of interpreting services12.
To sum up, the novice’s inability to awaken all the background knowl-
edge results in the difference in the distribution of attentional resources
during SL text comprehension which will have a direct effect on the sub-
sequent process of meaning transfer. This is not to say that the mental
processes of professionals and novices are totally dissimilar (Bell 1991)
since they both attempt to transfer meaning from a SL text into a TL text
and thus perform interlingual operations. The differences, at least those
described through TAP research, occur “in the distribution and frequency
in the types of strategies, i.e. the quantitative aspects of the translation
strategies” (Lörscher 1992: 208) as well as in the control both novices and

–––––––––
12
(Anna Riitta Vourikoski, a member of the Finnish Association of Translators and
Interpreters – private communication).
222 Chapter 5

professionals have over the translation process due to the different KINs
which are created in their translating minds (Whyatt 2010). Possibly the
underlying cause of the difference between Knowledge Integration Net-
works (KINs) created by inexperienced (i.e., L2 learners/users), novices
(translation trainees) and experienced translators (practicing profession-
als) reflect the fundamental differences in their conception of meaning.

5.3.2. Approaching meaning

As I discussed in chapter 3 and partially in chapter 2 student translators

have a very simplified perception of meaning. This is most probably due to
the fact that their conceptual proficiency is quite low and in the process of
development. Textual and contextual meaning is frequently ignored for the
sake of lexical meaning (Taylor 1990, Dodds 1999, Whyatt 2009a, 2009b).
Hence, beginning translators ignore the contextual cues and process mean-
ing on a micro-level, employing micro-strategies to solve the problems they
encounter (Hönig 1991). They rarely understand that comprehension is
knowledge of the language + extralinguistic knowledge (Gile 1993: 69), as
well as that in understanding language one is constantly involved in a dy-
namic on-line interpretation and construction of meaning (Fauconier 1998,
Whyatt 2007b). Their focus on form and on horizontal processing fails to
acknowledge that a text has its inner and outer context (Gorlee 1994) and
understanding will sometimes require some knowledge of facts (political,
historical, geographical, scientific, or other) rather than adherence to se-
mantics. Their ignorance comes from the lack of translation experience ac-
companied by corrective feedback which would make them aware of the
fact that in language everything conspires to convey meaning (Lakoff 1982,
Wierzbicka 1991). As observed by some TS scholars, the conviction that to
know a word is to know its form and meaning has frequently led many
translation students astray (Kussmaul 1995, Rogers 1996, Whyatt 2008). It
seems plausible that for novice translators words are treated as meaning
units whereas for experienced translators words are in fact entries of differ-
ent kinds of information and which aspects of the information are accessed
is task and context specific (Seguinot 1997: 115).
Professional translators have a much more comprehensive multilay-
ered perception of meaning. They approach the meaning in a functionally
vertical way. They are aware of the pre-existing knowledge they have and
the discourse-induced knowledge (Gile 1993: 71) they activate when they
 Becoming a professional translator … 223

are reading the SL text for translation, and for which they integrate all
their relevant knowledge into an intricate KIN. They are also aware that
in their efforts distributed between meaning comprehension, transfer and
reformulation they may have to acquire some more knowledge, “addi-
tional information, both linguistic (generally terminological, and some-
times stylistic) and extra-linguistic” (Gile 1993: 71). In this respect trans-
lation is a knowledge-based activity for professionals (Wilss 1996)
whereas for student translators (L2 learners/users and translation trainees
who begin to translate) translation is a language-based activity in the nar-
row linguistic sense.
To sum up, professional translators approach meaning from a wider
vantage point of their KIN at the level of their underlying conceptual base
where the semantic knowledge is always subject to higher order thinking.
Notions such as contextual appropriateness, effect on the reader, text type
are primary to out-of-context semantic L1 – L2 correspondences. Nov-
ices, still very much uncertain about their language performance feel safer
sticking to formal accuracy and semantic equivalence. They access their
mental lexicon via lexical rather than conceptual links which frequently
leads to mistranslation or translationese. As discussed in chapter two their
verbal fluency frequently exceeds their conceptual proficiency. Conse-
quently, when transferring meaning without a solid Common Underlying
Conceptual Base (Kesckes and Papp 2000, Kesckes 2007) on which inte-
grating knowledge into a KIN could rely is very difficult and the risk of
producing errors is very high. This fundamental difference in perceiving
form – meaning relationships points to novices’ inability to build a KIN
which would include any other domains apart from their linguistic knowl-
edge since even pragmatic knowledge seems to be missing. The differ-
ence in KINs which are created by novices and professionals in terms of
approaching meaning is also visible in the clear difference in using exter-
nal memory resources in the translation process.

5.3.3. Approaching external resources to aid translation

Student translators show a tendency to over rely on dictionaries which they

treat as “the repository of a linguistic community’s knowledge of words”
(Roberts 1997: 2), sometimes to the point that most of the entire time de-
voted to translating a text is primarily used to consult dictionaries (Varan-
tola 1998: 179). Whyatt (2006a) showed that student translators used up to
224 Chapter 5

50% of their translation time to consult dictionaries in a fairly simple trans-

lation task from English (their foreign language) into Polish (their native
language). Atkins and Varantola (1998) and Mackintosh (1998) are among
very few scholars who studied the use of dictionaries by inexperienced
translators. They reported that most of dictionary look-ups took place dur-
ing the first draft translation and not as recommended by Roberts (1992:
60) during the post-draft revision. Atkins and Varantola (1998) found a
clear preference for bilingual over monolingual dictionaries. For L2 to L1
translation the disproportion reached 72% of all the look-ups in bilingual
dictionaries as compared to only 28% in monolingual ones. In L1 to L2
translation the frequency of use for monolingual dictionaries increased to
37% of all the look-ups performed suggesting that student translators use
monolingual dictionaries to understand the SL text (to decode meaning)
rather than to encode meaning. Another observation reported by Atkins and
Varantola (1998) and Mackintosh (1998) referred to the lack of self-
confidence in making decisions which resulted in the need to find reassur-
ance in dictionaries referring not only to the semantic information but also
to the grammatical information about syntactic behaviour of words (e.g.,
collocation, countability, etc.) and discourse indicators (style, tone). Obvi-
ously, there is a problem of communicative confidence (Whyatt 2009a) in
L2 use. However, Whyatt (2006a) reported that students overuse dictionar-
ies as a source of expert knowledge about words or as an authority to dispel
their doubts only initially. With growing experience they learn not to expect
miracles from dictionaries (Varantola 1998: 181), they start to mistrust dic-
tionaries and develop a more assertive attitude to their own choices or
hunches. In other words, they start to understand that it is their skills as dic-
tionary users which will “determine the ultimate success or failure of the
dictionary search” (Varantola 1998: 184). Apart form the time consuming
consequence of frequent dictionary consultations, it seems that they are in
fact disruptive to the translation process as such which is starting and stop-
ping all the time. It is possible that for inexperienced translators using dic-
tionaries is in fact a distraction in the translation process, which as a novel
task they try to consciously control. As discussed above controlled-
processing uses cognitive capacity which is in limited supply. Distractions
are disruptive to consciously controlled processes whereas they are not dis-
ruptive to processes which have been automatized. As stressed by Donald
(2007b: 75) we usually trade off accuracy and speed in tasks which exceed
our capacity for conscious control especially when we have to multi-task
 Becoming a professional translator … 225

like in translation. This is true of novice translators, despite their efforts,

speed and accuracy, as well as communicative quality are frequently com-
promised. When the same task is performed in an automatic or semi-
automatic mode we can tolerate disturbances13. It is then possible that pro-
fessional translators do not find consulting dictionaries equally disruptive to
the flow of the translation process.
Although there is a dearth of empirical research on how professional
translators use reference material, some suggestions can be inferred from
translation process studies (Hansen 2002, Dimitrova 2005). Generally, it
can be expected that professional translators rely first and foremost on
their internal bilingual mental lexicon which with their experience in
translating must have been reorganized to allow for speedy and efficient
cross-linguistic lexical access (Whyatt 2006b). Although there is no em-
pirical evidence it seems justifiable to predict that their bilingual mental
lexicons develop a new functional architecture which is interlingual rather
than bilingual (Presas 2000, Paradis 2009). Chmiel (2004) discussing in-
terpreters suggested that the effect of constantly tapping into the LTM
store of vocabulary becomes inscribed in the form of tags explained as
“individually developed microstrategies which also activate additional
semantic and episodic memory traces” (Chmiel 2004: 111). These tags are
then used as a system of warnings (e.g., watch out for potential false
friend) or incentives (it’s fine, a well tested recipe). As it can be expected
the mental lexicons of experienced translators are not only functionally
reorganized but they contain significantly more lexical items with more de-
tailed information about their grammatical behaviour (morpho-syntactic in-
formation gathered in lemmas) and contextual/situational (socio-pragmatic)
rules. Experienced translators know that most of this information is not
available in bilingual dictionaries.
Other features of the professionals’ approach to dictionaries are the
consequences of their dynamic approach to meaning discussed above. In
contrast to novice translators they are not focused on finding exact seman-
tic equivalents, a TL noun for a SL noun, a TL verb for a SL verb since
they do not mentally process information in a local but in a more global
fashion. Vertical processing allows for a wide use of compensation strate-
gies (Piotrowska 2002) and does not lead the translator into a ‘word trap’.
Novice translators frequently become locked in one lexical item which
–––––––––
13
Donald gives an example of driving a car. Anderson used the same parallel.
226 Chapter 5

they struggle to find consulting numerous dictionaries instead of reformu-

lating the content and expressing the intended sense in different SL means
(e.g., by paraphrasing a sentence). Experienced translators see the text as
a whole, integrate all the relevant knowledge domains and via their
Knowledge Integration Network (KIN) they prime access to their mental
lexicon (see chapter 2 on mental lexicon and lexical access). Since they
have accessed it on many occasions alongside the cross-linguistc route
they reap the benefits of well paved connections which have left memory
traces (Squire and Kandel 2009, Paradis 2009). For novices these connec-
tions sometimes do not exist and even if they access the needed lexical
item they lack self-confidence whether or not it is the best one in this par-
ticular case so they have to find confirmation in dictionaries. Experienced
translators have more confidence in making choices. If the memory inhib-
its retrieval due to fatigue or a temporary block (Rose 1991) they use dic-
tionaries to trigger access. It can be tentatively assumed that professionals
use dictionaries only if they have to, that is if a word is for them an empty
frame (Kussmaul 1995) and understanding its meaning from context is
impossible or risky. Even in such cases they usually do not turn to a dic-
tionary during their first reading, but only in the subsequent stages of the
translation process when more detailed comprehension is needed. They
use bilingual dictionaries “to add nuance to meanings already established
in their minds or to stimulate the search for solutions” (Kelly 2005: 16).
In their look-ups their search is narrowed down by the pragmatic factors
which they have already established. Professionals are more likely to re-
search data bases for factual information which they need to solve a trans-
lation problem (Atkins and Varantola 1998). They also know from experi-
ence that they will not find non-dictionary type of information in diction-
aries whereas student translators hope they will. Experienced translators
also do not reach for dictionaries (which provide context free information
about word use) if they need information referring to longer stretches of
text than a lexical item to solve a context-dependent problem (Varantola
1998: 184). What is more they may sometimes consciously delay using a
dictionary not to slow down the translation process. Some scholars re-
ported that professional translators are likely to accept partial solutions
which they mark as temporary and later they return to these stretches of
the text during the subsequent revision of the draft translation (Dimitrova
2005, Roberts 1992). Although the question of tools at the translator’s
disposal deserves much more attention in view of what is now available
 Becoming a professional translator … 227

thanks to information technologies (translation memory systems, CAT

tools) let us sum up the differences between novices and professionals
and assess them with respect to the proposal of KIN.
To sum up the above observations it can be assumed that professionals
reap the benefits of their ability to integrate knowledge (KIN) also when it
comes to using external memory tools such as dictionaries. Their search is
frequently substantially narrowed down and therefore faster whereas nov-
ices look for words that they need without the contextual constraints which
are in their mind unavailable since they have not integrated all the vital
kinds of knowledge (an incomplete KIN). It has to be remembered how-
ever, that the professionals’ experience in integrating knowledge, especially
their bilingual knowledge base, will make them more confident in terms of
lexical choice as well as will mean that their bilingual mental lexicons will
be significantly larger than those of inexperienced translators. It will also
have a different functional architecture (interlingual links). Consequently,
not only the manner (using lexical and conceptual links) but also the speed
of interlingual lexical access will be aided by their experience. Similarly, it
is to be expected that with professional experience in translating the use of
both language systems has been repeatedly practiced to result in the semi-
automatic or automatic application of grammar structures and a possibly
fairly rich assembly of interlingual ready-made word combinations used in
familiar contexts. All this will lower the number of dictionary look-ups and
save time as the translator becomes not only a more competent L2 user but
also a more self-confident L1 and L2 user. The proposal of KIN stresses the
importance of integrating all kinds of knowledge kin (related to like in a
family) to the task at hand and predicts a priming effect on access to the
formal linguistic means, e.g., words, structures, discourse patterns, etc. Fi-
nally, the time consuming effect of frequent dictionary use by inexperi-
enced translators lends support to the possibility that for novices dictionary
searches are disruptive to the flow of the translation process whereas pro-
fessionals, who have automatized many aspects of the translation process,
can tolerate these disruptions with no detrimental effect on the flow and
fluency of the translation process. This difference could be explained as a
stimulating effect of an adequate KIN produced by professionals and as an
inhibiting effect of an incomplete KIN characteristic of inexperienced
translators. Further differences in the distribution of attentional resources
during the translation process are visible in how novices and professionals
monitor their on-line performance.
228 Chapter 5

5.3.4. Approaching the translation process: self-monitoring and meta-

cognitive supervision

Translation process studies, especially TAP research has shown that trans-
lators, both novices and professionals monitor their processing, but there
are quantitative and qualitative differences in how they apply their meta-
cognitive supervision over their performance.
Student translators tend to process information in a linear fashion. They
tend to stay on the micro-level and rarely move further ahead or revise the
stretches translated so far to pick up more contextual clues (Lörscher 1992,
Gerloff 1986, Hansen 2002). They rarely reflect on the macro-level infor-
mation, including the function and addressees of the TL text they create
(Kussmaul and Tirkkonen-Condit 1995, Whyatt 2000, 2007a). They show a
high degree of uncertainty when choosing one of the competing TL equiva-
lents for a processed SL item (Whyatt 2000, Whyatt 2009a). As evidenced
by Lörscher (1991, 1992) his protocol subjects who as foreign language
students had a limited experience in translation tended “not to check those
TL utterances, which they have translated and within which they did not re-
alize there was any problem, according to their sense. As a consequence,
the translations of the students more often than not reveal utterances which
contain grammatical errors, even in their mother tongue, violations of TL
text production norms, or which make no sense” (Lörscher 1992: 209).
Whyatt (2010) pointed out that in novice translators most of the mental re-
sources are spent on ensuring the progression of the translation process to-
wards completing the task apparently with no processing capacity left to
monitor the emerging outcome in its entirety. The monitoring reported in
TAPs is mostly devoted to achieving semantic equivalence (Gerloff 1986,
Whyatt 2007b). Most or almost all novice translators do not provide on-line
revision of their emerging translations from their broadly integrated knowl-
edge network (KIN). As they write down their translations, they rarely
question their stylistic quality and focus on correcting local surface errors.
The metalinguistic statements show a lot of concern for semantic accuracy
at the expense of pragmatic appropriateness (Whyatt 2007a, 2009b).
With growing experience the amount of conscious control over the
translation process to ensure good communicative quality increases. Krings
(1986a) investigated 8 advanced translation students and reported that
monitoring strategies constitute an important part of the translation process.
“A characteristic feature of the subjects’ translational procedure was the al-
 Becoming a professional translator … 229

ternating occurrence of retrieval and monitoring strategies. After having re-

trieved a potential equivalent the subjects normally switched over to moni-
toring” (Krings 1986a: 272). Sometimes although in the minority these
strategies resemble self-monitoring observed in free writing and oral pro-
duction and refer to the correctness of a retrieved L2 item rather than its
appropriateness as an equivalent. In the vast majority of cases, however,
monitoring used what Krings calls ‘spot-the-difference strategy’ when as-
sessing the appropriateness of L2 equivalents for particular SL items. The
subjects compared the SL and the TL items looking for differences in terms
of meaning, connotation, style and use. As soon as they found a difference
they rejected the potential equivalent and undertook a new retrieval at-
tempt. Krings noted that monitoring showed that “the learner’s intuitions
concerning specific L2 items play an important role in the monitoring of
the potential equivalents” (Krings 1986a: 272). However, Kussmaul (1991)
pointed out that conscious monitoring can in fact inhibit or reject more in-
tuitive and frequently contextually more appropriate solutions. Clearly
however, with more experience in translation advanced translation students
self-monitor their performance and show signs of metalinguistic awareness
(Buchweitz and Alves 2006). Malakoff and Hakuta (1991) pointed to the
importance of monitoring translation performance which they observed in
natural translators as a sign of metalinguistic awareness, an important factor
in the process of translation skill development (see chapter three).
Professional translators, who when translating (or writing down their
translation) have performed an adequate KIN, seem to monitor their
emerging texts from a global rather than local perspective. As we assumed
earlier through their professional experience with interlingual transfer
professional translators are better at controlling the way they use their L1
and L2 and they tend to keep the two language systems safely apart to
avoid interference (Odlin 1989, Kesckes and Papp 2000, Jarvis and Pav-
lenko 2008). Their language control (Green 1993) and metalinguistic
awareness (Bialystok 2001, Malakoff and Hakuta 1991) is a part of their
meta-cognitive supervision which is not absorbing their consciousness as
much as in novice translators. Consequently, professionals will differently
distribute their mental resources to monitor the emerging product and the
monitoring will not be as time and energy consuming as it is for inexperi-
enced translators. Donald’s metaphor (Donald 2007b: 75) of the piano
player who while focused on the entire interpretation will nevertheless
notice (in his/her awareness) any disturbances at the lower level is very
230 Chapter 5

useful here. Having easier lexical access due to their memory traces, more
confidence about the meaning of words they use they are more likely to
monitor their translations for style and pragmatic appropriateness. In their
meta-cognitive supervision they will make a conscious effort to objectify
their communicative results which is rare among inexperienced transla-
tors who frequently cannot distance themselves from their translations
(see Lörscher 1992: 209).
Acknowledging the existing idiosyncrasy and variability in self-
monitoring and providing on-line revision (Séguinot 1997), self-
monitoring by professional translators shows an interplay between con-
trolled cognitive problem solving and intuitive processing which is a
commonly acknowledged feature of the translation process (see Hönig
1991, Kussmaul 1995). Monitoring during the translation process shows
that both levels of language processing can be subject to the supervising
meta-cognitive awareness. Human beings can run many different auto-
matic tasks keeping them at the same time under conscious supervision
(Donald 2007b: 75). Monitoring the emerging product is probably as
automatic for professional translators as monitoring our speech when we
speak our native language (see Donald 2007b: 75).
Séguinot (1989a: 5) is among not many TS scholars who point out that
self-monitoring is a part of the professional translation process.

The translation process is not a step-by-step linear progression. When we

translate we are actually performing a number of tasks at the same time.
We monitor our output and tend to correct mechanical errors as they oc-
cur. We do not search for words one at a time, wait until the search is
successful, then search for a new word. The psycholinguistic research
suggests that the unconscious operations involved in producing language
can simultaneously pursue different options (Séguinot 1989a: 5).

This parallel processing, however, is only available to those with substan-

tial experience in translating and who have developed automatic or semi-
automatic routines that can be used in the multi-tasking without using up
cognitive capacity in excessive amounts. It is available to those who have
learned to produce an adequate KIN and use it for their meta-cognitive su-
pervision in parallel processing at the lexical and conceptual level irrespec-
tive of the language. However as observed by Malakoff and Hakuta (1991)
the basic condition is the routine and confident use of L1 and L2, including
its grammar and lexis combined with metalinguistic awareness of linguistic
 Becoming a professional translator … 231

relativity as an expression of an intercultural/interlingual relationship be-

tween the SL and the TL. If the use of L1 and L2 is non-routine (i.e., as
metalinguistic knowledge rather than communicative competence) but con-
sciously controlled (for example during L2 lexical access or in the applica-
tion of grammatical rules in beginner/intermediate L2 learners) cognitive
resources are used for language control. Novice translators for example, try
to exercise cognitive control even over their L1, which frequently leads to
memory blocks like in the case of a female student translator trying to con-
sciously recall which verb the Polish word zysk [profit] collocates with
(Whyatt 2010) without success and to her own annoyance.
In this respect the fact that novice translators or student translators pro-
duced more self-monitoring verbal data than professional translators in
TAPs can be interpreted in terms of the qualitative difference in the way
they monitor their translation process. While for the inexperienced ones
monitoring is conscious and cognitively controlled for professionals moni-
toring might be an automatized routine run under conscious supervision.
Donald gives the example of monitoring our speech in an automatic mode
when we do not have to think about the act of speaking itself, choosing
words and checking for meaning. To quote, “When we are speaking our
highly automatized native tongue, we can do many things at the same time
as we speak, provided that we have automatic routines for doing them”
(Donald 2007: 75). This kind of multitasking is only possible if we have
automatized some of the tasks into routines. It can, therefore, be tentatively
concluded that professional translators routinely self-monitor their per-
formance and perform on-line editing because they apply meta-cognitive
supervision prepared by an adequately integrated knowledge network
(KIN). Novices tend to integrate knowledge as they self-monitor their per-
formance and as a consequence their knowledge integration is only partial
(i.e., frequently only metalinguistic knowledge and linguistic competence).
To conclude, it seems that inexperienced translators monitor their
emerging translation by an application of rules (i.e., grammar) whereas
professional translators use their intuitions based on their KIN derived
from their experience (Séguinot 1989b: 80). The importance of self-
monitoring and self-reflection is frequently stressed by translation train-
ers echoing Toury’s (1995) call for fostering in an aspiring translator an
“internal kind of monitoring mechanism, which can operate on the (in-
terim) product as well as on the act of translation as such” (Toury 1995:
250). Kiraly (1995) points to the need to encourage the development of
232 Chapter 5

a functioning translation monitor through error analysis. It should be

admitted however, that the empirical data on self-monitoring among pro-
fessional translators is limited although it must have a significant impact
on the quality of the final product. Similarly, there is not a lot of re-
search into the revision stage of the translation process which aims to
ensure the high quality of the TL text, although it seems to be the care
for the formal quality of the translation that is tacitly assumed to divide
natural and professional translation. The differences in how novices and
professionals revise their translations are quite revealing with respect to
the proposal of KIN.

5.3.5. Approaching revision

“We all have had the experience of being able to revise other people’s work
more easily than our own” (Séguinot 1989b: 79). Hansen (2008) addressed
this tendency in a paper entitled, “The speck in your brother’s eye – the
beam in your own: Quality management in translation and revision”. Over-
all however, it has to be admitted that the revision activity which can be de-
fined as an “execution of an overall text improvement plan that incorpo-
rates a series of local, functional corrections” (Roussey et al. 1990: 54), is
an under-researched area in translation process studies. The scant data
available allows one to assume that also in this aspect there is a substantial
difference between novices and professionals. It is likely that the different
perspective on revision is due to the differences in how novices and profes-
sionals integrate knowledge into temporary networks (KIN) built for the
purpose of translating a SL text into its TL representative.
First and foremost, inexperienced translators, for example language
learners, rarely revise their translations. Their processing effort seems to
be focused on finishing the task. Jakobsen (2002) studied four profes-
sional translators and four translation students and noted that there is a
difference in the distribution of time devoted to drafting the translation
and post-drafting revision. Professionals seemed to have spent less time
drafting their translation but more time revising it in the post-drafting
stage although they introduced fewer changes than the less experienced
translation students. This finding was consistent with Dimitrova’s (2005:
106) study which focused on post-drafting revision and reported that in-
experienced translators (translation students) introduced a lot of changes
in the post-drafting revision of their translations. Although there were
 Becoming a professional translator … 233

only 4 professional translators in the subjects studied by Dimitrova, the

author concluded that the more experienced the translator the fewer
changes he/she introduces in the post-drafting revision. Jakobsen (2002)
observed that both professionals and inexperienced translators did more
revising when they were translating into their L2, although as pointed
out by Superczyński (2009: 24) it was not clear if more revising meant
more time devoted to revision or more changes introduced when revis-
ing the TL text.
Roussey et al. (1990: 53) reported that there is a difference between
professional translators and less experienced translators not only in terms
of the amount of time devoted to revision but also in terms of the strate-
gies used in revising the TL text. Inexperienced translators tend to focus
on local problems or surface errors and they tend to read through the text
in search for mistakes applying what Roussey et al. (1990) calls ‘local-
then-global strategy’. Professional translators in contrast to those less ex-
perienced in the task treat revision as an important part of the translation
process (Mossop 2001). They are more likely to balance the ‘local-then-
global strategy’ with ‘global-then-local strategy’ or as suggested by
Roussey et al. (1990: 54) they use a ‘simultaneous strategy’ and revise the
TL text paying attention to all the linguistic levels of the text including
primarily the macro-structural level. Asadi and Seguinot (2005) studied
nine professional translators and reported on the different attitudes to-
wards revision. Some translators seemed to have created their translation
in their mind first and made no or very few changes in the revision stage.
Some seemed to “translate by revising” (see Mossop 2007) introducing a
lot of changes in their on-line drafting of the translation and very few
changes during the post-drafting revision as if they immediately revised
their translation while typing it down and only very few changes were in-
troduced during post-drafting. Lauffer (2002) who observed three transla-
tors using a combination of observational methods including Translog14,
Camtasia15 as well as TAPs noted that
–––––––––
14
Translog is a key logging programme which records all the keyboard activity dur-
ing the translation process. It is used to study the translation process (see Jakobsen
1999).
15
Camtasia called a ‘screen spy’ is a computer programme which allows one to re-
cord everything which is opened on the computer screen during the translation process.
It is used to study how translators search for information in external memory sources
(Lauffer 2002).
234 Chapter 5

revision reoccurred many times throughout the translation process. Re-

vision strategies included reading and rereading the text, comparing the
source and target texts for accuracy and idiomatic language, verifying
and changing lexical choices, adjusting grammatical structure changing
word order, revising syntax and improving the overall flow of the text.
Revision was done both immediately as the translators typed, and later
when reviewing the sentence, paragraph and then the complete text
(Lauffer 2002).

From the proposed KIN, it seems that the manner in which translators carry
out the self-revision of the finished translation product shows their differing
ability to create an adequate KIN for a particular task. Novices hardly ever
revise their TL texts. Translators with very little experience spend more
time revising but focus on local problems without the global purpose ori-
ented meta-cognitive supervision which relies on their knowledge integra-
tion network. Novices find it difficult to distance themselves from the SL
text while revising their translations, their focus on form brings about more
care for local accuracy in terms of formal features of the TL words and
phrases equivalent in meaning to their SL counterparts. Professional trans-
lators have a global rather than local attitude (i.e., due to their KIN). Some
professional translators leave the TL text aside after the first draft and re-
vise it some time later when they have gained some distance which will al-
low them to revise the translation with the fresh eye of an objective reviser.
This tendency is significant. In a way, it might suggest that by leaving their
translation aside, they feel the need to re-frame from the knowledge inte-
gration network they have been using so far to produce the draft translation.
In the final stages of preparing the TL text to enter the TL reality, which
will happen as soon as it is sent by the translator to the client who commis-
sioned it for a specific purpose, the KIN has to be readjusted to make sure
that the translation is a viable part of the TL reality.
With not much data available, the concept of re-framing in order to
carry out a successful self-revision is only a hypothetical, though a plau-
sible assumption confirmed by the problems novices have in terms of cut-
ting themselves off from the SL text and revising the new TL text as a text
in its own right. If revision marks a definite point of departure from the
SL text, it requires re-framing with focus on the target language and atten-
tiveness to all levels of language use (language as knowledge and lan-
guage as social practice, see chapter 1).
 Becoming a professional translator … 235

We must also remember that in professional practice, translators apart

from self-revision frequently rely on cooperating with other professionals,
revisers, proof-readers and editors who ideally collectively ensure the
quality of the final translation. In the less ideal world of freelance transla-
tion including others is frequently a cost which is cut back on for eco-
nomic reasons (see chapter 6). Including others might be the best way but
it is still the translator who has to develop the skill of self-revision to
eradicate as many errors or inadequacies as possible before approaching
others as revisers. In this respect, a practicing translator is a one-man
band and it becomes even more regular to maintain a self-critical ap-
proach to one’s own translations.
Hansen (2008: 274) suggested that revision competence is closely re-
lated to translation competence but it is partly different. In effect good
translators are not necessarily good revisers and therefore training is
needed to develop the skill of revision. In her own words, “the necessary
presuppositions of revision, which are attentiveness to pragmatic, linguis-
tic, stylistic phenomena and errors, the ability to abstract or distance one-
self from one’s own and others’ previous formulations, fairness, and ex-
plaining and arguing – these can be trained at universities, in separate
Masters’ courses on revision” (Hansen 2008: 275). Whether revision is
treated as a separate skill or not, it is an important part of the translation
process16 in professional translation (Breedveld 2002). Accordingly the
ability to self-revise one’s translation is a part of the human skill to trans-
late to reach professional standards.
To sum up, considering the above arguments the ability to produce a
KIN for a specific translation task is developmental in nature and re-
flected at every stage of the translation process from approaching the SL
text to revising the first draft translation. From the comparison of differ-
ences in how novice translators and professionals approach various stages
of the translation process it is possible to conclude that what divides them
is their differing ability to activate, integrate and apply knowledge of all
kinds. Novices seem to be at the beginning of this evolutionary journey
and make a conscious effort to integrate only the knowledge of the two
language systems which in itself is difficult (see chapter 2) especially
when viewed from the communicative competence perspective. Accord-

–––––––––
16
In traditional divisions of the translation process revision is assumed to be a part
of TL text production.
236 Chapter 5

ing to the cognitive learning principles the ability to activate and integrate
knowledge proceeds from the conscious or deliberate effort to semi-
automatic and automatic application. However, as stressed by Donald
(2007b) automaticity is a product of practice and conscious rehearsal and
as such is always open to further modification. In translation skill devel-
opment from the stage of natural translation ability towards professional
expertise, the role of awareness and the rehearsal of skills is essential. The
role of self-involvement in the refinement of skills past the natural ability
to translate shared by all bilinguals points to the importance of conscious-
ness in the learning process. As observed by Donald (2007b: 78), “there is
very little evidence of complex skill learning outside of consciousness”.
Since it is widely assumed that translators learn from experience but it is
infrequently pointed out what they actually learn, the proposal of KIN
makes a prediction that through experience combined with corrective
feedback and/or self-reflection translators learn to activate, integrate and
apply all the knowledge relevant for (kin to) a particular translation task.
Most importantly, however, the parameters set by an adequate KIN pro-
vide the grounds for meta-cognitive supervision and on-line quality con-
trol during the drafting of the first version of the translation and at the
stage of the post-draft revision. Finally, re-framing the KIN is needed to
ensure that the TL text is not only a pragmatic match of the SL text but
also a viable and communicatively comprehensible part of the TL reality.
The discussion of the different approaches that inexperienced and ex-
perienced translators show at all the discussed stages of the translation
process points to the local (inexperienced and novices) versus global (pro-
fessionals and experts) perspective. It therefore seems that the ability to
build KINs allows translators to see the SL text from their background van-
tage point and supervise how and to what effect the contents of the SL text
is transported into a TL text which then becomes a part of the TL reality
when it is used by members of the TL community, printed, published or
made otherwise available to TL users. The process of creating an adequate
KIN is dynamic in nature and starts with the first encounter with the SL text
which needs translating. As pointed out by Séguinot (1989a: 6) even in
comprehending the SL text “we are constantly making predictions about
what is ahead. We do this on the bases of educated guesses. The education
comes in several forms: experiences stored in the form of scripts, scenarios,
frames, schema, i.e., patterns, which include knowledge about types of
texts, language patterning and content information about the way the world
 Becoming a professional translator … 237

operates”. The ability to integrate knowledge is possibly what differentiates

a novice and a professional. More research into how this ability develops
parallel to translation experience can allow for a more fine-grained distinc-
tion between a novice, a trainee a professional who begins his/her career, an
experienced professional (Hoffman’s 1997), and a translation expert, as
well as a specialist.17
It seems that the ability to integrate knowledge is a slow global proc-
ess which requires not only experience but conscious self-learning from
the experience of translating. As such the proposal of the evolving ability
to integrate knowledge in the process of translation skill development has
another consequence. It assumes that translators who consciously and in-
tentionally develop their human skill to translate are not only applied lin-
guists and intercultural mediators but first of all expert learners who learn
to reorganize and integrate knowledge.

5.4. The translator as an expert learner

Many assumptions have been made about the translator’s knowledge and
expertise but less frequently translators have been seen as learners them-
selves as if this perspective was a threat to their professional competence.
Similar scant attention has been paid to the fact that the experience of trans-
lation has an accelerating effect on the bilingual knowledge base of the de-
veloping translator. Translators by having to handle both working lan-
guages in the translation process gradually acquire a wide range of vocabu-
lary exceeding the average size of the mental lexicon needed for general
communication. The pragmatic knowledge, sensitivity to discourse patterns
governing the choice of register and style and the awareness of linguistic
relativity that translators gradually accumulate parallel to the growing range
of their translation experience has rarely been acknowledged in the discus-
sion of the development of translation as a human skill. By hypothesizing
the concept of KIN and suggesting that the development of expertise in
translation relies on the translator’s growing ability to integrate knowledge
I see the translator who chooses to become a professional as an expert
learner. As indicated in chapter 4 this perspective allows to account for the

–––––––––
17
See Jääskeläinen (2010: 216) for more detailed distinctions in expertise research, i.e.
professionals, routine and adaptive experts and specialists following Hatano and Inagaki
(1992).
238 Chapter 5

development of translation expertise supported by institutionalized training

and/or reliant on self-learning and self-coaching. The development of trans-
lation expertise is most likely a slow process. Below I present two perspec-
tives on the translator as an expert learner: the translator as a Systems Intel-
ligent person and the translator as a self-confident expert.

5.4.1. The translator as a Systems Intelligent person

Peter Newmark (1969: 85) said, “any old fool can learn a foreign language
but it takes an intelligent person to become a translator”. When I shared this
quote with my translation students they found it very encouraging, not to
say flattering. The statement superficial as it might seem provides some
food for thought. Translators through their work participate in many inter-
related systems like language (L1 and L2), culture (L1 and L2) and respec-
tive societies. Throughout their professional life they learn to function
within these systems and interact with them and they might even have the
power to influence them and introduce change (Saarinen & Hämäläinen
2007). In a way they become Systems Intelligent persons.
Although much of Systems Intelligence philosophy is devoted to soci-
ology and grounds itself on the ancient premises according to which the
ultimate goal of human behaviour is the betterment of life, it also offers a
feasible framework to view the gradual development of translation as a
human skill compatible with the framework offered by KIN proposed in
this chapter. In many ways the translator as an expert learner learns to act
within many interconnected multifaceted systems including: the two
working languages (in themselves interconnected knowledge systems as I
discussed in chapter two), the two respective cognitive cultural networks
of L1 and L2 (Donald 2006), the author of the SL text, the client who
commissioned the translation driven by a purpose, the viable text typo-
logical norms, and any other systems including the translator as a cogni-
tive system with imposed limitations discussed in the four assumptions on
which KIN rests. In the process of interacting with the many systems
which are intrinsic to translation activity the translator as an expert learner
has to develop his/her Systems Intelligence. To quote,

By Systems Intelligence (SI) we mean intelligent behaviour in the

context of complex systems involving interaction and feedback. A
subject acting with systems intelligence engages successfully and
productively with the holistic feedback mechanisms of her environ-
 Becoming a professional translator … 239

ment. She perceives herself as part of a whole, the influence of the

whole upon herself as well as her own influence upon the whole. By
observing her own interdependence in the feedback intensive envi-
ronment, she is able to act intelligently (Saarinen and Hämäläinen
2007: 51).

The assumptions of Systems Intelligence include several ideas which are

present in the translator’s behaviour:

– whole is more important than parts

– ‘part’ and ‘whole’ are relative abstractions that are always subject to
redefinition by changing the perspective
– Systems approach begins when you start to perceive the world through
the eyes of another person
– Systems approach looks beyond isolated linear cause-and-effect
chains for interconnections and interrelations

The relevance of Systems Intelligence to the development of translation

as a human skill hinges on the primacy of the whole, multifaceted inter-
connections and interdependencies that have to be accounted for by the
translator. Essentially it means that the translator as an expert learner
develops what can be called ‘intelligence-in-action’, a kind of under-
standing which “will be judged by its practical outcome and manifesta-
tion in the conduct” (Saarinen & Hämäläinen 2007: 56), the actual trans-
lating performance and its final product.
As such this expertise clearly relates to what Aristotle called practi-
cal rather than theoretical reason and is therefore more focused on
‘knowledge how’ rather than on ‘knowledge that’. Consequently, it ex-
ceeds the strictly cognitive perspective according to which we con-
sciously apply learned patterns and allows for integrating all possible
capacities which include instinctual, intuitive, tacit, subconscious, un-
conscious and inarticulate aspects that cannot be accounted for within a
strictly cognitive perspective. This all inclusive nature of system intelli-
gent behaviour places the Systems Intelligence framework in a much
wider perspective that the strictly cognitive one offered by Gardner’s
Multiple Intelligence and Systems Thinking. As explained by Saarinen
and Hämäläinen (2007). Systems Intelligence although inspired by Sys-
tems Thinking literature (Churchman 1968, Checkland 1999, Flood
1999, Senge 1999) aims to account for the practical, active thinking
240 Chapter 5

used by human agents “in real life situations involving complex systems
of interaction with feedback mechanisms” (Saarinen and Hämäläinen
2007: 51).
In translation studies literature there are authors who have indicated
the interactive aspect of learning to translate to reach a professional
level. Toury (1995) suggested the concept of socialization according to
which a beginning translator has to learn to become a member of the
translating community, by accepting its norms and working out his/her
own understanding of the profession. Kiraly (1995) focused more on
the cognitive aspects developing what he called the translator’s self-
concept, viewed as the understanding of one’s active role in the trans-
lation process. In his later works Kiraly (1997, 2004, 2005) stressed
the socio-constructivist ideas of the learning self in response to dy-
namic interaction with a feedback rich environment. Other scholars
stressed cognitive and intuitive processing (Hönig 1991) in translation
and the concept of creativity (Kussmaul 1991, Mackenzie 1998) in the
translation process when reason goes together with ‘gut feeling’ (Rob-
inson 1997). In this respect Systems Intelligence is broad enough to
encompass all the aspects of translation as a developing human skill
which combines personal mastery with socially owned mental models,
shared vision, team learning and systems thinking18 (Senge 1990).
When approaching the translator as an expert learner in the process of
developing the skill to translate it seems that there must be an evolution from
the learner’s personal mastery into systems thinking. The translator develops
his/her capacity (personal ability/mastery) to use the two (or more) languages
to bridge language barriers for other members of language communities who
cannot communicate otherwise. To do so translators have to be aware of their
mental models formed by their unique culture. The theory of Systems Intelli-
gence forms a fundamental link between personal mastery and systems
thinking. Personal Mastery is defined by Systems Intelligence as the individ-
ual’s ability to fully use one’s unique potential “to maneuver successfully in
the various systems structures that constitute the environment…” (Saarinen
and Hämäläinen 2007: 58). Mental models, one’s models of thinking deter-
mine what actions are considered necessary and possible. In a way “thinking
transforms into action and repeated action into habits” (Saarinen and

–––––––––
18
Peter Senge (1990) in The Fifth Discipline defined the above as key themes in
learning organizations (for us translators).
 Becoming a professional translator … 241

Hämäläinen 2007: 51). This simple chain of transformation describes the

translator’s progression in the evolution of his/her human skill to translate
where declarative knowledge about translation becomes proceduralised and
through repeated experience combined with self-reflection forms a repertoire
of routines, well-tested procedures and reliable work habits.
However, as observed by Senge (1990) people are mostly unaware of
their mental models, their modes of thinking which drive and direct their
actions. Similarly, translators are unaware of their procedures in their
entirety. This however, does not mean that some mental models are not
subject to self-reflection. A translator is a self-correcting learning system
where the impact of experiences is always idiosyncratic and sometimes
incidental. In Systems Intelligence thinking about thinking defined as a
fundamental meta-level capacity is crucial and includes:

– acknowledging that one’s actions and behaviours reflect the way we

think (our mental models, assumptions, beliefs, interpretations, etc.)
– acknowledging that one’s thinking can be highly idiosyncratic, one-
sided and thus distorted with reference to the holistic systemic envi-
ronment
– seeing the need to mirror one’s mental models and engage in meta-
level thinking about my own thinking in order to act more intelli-
gently, to change “my behaviour and actions to be more in line with
my true aspirations, interests and the parameters at hand, as they ap-
pear in the environment in which I operate” (Saarinen and
Hämäläinen 2007: 60)
– seeing the need for meta-level reflection on my own framing of the
environment as the holistic interactive system to overcome the idio-
syncratic limitations to improve the effect of one’s actions.

It is possible that the evolution and development of translation as a human

skill requires limitless re-framing of how the translator perceives the two
language systems and the translation environment as a complex arena of
interactive systems. As pointed out by (Saarinen and Hämäläinen 2007:
60), “[i]t is a well known fact of cognitive science and creativity research
that re-framing is a key to new opportunities, higher productivity and
creativity at large”. The motto of this chapter by Séguinot (1989b: 80),
“The key feature of expertise in the performance of skills is the ability to
restructure knowledge” lends support to this perception.
242 Chapter 5

Similarly, the role of self-reflection in translation expertise develop-

ment signaled in chapter 4 is very much in line with Systems Intelligence.
To quote,

Thinking about thinking is about identifying one’s favoured framing

patterns, challeging them and adjusting them accordingly. It is clear that
the possibilities to re-frame the holistic, feedback intensive structures
around self, as well as their relations to self, are literally limitless. At
the same time one is likely to have gotten stuck and stationed to some
particular framing (Saarinen and Hämäläinen 2007: 60).

A mature translator as an expert learner (Schön 1983, Ertmer and Newby

1996) and a Systems Intelligent person will acknowledge the limitations
of his/her own thinking and mental models, re-framing and re-adjusting
his/her Knowledge Integration Network (KIN) according to the changing
interactive environment and feedback structures. This continuous re-
thinking of one’s own thinking is fundamental to the development of
translation as a human skill. The translator as an expert learner through
the repeated activity of translating and by reflecting upon the choices
made builds up a repertoire of experiences of building KINs, hypothesiz-
ing solutions, assessing them, rejecting, accepting, etc. As an expert
learner the translator learns to self-govern his/her learning process, re-
frame if necessary unproductive ways of going about the translation task.
In effect the translator develops his/her Systems Intelligence and becomes
more and more self-confident, more and more empowered (Tymoczko
2007) to solve problems and make decisions either cognitive or intuitive
while at the same time being aware of moving within certain organized
and rule-governed systems, e.g., L1 and L2, the respective socio-cultural
realities, the specific knowledge domains and the people who participate
in communicative interaction made possible by the translator. As put by
Tabakowska (2003), in the game of translation, the translator deals the
cards but he/she is not the only player. Irrespective of the fact whether
one subscribes to the view of general intelligence (the ‘g’ theory) or to the
framework of multiple intelligences (see Waterhouse (2006: 211), and
Gardner and Connel (2000: 292) for different frameworks) complex skills
like translation require a synergy of knowledge and skills applied by a
Systems Intelligent translator. This integration of different factors accom-
panied by self-reflectivity has also a wider ethical dimension when one
considers the fact that translators are in many ways social agents (Chau
 Becoming a professional translator … 243

1991), hence their Systems Intelligence becomes paramount. As observed

by Tymoczko (2007: 220), “Only self-reflectivity can alert translators to
the various constraints – internal and external – that they face in pursuing
their ethical, political, and ideological goals in translation, making it pos-
sible to come to grips effectively and strategically with the complexities
of their historical and cultural contexts”.
To summarize the translator is a constant expert learner as frequently
admitted by authors of guides and experienced translators. Through inte-
grating knowledge for the purpose of translation he/she constantly re-
structures the knowledge structures to be able to perform in a more effec-
tive way both in terms of the quality of translation and the time it takes to
produce a translation. In this way one can say that like in any other com-
plex skill, the translator becomes more confident and has more trust in
his/her own abilities and is more aware of his/her limitations. The mem-
ory routes have been traversed so many times that it takes much less time
to retrieve information, or a desired TL word. Since the notion of self-
confidence is an essential product of expert learning and an essential as-
pect of expertise let us look at how translators develop self-confidence.

5.4.2. The translator as a self-confident expert

The concept of self-confidence is essential when discussing skill devel-

opment. Hönig (1991: 88) pointed out that “self-confidence is fundamen-
tal to effective and successful translating” but not many scholars have
tried to dismantle what this really means and where does self-confidence
come from. Whyatt (2009a: 379) discussed the concept of communicative
confidence with reference to communicative competence and pointed out
that both notions are not necessarily parallel when it comes to L2 learners’
performance although people tend to assume that self-confidence sub-
sumes competence. In her paper entitled “Building L2 communicative
confidence through interlingual tasks”, the author demonstrated that trans-
lation tasks which abound in open-ended problems challenge the L2
learners’ communicative confidence by creating doubt which has to be
overcome in order to make a decision on how to solve a particular prob-
lem. Confidence like courage, is the quality which is summoned when the
need arises. Confidence is needed when in doubt and in need of making a
decision and courage is needed in the face of fear and in need of deciding
how to act. To quote,
244 Chapter 5

Since many translation problems are open-ended, in the sense that there
are no readily available rules/formulas that can lead to appropriate solu-
tions, the task requires creative thinking with fluency and flexibility in
the choice of strategies as the most wanted companions. This puts the
L2 user [here translator] in situations when the outcome is uncertain
and, as indicated above, when doubt arises and the outcome is uncertain
confidence is needed to make decisions (Whyatt 2009a: 379-380).

Many TS scholars observed that when making decisions novice and ex-
perienced translators differ primarily in the length of time needed to han-
dle a particular problem. Experts are characterized by accurate and reli-
able judgments but also by ‘economy of effort’ even in the face of ‘rare or
tough cases’ (Hoffman 1997: 199-200). Some scholars point to expertise-
based intuition defined as “intuition rooted in extensive experience within
a specific domain” (Salas et al. 2010: 941) and stress its role in efficient
decision making grounded on self-confidence. Donald (2008: 191) ex-
tends the perspective from a personal to the social dimension of decision-
making saying that although “decision making is a very private thing, in-
dividualized and personal, it also has a cultural dimension”. As stressed
by Donald (2008), since the way we acquire knowledge and skills is al-
ways supported by the culture in which we grow up and become educated
in, the decisions that we make “engage learned algorithms of thought im-
ported from culture”, whether we use language, play chess or conduct
business negotiations. Some decisions are made quickly and some take a
long time especially if they are going to affect other people like in the
case of translation problems. TAP research has demonstrated that novice
translators with some limited experience in translation take a long time to
make decisions which is frequently interpreted as a sign of low or no self-
confidence. They constantly look up words in dictionaries to find confir-
mation for their decisions because translation questions their knowledge
of vocabulary (Whyatt 2009b) and even in their native language they lose
their language instinct and their self-confidence in their language skills.
The combination of personal decision-making with its social dimension
seems ecologically valid for professional translators and possibly requires
personal yet socially grounded self-confidence.
Why does translation challenge our linguistic self-confidence when
we become involved in the activity? The reasons are various. First of all,
we know that we are not speaking for ourselves, we only give our voice to
somebody else and we feel intrinsically obliged to do the best we can for
 Becoming a professional translator … 245

the original communicator to communicate his/her message clearly and

without changing its sense and purpose. As many professional translators
have frequently experienced interpreting the sense and communicative in-
tentions is not always easier if the SL text is in one’s native language. For
example, let us consider a brief excerpt from an article I once was asked
to translate:

Prace badawcze wykonane w obrębie infiltracyjnego ujęcia wody „Dę-

bina” w Poznaniu miały na celu rozpoznanie zmian w aluwiach rzeki
Warty (w strefie aeracji – do głęb. 1,5 m) zaistniałych pod wpływem
wymuszonej infiltracji wód powierzchniowych. Zidentyfikowano efekty
kolmatacji chemicznej, biochemicznej i mechanicznej oraz wskazano na
uwarunkowania sprzyjające jej rozwojowi. Podatność osadów na
przeobrażenia wywołujące kolmatację wynika głównie z: anizotropii
przepuszczalności, związanej z dużym zróżnicowaniem facjalnym
osadów (zarówno pionowym jak i lateralnym), licznych przewarstwień
mułków, spowalniających lub ograniczających infiltrację, a jednocześ-
nie ulegających rozmywaniu i przemieszczaniu w zawiesinie, obecności
w aluwiach szczątków organicznych (Skolasińska 2003: 7319).

I still remember what I felt having read the article. Although it was writ-
ten in my native language, it was an alien register and my native speaker
self-confidence was challenged and of no help. I had no background
knowledge to interpret the meaning of words and creating an adequate
KIN required a lot of factual research and as a result was extremely time-
consuming but, to my gradual self-discovery, not impossible. Then sup-
ported by parallel texts in English (since there was no adequate dictionary
of geological terms available at the time) and with expert advice from a
geologist with some although limited knowledge of English but extremely
useful as a valuable source of terminology, I did translate the article
which was later published in a research journal. I frequently self-reflected
on how much I learned translating that article not only about geology, as
this knowledge was only transient and temporarily needed for my KIN,
but about translation and self-confidence. Through integrating all the
knowledge which was needed to translate the article and having built an
adequate KIN I gained grounds for my self-confidence.
–––––––––
19
Skolasińska, Katarzyna. 2003. „Kolmatacja osadów w strefie aeracji pod wpły-
wem infiltracji wód powierzchniowych (na przykładzie ujęcia wody ‘Dębina’ w
Poznaniu”. Przegląd Geologiczny 51/1: 73-78.
246 Chapter 5

The above example illustrates that the translator’s self-confidence is a

fluid quality summoned by the need to solve problems and make deci-
sions. It is not something the translator has once and for all because in to-
tally novel situations and unknown knowledge domains it has to be sup-
ported by the knowledge acquisition needed to build a network of knowl-
edge (KIN) so that translating can be done with confidence. If the transla-
tion is done in a familiar context, in an area that we have previously cre-
ated KINs for, then our decisions will be made with confidence, faster and
more adequately. If we translate in an area which we do not have much
knowledge about our cognitive expenditure to create a KIN will be larger
to make us feel self-confident and knowledgeable enough to give our
voice to other experts in the fields in which our expertise is limited.
Lack of factual background knowledge is not the only challenge to the
translator’s self-confidence. The fact that for a vast majority of translators
one of the working languages is not the language they feel completely at
home with (L2, language B) is a major threat to feeling self-confident
when using it in translation. As discussed in chapter 2, the L1 is stored in
our procedural (implicit) memory (meaning we use it without the need to
explicitly apply rules) and the L2 if learned consciously as knowledge is
stored in our declarative (explicit) memory. Bearing in mind that transla-
tion is the conscious and deliberate use of both language systems with the
imperative to activate only the TL in the actual translation performance
the procedural/declarative divide may prove tricky for the translator’s
self-confidence. Whyatt (2007b: 142) reported that L2 advanced learners
of EFL at university level felt annoyed by their slow lexical access to their
L1 words and their lack of certainty as to how to combine words into
natural L1 collocations.
When one consciously tries to retrieve an L1 morphosyntactic pattern
from the bilingual mental lexicon the information is not explicitly avail-
able but accessible as an implicit procedure. In other words, if we ask
ourselves, z czym się kolokuje zysk? [what does the word profit collocate
with?] (Whyatt 2010) we are approaching our native language knowledge
from a metalinguistic declarative perspective. This means that we are
running the risk of slowing down lexical access or transferring an L2 pat-
tern (which is consciously available because it was learned) and we might
say zrobić zysk (to make a profit) displaying an influence of our L2 on our
L1. It is possible that the lack of self-confidence disturbs the interplay be-
tween cognitive and intuitive processing.
 Becoming a professional translator … 247

As pointed out by Donald (2008), “unconscious or “intuitive” deci-

sions are often the best, and many successful decisions occur in an auto-
matized manner, in highly over-predicted situations” (Donald 2008: 191).
In the context of translation this intuitive mode is frequently blocked off
either by lack of experience in traversing the procedural/declarative di-
vide in novice translators or by problems with language control (see chap-
ter 2) due to fatigue in experienced translators, for example. In both cases
low self-confidence is visible in the translator’s performance.
When translating into one’s L2 the challenge to one’s self-confidence
usually comes from the incomplete communicative competence in the for-
eign language. The pragmatic knowledge or discourse grammar (Paradis
2009) is likely to be compromised due to gaps in L2 socialization. As a re-
sult, in the use of L2 for translation not only the inexperienced face numer-
ous problems referring to whether or not a particular word can or cannot be
used in a given context (Whyatt 2009b). The knowledge needed to make
such a decision is usually too subtle to teach (Nida 2002) and has to be ac-
quired by practice either in communicative situations or in translation. Pro-
viding optimum conditions for learning all the little nuances in meaning as
being dependant on the context and situation remains a challenge for L2
teaching/learning methodologies and is of viable interest to translation
training pedagogy. More confident L2 users are more likely to develop into
self-confident translation professionals.
It seems that the notion of self-confidence is a key factor in the devel-
opment of translation expertise and a product of repeated translation ex-
perience. It is based on sound judgment of one’s true abilities and on the
awareness of and adherence to job ethics. Yet, it is tempting to ask what
drives the development of self-confidence in those people who aspire to
become professional translators. Is it the promise of a well-paid job, or is
it simply passion for the activity in itself? The first reason seems unlikely
in the present situation in which the profession is socially undervalued
and underpaid. The role of affective factors in translation skill develop-
ment to reach the level of expertise seems a better candidate as a driving
force behind the translator who as an expert learner engages in the process
of building his/her self-confidence.
248 Chapter 5

5.5. Undervalued affective factors in TC acquisition

The fact that those who seek to develop translation skill find translation a
challenging, demanding and rewarding activity plays an important part in
the transition from translation as a bilingual ability to translation as a skill
which can be trained, practiced and eventually mastered sufficiently
enough to allow one to call oneself a self-confident competent profes-
sional.
As observed by Shreve (1997: 125) translation skill will exceed the
bounds of the natural untrained ability only if a bilingual person will in-
tentionally choose to work towards it and for whom translation becomes a
“deliberately sought out” communicative experience. Consequently, TC
acquisition can be seen as a learning process of a highly motivated expert
learner. This possibility is very likely for the simple reason that translator
training programmes place very high demands on candidates (i.e., L2
mastery discussed in chapter 2) and are highly selective and usually avail-
able to a small number of students. At the English Department where I
work the conference interpreting programme takes on 12 students out of
about nearly 200 EFL students. The MA course in written translation
which started in 2010 accepted 15 students out of 70 candidates. Having
gone through a tough recruitment process, it can be expected that the
trainees are indeed highly motivated.
Another motivation enhancing factor comes from the nature of trans-
lation as a purposeful communicative activity. Translation is a highly hu-
manistic act whose aim is to help others to communicate for various rea-
sons (Whyatt 2009b: 198). It is summoned to fulfill a strong human need
to share information, ideas and generally to interact with others whether
on a private or public level. The awareness of this social role is highly
motivating for the translator irrespective of the point on the developmen-
tal continuum, or “evolutionary space” (Shreve 1997) of translation as a
human skill. Natural translators, L2 learners or L2 users can translate
even if the formal aspects of their performance fall short of professional
standards. Jensen (1998) is emphatic that motivation especially internal is
of utmost importance for the long-term educational effect of the learning
process. Memory research has also confirmed the role motivation plays
for the retention of learnt material, as well as for the future recall of
needed knowledge both declarative and procedural (Squire and Kandel
2009). Translators as self-motivated expert learners through practice be-
 Becoming a professional translator … 249

come aware of how much diverse knowledge is frequently needed to al-

low for fluent translation. Much of this knowledge is language-related and
sometimes reliant on subtle socio-pragmatic rules. Negotiating doubts
when solving translation problems and restoring one’s self-confidence,
recovering from ups and downs in the translation process involves emo-
tions whether one is aware of them or not. In consequence the memory
traces are comparable to those which accompany a player in a game of
skill (Whyatt 2009a: 375) who takes the challenge, acts in a goal-oriented
way and receives a reward (even if the reword is the very experience of
playing the game).
It is also possible that the intense information processing required in
translation provides a very stimulating environment for interplay between
emotions and cognition. In this respect further internal motivation comes
from the complex intellectual demands of translation as a cognitive intel-
lectual task. Whyatt (2009b) reported that 91% of subjects (advanced L2
learners) said that they treated translation tasks as a kind of challenge, a
brain teaser which offers an intellectual virtual reality game in which the
L2 learner acts as a translator (see Gorlee 1994 on applying game theory
to translation). By assuming different social roles, e.g., when translating a
recipe, or a tourist brochure, etc. the translator enters the realm of cross-
cultural empathy and constantly learns new words and new concepts.
It is also plausible that translation tasks by their game-like nature pro-
vide students with strong motivation “the desire to communicate” what
Paradis (1992) calls the microgenesis of an utterance.20
Perceiving translation tasks as a kind of virtual reality game allows
one to resort to a variety of strategies which combine analytical thinking
and intuitive problem solving. In translating it seems that the translator
might find “an abstract, game-playing pleasure, marrying both objective,
analytical reasoning and more intuitive or creative thought processes”
(Jones 2002: 130, quoted after Chmiel 2004: 249). This aspect of doing
something with pleasure has rarely been discussed in translation skill de-
velopment as if it did not fit the educational context, or as if it might belit-
tle the serious aims that are ahead of translation trainers. To quote,

–––––––––
20
This strong motivation is typical of L1 acquisition but it is mostly missing in the
learning of an L2 in a school environment resulting in a lack of dopamine release (Schu-
mann, 1998, see also Damasio, 1994, 1999 on the neglected area of the role of emotions
in learning).
250 Chapter 5

Strangely absent in theoretical speculation on translation teaching have

been theories of play and game in language. This is all the more surprising
in that any attempt to theorize intuition in thought and creativity in lan-
guage must surely take into account the enormous cognitive contribution
of play in human development (Bruner et al. 1976) (Cronin 2005: 259).

As Volkovitch (1998: 242) explains, “the ludic dimension is in no way in-

compatible with rigour, even if the latter is hidden by way of modesty or
design” (quoted after Cronin 2005: 258). Yet, as observed by Gorlee (1994)
the translator involved in the activity fits well in the description of a ra-
tional player who challenged by the task to translate will mobilize the entire
cognitive potential. This is noticeable in translation classroom discourse
with multiple voices being heard in which a wealth of various pieces of
knowledge surface and might contribute to finding a solution to a transla-
tion problem. It does seem that the learners in a collaborative effort (see
González Davies 2004) are involved in a game of sometimes hit and miss
with frequently amusing effects which create a good-humoured atmosphere
favourable for creative thinking (see Kussmaul 1991, 1995). When two
languages and two worlds meet in translation, there is a frequent unin-
tended collision of concepts and forms. Having taught translation for 17
years I have frequently observed what I called ‘a laughing test’ which de-
scribes the students immediate reaction of bursting out with laughter when
a similar amusing effect is unintentionally achieved by their peer, a novice
translator. Only later the analysis and the explanation of the error which re-
sulted in evoking unwanted associations in translation follows. Nobody
takes offence and the only conclusion which is drawn is in the need to be
vigilant about the fact that languages play their own games if not kept un-
der control. Accordingly, Kussmaul (1991, 1995) saw the potential of the
ludic elements of translation as a creative process of problem solving,
pointing out that the subjects who participated in his TAP studies during the
incubation stage diverged from conscious problem solving into laughter
and fooling around, playing with language and frequently this relaxed at-
mosphere resulted in illumination and finding a solution which would not
be found by analytical reasoning. As Cronin (2005: 258) points out,
“[t]here is an awareness therefore, of ludic potential in translation peda-
gogy, but the actual theorization of play itself is almost wholly absent from
speculation (see Cronin 1995)”. Still, the fact that translation as a task is en-
joyable and intellectually rewarding is enhancing the motivation of the
learner involved in the process of TC acquisition.
 Becoming a professional translator … 251

For practicing and professional translators, the effect of the translation

experience is highly motivating, on the one hand by the financial gain
(being paid for the work) but this external motivation is not of primary
importance. As many practicing translators would agree the actual work
involves long hours in front of the computer screen, intensive effort to
meet the deadlines and frequently no feedback from the client apart from
the frequently delayed paycheck. The internal motivation is then more fu-
eling on an everyday basis possibly drawn from the unlimited intellectual
demands of translation tasks which continuously expose lack of knowl-
edge or lack of communicative confidence whether the knowledge one
has is applicable to a specific translation task. Overcoming these limita-
tions and playing with the uniquely flexible tool that language is (Dela-
bastita 1993) seems to be a potential source of self-fulfillment. Newmark
(1999) in his interview with Monica Pedrola admitted that he likes to
translate because translating is “a crossword puzzle, it’s something very
attractive, I love to translate for the same reasons: because you have to try
to fit things in; it’s not only work it’s a game as well as work and it helps
you to understand people” (Pedrola 1999: 20).
All these affective factors show that there is a role emotions play in
the process of TC acquisition and as such require more attention (see
Hayes 1996). Learning to integrate knowledge and build an adequate KIN
for each translation task creates the need for cognitive adaptation and
creativity. The above mentioned translation-inherent motivation enhan-
cers are well known to translator trainers but they also provide explana-
tion that translation as a human skill is well suited to self-development,
self-improvement and self-mastery when the translator is perceived as an
expert learner.

5.6. Conclusions

Drawing on what is known about the nature of translation with respect to

its function and process I have hypothesized that the development of
translation as a human skill to reach the level of professional expert
knowledge is dependent on the ability to integrate all the knowledge re-
quired for a task at hand. Translators to translate a text have to create in
their mind what I termed a Knowledge Integration Network (KIN). The
KIN is in this respect similar to the model of translation competence put
forward by Cao (1996: 328 after Piotrowska 2007: 127) in which atten-
252 Chapter 5

tion is drawn to the fact that in translating different sets of knowledge are
activated. In my proposal it is not only activation that is important but the
integration of the needed knowledge domains. It is through the integration
of the knowledge networks that linguistic tools are prepared for the actual
process of writing down the TL text. KIN is then an expression of the
translator’s cognitive adaptation to a particular translation task (see
Buchweitz and Alves 2006).
It is expected that KIN is performed in preparation for the translation
of every text. However, in its nature it is a dynamic constellation which is
flexible and subject to on-line modification as the translator progresses
with his/her task. It remains active throughout the whole translation proc-
ess, including revision and editing. It responds to the meta-cognitive su-
pervision and the actual physical articulation of the process into a prod-
uct, i.e., the typing of the translation. In simple terms KIN acts as a stock-
taker. It either assures the translator that ‘yes, you have all the necessary
knowledge that is needed to translate this text’ or it informs explicitly or
implicitly, ‘you need to complete your KIN to translate this text because,
for example:

1) The area is unfamiliar, e.g., your knowledge of geology is next to

none you will have problems understanding the text although it is
in your native language.
2) Because of the missing conceptual knowledge you are also missing
adequate vocabulary in both SL and TL lexicons to talk about it.
3) You have all the conceptual knowledge but since you have not
dealt with this knowledge domain in your TL (i.e., L2), you will
not have the necessary terminology. For example, names of plants
mentioned in the description of the park which is for sale together
with the palace in the business offer you are translating’.

It seems that the ability to integrate knowledge essential for the develop-
ment of translation as a human skill involves and possibly starts at the
level of bilingual knowledge. First of all the knowledge components un-
derlying the use of two languages have to be integrated including linguis-
tic competence/knowledge, pragmatic knowledge and metalinguistic
knowledge. The integration however is always responsive to the commu-
nicative needs of the situation in which we want to use language. In our
L1 communication we rely on linguistic competence and pragmatic
 Becoming a professional translator … 253

knowledge but when needed we will rehearse our words and make a con-
scious metalinguistic effort to find the most pleasing or most hurtful ones.
In L2 communication we will consciously monitor our performance to
protect ourselves from grammatical errors, we will more clearly assess the
socio-cultural context not to come across as somebody impolite using for
example, pragmatically inappropriate forms of address. In translation we
are constantly maneuvering between the two systems with their intricate
socio-cultural connections. Frequently we are treading on thin ice and
self-confidence that the situation is manageable and that we will remain in
control comes from our trust in the ability to integrate all the necessary
knowledge for the task at hand. The precise parameters are different for
every translation task. The outcome of the knowledge integration process
is different and of different use for every translator depending on the stage
on the developmental continuum but the general principle of activation
and integration is always there. What is more, the effect of the ability to
integrate knowledge results not only in conceptual activation but also in
lexical activation by either producing a priming effect on lexical access or
informing the translator that there are gaps in knowledge and external
sources have to be used to provide information and vocabulary. As such
an adequately created KIN stimulates and activates the translator’s long
term memory, the translator’s most important tool (Bell 1991, Wilss
1999). Research shows that inexperienced translators frequently become
locked in at one knowledge domain, e.g., the lexical knowledge. Although
memory navigation is an essential aspect of the translation process and
significantly decisive in terms of the speed and quality of the translation
performance it has not received extensive attention confirmed by empiri-
cal data. KIN attempts to provide a theoretical framework in which the
human skill to translate relies heavily on the enhanced retrieval capacity
from the translator’s LTM and/or on the research skills to find and apply
whatever is unavailable in the internal memory in external memory ar-
chives (these include reference books, dictionaries, data bases, and re-
cently most frequently the Internet).
It is self-evident that today’s technology has transformed the way trans-
lators integrate knowledge and create intricate active networks which pro-
vide background knowledge, facts and terminology needed to translate a
particular text. As predicted by Wilss (1999) thanks to IT and CAT tools
knowledge is literally available at one’s fingertips and the capacity for stor-
age and the retrieval of information is being constantly expanded. Still the
254 Chapter 5

need to select and integrate what is adequate for a particular translation task
remains the responsibility of the translator. Kelly (2005) makes an interest-
ing observation and notes that experienced translators of today might have
quantitatively more knowledge in their personal database whereas those
who train to be translators at the present moment instead of having a lot of
declarative knowledge themselves posses better research skills and know
where to search for what they need to know. In both cases the experience of
building an adequate KIN will result in memory traces each and every
translation experience leaves in the translator’s LTM. These traces become
stored in the LTM and can be reactivated in future tasks. As a result, when
translating in well known domains and scenarios the translator will rely on
his/her routine performance (i.e., quick lexical access via lexical links) es-
pecially when the demands placed on the translator seem to exceed his/her
processing capacity. It is only through automatizing certain subtasks that a
complex skill such as translating can be developed to the level of expertise.
Wilss (1996: 102) put it briefly: “as experience with translation increases,
the demand of cognitive expenditure decreases”. It seems that the process
of developing translation expertise as the highest stage in the evolution of
translation as a human skill can be seen as the growing ability to perform
all the multitasking within the limited capacity of the human mind. This
multitasking is achieved through knowledge integration which is needed to
translate with efficiency. The ability to integrate knowledge can be devel-
oped either through translator training or through self-learning and self-
coaching. Accordingly, the translator devoted to expertise acquisition can
be perceived as an expert learner, a systems intelligent person who by prac-
tice combined with self-reflection develops his/her professional self-
confidence. Finally, it has also been suggested that affective factors play an
important role in the process of translation expertise development seen as
the ability to integrate and apply knowledge.
The proposal of KIN discussed in this chapter might seem yet another
theoretical speculation. Many questions arise: Can it be empirically
tested? How and where can we find evidence that depending on the stage
of translation skill development, the translator will differ in his/her ability
to integrate knowledge? It seems unlikely that this particular aspect of
translation skill can be thoroughly researched within one discipline of
Translation Studies. It is much more probable that an interdisciplinary ef-
fort is needed including such disciplines as psychology both experimental
and cognitive, psycholinguistics and neurolinguistics which due to their
 Becoming a professional translator … 255

research methodologies and rampant recent technological advances have a

lot to offer in this respect. The words of James Holmes (1988), come to
mind that ‘we should go back to a laboratory’ in order to hypothesize and
test our hypotheses in empirical studies. Chapter 6 and 7 are a modest
contribution to empirically validate the hypothesized ability to integrate
knowledge in the process of translation expertise development.
Chapter 6

Empirical investigation into the development

of a Knowledge Integration Network in
translators – part one

The purpose of this chapter is to empirically verify the emergence and de-
velopment of the ability to integrate knowledge in the process of the evolu-
tion of translation expertise. Drawing from the available understanding of
translation as a human skill and recognizing its bilingual foundations, the
theoretical proposal of a Knowledge Integration Network discussed in
chapter 5 will be tested from two complimentary vantage points, the trans-
lator and the translation process. This chapter will analyze the first pool of
data in which the Knowledge Integration Network will be sought. The data
comes from questionnaires completed by translators who occupy different
stages on the evolutionary continuum of their translation skill development.
The groups of subjects are described with reference to their bilingual foun-
dations and their translation experience. The purpose of each questionnaire
is described and the data are analyzed with respect to the theoretical pro-
posal of perceiving the progression on the continuum of translation skill
development as a process of knowledge integration. The detailed analysis
of the views and opinions of potential and practicing translators is carried
out in line with the evolutionary perspective presented so far in this work.
The discussion of the results aims at detecting implications for the educa-
tional purposes of translator training as well as at pointing to further re-
search into the development of translation as a human skill.

6.1. The purpose and the subjects of the questionnaire study

The aim of the questionnaire study was to gather data from a statistically
relevant number of people who are either potential or practicing transla-
tors and whose opinions can be analyzed to provide a representative pro-
file of a translator-to-be and a translation professional who earns a living
providing translation services. A battery of questionnaires included differ-
258 Chapter 6

ent sets of questions prepared for each of the four groups of respondents.
Some questions appeared in all the questionnaires and some were specific
to the stage the subjects occupied on the developmental continuum as
translators. The assumption behind the questions chosen for the study was
that by sharing views and experience the subjects will shed some light on
the development of the human ability to translate and support the theo-
retical proposal of knowledge integration networks (KINs). Both the
questionnaire method and the purpose of the investigation meant that this
part of my research did not aim to test hypotheses but intended to present
a broad picture of those who are involved in translating, either as EFL
students who might seek career in translation or as professional practitio-
ners. The observational data was then subjected to a detailed analysis with
respect to the proposal of KIN presented in chapter 5. The subjects were
randomly chosen to represent a particular stage in the development of the
translation skill continuum, ranging from absolute novices to profession-
als with up to 40 years experience in translation.
The first group of subjects (1BA) consisted of 80 students of English as
a Foreign Language at the end of their 1BA year in the School of English at
Adam Mickiewicz University in Poznań, Poland which is the largest centre
for English studies in Poland and in Europe. The School of English has the
reputation of high standards of teaching English as a foreign language and
many students come from all over Poland to study English and become
teachers, translators or language experts employed in various areas. The
courses offered to 1BA students apart from some content courses run in
English such as British/American literature or linguistics and cultural stud-
ies focus on the development of the four basic language skills speaking, lis-
tening, reading and writing. There is no translation course offered to 1BA
students which allows to treat them as a group representing absolute begin-
ners in translation, but also as a group of potential translators if they decide
to pursue such a career in their future professional life. At the time of the
questionnaire the 1BA students had just completed their first year of study-
ing English in a structured learning environment at university level. In a
way their bilingual foundations have only begun to be structured. In terms
of language proficiency, having passed the final Practical English exam
(PNJA which stands for the Polish Praktyczna Nauka Języka Angielskiego)
they are at the level comparable to grade B/C of the CAE (Cambridge Ad-
vanced Exam in English). The questionnaire prepared for the 1BA group
aimed to elicit the subjects’ personal ideas about translation as a profession
 Empirical investigation into the development of … 259

and activity. Indirectly it also elicited their assumptions and expectations

about the necessary skills expected of translators.
The second group of subjects (2BA) consisted of 40 students of English
as a Foreign Language at the end of their 2BA year at the same department
as the 1BA group. At the time of the questionnaire they had just completed
the second year of studying English in a structured learning environment at
university level. Their bilingual foundations have become more consoli-
dated and more advanced. In terms of language proficiency having passed
the final Practical English exam for 2BA they are at the level comparable to
grade A of the CAE (Cambridge Advanced Exam in English). The major
difference with the 2BA group apart from having studied English at univer-
sity level for two years was that during the second year at university all the
students had attended a translation course. Although the translation course
is not meant to prepare them to become professional translators as it first of
all aims to improve their language proficiency, it nevertheless is structured
according to the professional understanding of the nature and purpose of
translation. In this way during the translation course the 2BA EFL students
experience translation as an activity and become aware of the challenges
and requirements which are involved in intercultural communication via
translation. The tasks range from translations of short non-literary texts in
several areas chosen for a yearly course (e.g., sport, business, mass media,
tourism, politics, etc.) to field work projects which aim at raising awareness
of the communicative function of translation as a service to community.
Tasks also include critical evaluations of translations available on the open
market (e.g., restaurant menu translations, tourist information available on
the internet, etc.). Initially students translate from English, their foreign
language into Polish, their native language, and in the second semester they
also translate from Polish into English. As for classroom procedures most
work is done in the spirit of cooperative learning, including collective effort
and open discussions of translation problems. Peer correction is frequently
used to encourage taking the receiver’s perspective on a translation as a fin-
ished product. Experience shows that through translation students become
more aware of language as a tool and develop their metalinguistic ability as
well as metacognitive skills practiced during the translation process (see
chapter 3). The questionnaire prepared for the 2BA group aimed to elicit
some information about their bilingual foundations as well as the subjects’
personal opinions about translation as an activity. Following their personal
experience of translation the questions included issues referring to the ef-
260 Chapter 6

fect translation experience has had on their language skills with the as-
sumption that their answers might provide support for the developing abil-
ity to integrate knowledge in the initial stages of the evolution of translation
as a human skill.
The third group of subjects (2MA) included 40 students of English as a
Foreign Language at the end of their final year at university that is just be-
fore their graduation and considered ready to start their first job. Most of
them will work as teachers of English or as language experts employed on
the open professional market, some might pursue careers in translation or
join an academic research community as PhD students. At the time of the
questionnaire they had just completed the second year of their MA pro-
gramme pursuing the study of English at a highly advanced level in a struc-
tured learning environment at Adam Mickiewicz University in Poznań.
They had all passed their last practical English exam at the end of their
1MA year. Their bilingual foundations had been assessed as solid in terms
of speaking, listening, reading and writing in English. In terms of language
proficiency having passed the final Practical English exam for 1 MA stu-
dents they are highly proficient at the level comparable to grade A/B of the
CPE (Cambridge Proficiency Exam in English). Apart from the translation
course which is obligatory for the 2BA students the 2MA students had not
been exposed to any more translation practice in a structured learning envi-
ronment. The questionnaire prepared for the 2MA group replicated some of
the questions prepared for the 2BA group. The questions were aimed at as-
sessing the subjects’ bilingual foundations relying on their self-assessment
as well as at eliciting their opinions about translation as an activity which
they experienced three years prior to the questionnaire study (as 2BA stu-
dents). The specific interest was in any detectable changes in the subjects’
opinions about translation as an activity and the required skills between the
three groups of potential translators (1BA, 2BA, 2MA).
The fourth group of subjects (PT) included 40 professional translators1
who have been providing translation services on the open market. Com-
pared to the three previous groups, the group of professional translators was
the least homogenous in terms of age and professional experience as prac-
ticing translators. There were 12 translators with professional experience
from 1 to 9 years, 17 translators who have been providing translation ser-

–––––––––
1
I use the term professional translator as synonymous with practicing translator, i.e.,
providing translation services on an open market.
 Empirical investigation into the development of … 261

vices for 10 to 19 years and the remaining 11 translators have been in the
profession for 20 to 40 years. The vast majority of the professional transla-
tors were members of the Association of Polish Translators (Stowarzyszenie
Tłumaczy Polskich) which is the oldest association of professional transla-
tors in Poland and the only one that gathers translators of all specializa-
tions, literary and non-literary, written and oral. Being listed as a member
of the Association is in a way a guarantee of high professional standards as
to become a member of the Association one has to meet strict requirements
(see the STP webpage at: http://www.stp.org.pl/). Since in Poland like in
many other countries (e.g., the USA) accreditation is not required to pro-
vide translation services there are many translators who are providing trans-
lation services and enjoy a good reputation while not being members of the
Association of Polish Translators. In this respect their length of practice
seems the best indication of their professional standards (see Shreve 1997,
Chriss 2006, Gouadec 2007). For the majority of the translators (35) Polish
was their native language, L1 (language A in the TS terminology) and Eng-
lish was their first foreign language, L2 (language B in the TS terminol-
ogy). There were 2 natural bilinguals (English-Polish and Polish-German)
and 3 translators for whom English was their native language (language A)
and Polish was their language B. A vast majority of the professional trans-
lators who filled out the questionnaire are freelance translators and provid-
ing translation services is for them not their only source of income. Some
of the translators who responded to the questionnaire are currently em-
ployed in the EU. In other words the only criterion for all the respondents
in the fourth group (PT) of the questionnaire study was providing transla-
tion services on an open market. The questionnaire prepared for profes-
sional translators aimed at eliciting information about their educational
background, bilingual foundations, attitude to translation as an activity and
specific procedures applied when translating. The last set of questions en-
couraged the professional translators to reflect upon their professional ca-
reer and if possible give advice to beginning translators.
All the groups of subjects completed their questionnaires anony-
mously and voluntarily. Although, the questionnaire for professional
translators was sent by e-mail, the identity of the respondents was not in
any way considered in the study. All the respondents did not receive any
incentive or remuneration apart from my gratitude and sincere thanks for
sharing their experience in a collective endeavour to increase our under-
standing of translation as an activity, as a human skill and as a profession.
262 Chapter 6

6.2. Discussion of the results

The results will be discussed in the evolutionary perspective of translation

skill development. Starting with ideas and assumptions about translation
as a profession and as an activity expressed by potential and practicing
translators, through investigating the subjects’ personal language acquisi-
tion history which allows assessing the bilingual foundations on which
translation expertise can be built, the focus is next shifted to translation
professionals. The answers provided by the practicing translators to 50
questions as well as the comments they left on the questionnaires were
analyzed to provide empirical data on the evolution of translation as a
human skill. The evolution is reflected in the continuum from ideas and
assumptions verified by data on translation practice and reflected in the
self-assessment of transformations within the translator as an individual.
The invaluable assistance of all the subjects who participated in the ques-
tionnaire study, with the practicing translators as the most precious infor-
mants, has allowed to draw a developmental profile of a translator which
will be discussed in detail in the sections below and with reference to the
hypothesized ability to integrate knowledge into KINs.

6.2.1. Translation as a profession – assumptions and facts

A lot has been said about the lack of recognition and respect for the work of
translators in Translation Studies (cf. Hejwowski 2004, Snell-Hornby
2006). Contrary to the rather disillusioned underappreciated voices among
those studying the working conditions of professional translators which fo-
cus on drawbacks rather than advantages, translation as a profession enjoys
popularity among foreign language students. The 1BA group of subjects
with no structured experience in translation was asked to share their ideas
about translation as a profession. In total 80 students filled out a question-
naire consisting of 10 questions. The results show that their beliefs about
the profession are very positive. As many as 85% see the job of a translator
as interesting, 70% would like to pursue a career in translation in compari-
son with a mere 5% who stated that they would not like to work as transla-
tors in their future professional life, and 25% who chose the ‘I don’t know’
answer to the question: “In your future career would you like to use English
and work as a translator?” The willingness to become a professional trans-
lator in future remains quite high for the 2BA subjects but it is significantly
 Empirical investigation into the development of … 263

lower for the 2MA group with 47.5% of those who would like to work as
translators as compared with 30% who do not see themselves as translators
and 22.5% who remain undecided. The graph below shows the changes in
the subjects’ response to the question whether they would like to work as
translators in their future professional life.

1BA
80% 70%
70% 2BA
60% 52.50%
50%
47.50% 2MA
40%
40% 30%
30%
25% 22.50%
20%
5% 7.50%
10%
0%

Yes No I don’t know

Graph 1. The drop in EFL students’ willingness to take up a career in

translation.

The results showing a gradual decline in the percentage of the EFL stu-
dents who would like to become translators as they progress in their uni-
versity career from entering the academic community towards graduation
are open to interpretation. It is possible that the overwhelming 70% of the
1BA students hold an idealized view of a translator at work which is fre-
quently shared by those who start to learn a foreign language. The major-
ity of 1BA subjects see the job of a translator as appealing, interesting,
important and prestigious. This interpretation is corroborated by the an-
swers given by the subjects when asked to state three reasons why they
would like to work as translators. The most popular reasons were eco-
nomic including good pay followed by prestige, rewording work, interest-
ing (you constantly learn new things), challenging work, fun, flexible
working hours and work needed on the market. These positive features of
the translating profession are, nevertheless, accompanied by a more real-
istic view which is able to find the not-so-ideal aspects of the profession.
When asked to state three reasons why they would not like to work as
translators, the 1BA subjects came out with the following lists of reasons:
stressful, monotonous, dull, doesn’t pay well, hard work, pressure, re-
264 Chapter 6

sponsibility for making mistakes. Counterbalanced by these negative fea-

tures assumed to be a part of a translator’s life as a professional, the posi-
tive assessment is perhaps not idealized but lacking in experience. How-
ever, this is also not entirely true since out of 80 1BA subjects as many as
70 (87.5%) said that they had translated a text for somebody else.
This report of prior translation experience in 1BA students confirms that
translation as a skill is expected of bilinguals, and by many lay people as-
sumed to be a part and parcel of being able to use two languages for com-
munication. It is nevertheless quite likely that although the 1BA subjects
experienced producing a translation for somebody else, they had never re-
ceived structured feedback on how they actually performed but most likely
self-discovered that it is not as simple as they themselves had expected.
Still, testing the 1BA students with respect to the expectations that they as-
sociate with the job of a translator reveals the beginner’s perspective.
More detailed assumptions about the skills and qualities of a profes-
sional translator were elicited from the 1BA group indirectly by asking
the respondents to give three reasons why they themselves would (or
would not) make a good translator. Their answers show what skills or
qualities they feel are expected of a professional translator. The most fre-
quent reasons why the 1BA subjects felt they would make a good transla-
tor were included in the following answers: I’m creative, I have a rich vo-
cabulary, I have wide interests, I’m confident, I like learning, I enjoy writ-
ing, I’m a keen reader, I enjoy playing with words, I’m communicative,
I’m imaginative, I’m flexible, I’m meticulous. On the other hand, the rea-
sons why some 1BA subjects felt they would not make a good translator
having themselves not much experience in translating were included in
the following answers: I need to master the L2, I have grammar problems,
I have problems with remembering vocabulary, I lack self-confidence, I
lack self-discipline, I’m nervous when speaking in public. The above data
shows that the 1BA students do not only have an idealized view of the
profession but they possibly due to some, although much limited experi-
ence in translation, show awareness of the necessary skills intrinsic to the
job of a translator. What is more, the subjects in all of the groups of EFL
students stated their preferences for the kind of translation they would like
to perform. The changes in the preferences for oral versus written transla-
tion, or possibly both kinds of modalities are also significant and point to
the growing awareness of translation as a skill. As the graph below shows
it is written translation which seems to be a preferred kind among all the
 Empirical investigation into the development of … 265

subjects starting with 59% of the 1BA group going up to 75% for the 2BA
group and finally reaches 91% for the 2MA group.

100%
1BA
91%
90%
2BA
80% 76%

70% 2MA
59%
60%
50%
40%
30%
30%
20% 11% 12% 12%
6%
10% 3%
0%

Written Oral Both

Graph 2. EFL students’ preferences for oral versus written translation.

The preference for oral translation (interpreting) only is low at 11% for the
1BA group and 12% for the 2BA group, and it further drops to 6% for the
near graduates (2MA group). Similarly the percentage of those who express
their interest in both kinds of translation is lower with the time spent at uni-
versity: 30% for 1BA drops to 12% for the 2BA and down to 3% for the
2MA group. A possible explanation of the growing preference for written
translation and the decline in the number of EFL students who express their
interest in becoming interpreters includes their growing awareness of the
requirements of the task due to experience gained in translation classes. No
matter how limited this experience is, as compared to the experience avail-
able in translation training programmes, the results show that the EFL stu-
dents become aware of the psychological demands which are intrinsic to
the task of interpreting (including all its kinds) such as an excellent ability
to concentrate and divide attention between listening, understanding what is
being said and producing what has been said in the TL. The constant need
for multitasking while remaining in control of the flow of communication
as well as coping with time pressure and stress caused by being a vital link
in the intercultural communication of the participants (see Shlesinger 2000,
Pöchhacker and Shlesinger 2002) not to mention the sometimes unfavour-
able working conditions (e.g., lack of coherence and clarity in the utter-
ances produced by a SL speaker, external noise or fatigue) is too much to
266 Chapter 6

consider by a large number of EFL students. Further job requirements for

interpreters include excellent interpersonal skills, public speaking skills, the
ability to cope with the unexpected, thorough knowledge of the SL and TL
and cultural implications which are necessary in intercultural mediation,
and finally the readiness to travel and provide interpreting services when
and where they are needed are discouraging to many language students.
These high demands are able to be met only by a small number of EFL stu-
dents (Angelelli and Jacobson 2009), especially as they get closer to mak-
ing final decisions about their future professional life just before they
graduate. Possibly the majority of the EFL students, by the time their five
years of intensive language studies are over, have already discovered that
the requirements of interpreting as a professional career do not match their
language skills as well as their personality profile (see Schweda-Nicholson
2005: 109). They might more readily see themselves as translators who do
not have to appear in public but work usually in the privacy of their own
home. If working as freelancers they are their own bosses and can choose
when they want to work and when they can take a break. Obviously though
free lancing as a translator means that the income is never guaranteed there-
fore it is frequently seen as a secondary source of income to supplement an
elsewhere earned salary, usually in teaching English as a foreign language.
This interpretation is confirmed by the data collected from question-
naires filled out by professional translators. A vast majority of the profes-
sional translators described themselves as freelance translators and for a
slim majority the translation services they provide are not the sole source
of their income. Out of 40 professional translators 80% described them-
selves as freelance part-time translators and only 20% described them-
selves as full-time translators. Out of those 2 translators were also running
their own translation companies and 6 were employed in the EU transla-
tion services. For 45% of the professional translators, providing transla-
tion services is the sole source of their income whereas for 55% percent
of the respondents providing translation services supplements their in-
come as they work as professionals elsewhere. Out of 27 freelance trans-
lators who provide translation services part-time there were 16 university
lecturers, 1 PhD student, one editor, two retired academics, one retired
engineer and a former academic teacher, one psychotherapist and a person
employed in banking. Four respondents did not specify their major career
although marked that translating is not their only source of income.
 Empirical investigation into the development of … 267

As for the kind of translation services provided by the respondents, the

results differ a lot from the preferences reported by the EFL students in
the sense that the largest number of the professional translators (18 out of
40 = 45%) declared that the services they provide include non-literary
written translation and interpreting. There were 11 professionals who spe-
cialized in written non-literary translation, 7 who dealt with written liter-
ary and written non-literary texts. Only 2 professionals out of 40 were
providing interpreting services only and 2 translators marked all kinds of
translation that is written literary, written non-literary and interpreting
services. No respondents marked the written translation of literary texts as
the only kind of services provided by them. The graph below shows the
distribution of translation services provided by the respondents.

80% 72.50%
70%

60%

50% 45%

40%

30%
17.50%
20%

10% 5%
0%

Written Non-literary Written literary Oral only Both oral and w ritten
non-literary

Graph 3. The distribution of translation services with respect to modality

and kind provided by practicing translators.

The distribution of translation services shown in the graph above is

probably a reflection of the market demands. The majority of texts which
need to be translated on an every day basis are most likely non-literary,
including all the translation for the tourist industry, business, press agen-
cies, documents, brochures, manuals, academic papers, etc. The transla-
tors’ readiness to provide such services is a response to the market de-
mands as well as the result of their business-like approach to their profes-
sion. Not surprisingly then, as many as 38 translators (95% of all the re-
spondents) provide such comprehensive services.
268 Chapter 6

It is interesting to note that literary translation was not marked as the

only kind of service provided and that the seven translators (17.5%) who
marked providing translation services of literary texts were also involved
with the translation of non-literary texts. This can probably be explained
in terms of economy. Translating novels, for example means taking up
long-term projects, and unless one has a good publisher who will pay ad-
vance money, the income is delayed until the translation is accepted by
the publisher, or might be delayed until the book is published. The finan-
cial aspect is plausible especially for the novices on the translation mar-
ket. Out of the seven translators who specialized in literary translation
only two admitted that translation was the only source of their income.
This result shows that, as it has been advocated by many TS scholars, the
division between literary and non-literary translation is probably irrele-
vant from the translator’s point of view. Consequently, all the disputes
whether translation is an art or a craft seem unproductive when in today’s
professional world translation means business (Sofer 2006, Chriss 2006,
Gouadec 2007). The basic stock of skills required to translate is probably
the same for non-literary and literary translation but some additional
qualities might decide about the preferences to take up literary translation
or interpreting (Gile 1995/2009, Tabakowska 2003, Piotrowska 2007).
The latter seems especially interesting with only two respondents (5%)
focusing on providing exclusively interpreting services. However, when
grouped with those translators who opted for both non-literary written
translation and interpreting as many as 50% of all the respondents said
they were ready to provide their services as interpreters.
The results showing preferences for oral versus written translation
elicited from the professional translators are below matched with the ex-
pectations and preferences expressed by the EFL students to show the
change in the distribution of translation services with respect to both mo-
dalities.
 Empirical investigation into the development of … 269

100% 95% 1BA

91%
90% 2BA
76%
80% 2MA
70%
60%
59% PT
50% 45%

40% 30%
30%
20% 11% 12% 12%
6%
10% 5% 3%
0%
Written Oral Both

Graph 4. Differences in the distribution of modality preferences between

potential and professional translators.

The changes in the distribution of the preferred modality by practicing

translators show that professionals, possibly with experience, become
more versatile in the services they can provide. Although the division
between written translation and interpreting is generally agreed upon
and not many scholars would question the different demands both kinds
of translation place on the translator/interpreter (Gile 1995/2009), it is
possible that from the translator’s point of view both translating and in-
terpreting draw from the same pool of expertise and require knowledge
integration. Experience gained in both written translation and interpret-
ing remains available for future tasks and is potentially able to refine the
skill of integrating knowledge (KIN) in the translation process. This in-
dicates that the skill of integrating knowledge is transferable irrespective
of the modality, in which the process of translation will be finalized, ei-
ther as a text or as speech. Although these issues require a more thor-
ough analysis, it is possible that the experience gained in interpreting
and in written translation is mutually beneficial for translating in both
modalities, and yet it requires differently distributed attention and dif-
ferent mental effort management in the process. To sum up, with respect
to modality, the preference for written translation expressed by the ELT
students’ remains valid for the professionals, however, there is a diver-
gent tendency as far as combining the services of interpreting and writ-
ten translation is concerned with a growing number of translators who
270 Chapter 6

are providing both kinds of translation, oral and written. Does it mean
that with time and experience professionals become more versatile in
terms of modality and the focus on one modality (2 out of 40 focused
only on interpreting) is an exception rather than a rule, or is it again a
response to the market demands? Further data might shed some light on
the specialization issue.
In response to the question, ‘Do you specialize in particular fields
(registers?’), ‘No’ was a rare answer. Only 26% of the translators stated
that they do not specialize in specific domains but are ready to translate
any text if commissioned to do so. The majority of 74% stated that they
do specialize in certain specific areas. Quite frequently the respondents
listed more than one area and often there were several related fields men-
tioned, for example finance, banking, law, marketing or philosophy, psy-
chology, popular science, or medical sciences, chemistry, biology, or
popular science, non-fiction literature, children’s books. Rarely the areas
mentioned were unrelated, e.g., IT, science, history or musicology, legal
language, property. One respondent who specialized in engineering and
technology included a comment, ‘but not exclusively and depending on
the market situation’. One professional translator in response to the ques-
tion, ‘What advice would you give to a beginning translator?’ said,
‘choose an area and specialize in it’. The results thus show that the major-
ity of professional translators are not, as it might seem to some, a ‘Jack of
all trades, master of none’. They might be versatile as far as modality is
concerned but as far as knowledge domains are concerned they do prefer
to specialize in some areas.
This result is clearly supporting the proposal of a Knowledge Integra-
tion Network (KIN). Specializing in a certain area the translators are ac-
cumulating factual knowledge and parallel to it linguistic means to ex-
press concepts and terms familiar to a specialist in the area and alien to a
lay person. Every translation task will require integration of the knowl-
edge kin to the task which will become integrated and ready for use.
When KIN is set in motion, the conceptual activation will, at least in the-
ory, prime access to linguistic means in the target language or will inform
the translator about the need to acquire either the knowledge needed
and/or the linguistic means which are not available in the translator’s
LTM to express the conceptual content of the SL text for the TL readers.
Clearly, the richer the conceptual base for a given area the faster the proc-
ess of building a KIN should be for a new text, and consequently the
 Empirical investigation into the development of … 271

faster the translation process which follows. Indeed, one translator ad-
vised the beginning translators not to accept a job in an unknown domain
unless for their self-educational purposes.
Relying on the data presented in this section, it can be tentatively con-
cluded that the principle of knowledge integration networks (KIN) for the
purpose of translation, irrespective of the modality in which translation
takes place, is assumed even by beginner translators (see comments from
the 1BA subjects). Although beginners focus more on the need to consoli-
date and integrate their bilingual knowledge, the clear preference for spe-
cializing in certain areas expressed by practicing translators points to the
role of knowledge integration in the translation process. It seems that the
need to integrate knowledge for every translation task results in the on-
going process of building the translator’s knowledge base which will fa-
cilitate similar translation tasks in the future. This knowledge accumula-
tion effect of KINs is acknowledged by the professional translators and
perceived as an intellectual reward, specific ‘perks of the job’. One of the
respondents expressed it in the following way when giving advice to be-
ginning translators, “Be prepared and open for a lifelong learning process.
Regard that profession mainly as a challenge and a passion, and if you are
good enough money will come by the way ... ☺”.
In this analysis of ideas about translation as a profession gathered from
the ELF students (potential translators) and comparing them with the real
life experience of practicing translators it seems that what stretches be-
tween potential translators (1BA, 2BA and 2MA groups) and the practicing
professionals (the PT group) is a developmentally different attitude to
knowledge integration. For inexperienced translators or for those with lim-
ited translation experience, the integration is local and includes only the
two languages. For experienced practicing translators the integration is
global and includes their entire knowledge base (also the extralinguistic de-
clarative knowledge) resulting in the preference for specialization in some
chosen areas. This assumption is further corroborated by the strong prefer-
ence for the written modality of translation expressed by the ELF subjects
which gives more time for building KINs, and allows for the use of external
memory sources, and the very low preference for oral versus written mo-
dality in practicing translators. It seems that practitioners because of their
experience can reap the benefits of their ability to integrate knowledge
when performing in either oral or written modality. The differences in the
ideas about translation as a profession and the dynamic changes which
272 Chapter 6

were discussed in this section as far as the EFL students’ interest in pursu-
ing a career in translation is concerned allow to expect that there are major
transformations in the students’ ideas about translation as an activity.
Matching the novices’ views against the practitioners’ accounts in this re-
spect may throw more light on the evolution of translation as a human skill.

6.2.2. Ideas about translation as an activity

When asked to assess whether translating was in fact more difficult than
they had expected 77.5% of the 2BA group admitted that it, indeed, was.
This significant majority of the 2BA subjects must have shared common
expectations that the skill to translate is simply a matter of knowing two
languages (cf. Paradis 2005, Whyatt 2010). Most likely these expectations
have become verified by their own translation experience and the corrective
feedback they have received from their teachers as well as from their peers
during their yearly translation course. These results also point to the educa-
tional effect of translation practice on the subjects’ changing ideas about
translation. The growing awareness that translation is not an easy task of
switching between the two languages remains with the EFL students even
three years after they had participated in a translation course. As many as
75% of the 2MA subjects still admitted that translating was more difficult
than they had expected (see Whyatt 2008, 2009b). Both groups of subjects
with the same overwhelming majority of 87.5% agreed that translating
takes a lot of time and effort, admitting in this way that it is a time-
consuming activity. In both groups of subjects the majority of 65% in the
2BA group, and 72.5% in the 2 MA group admitted that they needed to
make breaks when working on a translation task. What is more as many as
87.5% of the 2BA subjects considered translation to be a tiring activity as
compared with 56% of the 2MA subjects who reported that after translating
a text they felt very tired. This finding might summon two interpretations;
one that there is a correlation between the L2 proficiency and the mental ef-
fort needed to translate and the second possible interpretation might point
to the subjects growing ability to integrate knowledge when translating.
As indicated in chapter five, bilingual knowledge is an essential part in
the process of building a KIN for translation. It is to be expected that for
2BA students this part of the knowledge structures is still very week and
deficient especially in the areas of conceptual competence and pragmatic
knowledge. Considering the course of studies in which EFL students are
 Empirical investigation into the development of … 273

provided with intensive language practice in all four basic language skills
throughout the following three years to reach the point of graduation as
2MA students, their bilingual knowledge is far more advanced and better
structured in comparison with 2BA students. Since all the EFL students
showed a preference for written translation a comparison of their writing
skills between the 2BA and 2MA years should prove explanatory for the
difference in their L2 proficiency. As 2BA students, they learn to write
short essays in English and their writing exam consists of writing a 500
word essay, which is expository, whereas as 2MA students they had already
had to write not only research papers but their BA diploma work in order to
complete their BA studies. At the time when they filled out the question-
naires which are being analyzed here, the 2MA group had completed their
MA theses and the students were awaiting their final MA exam. It is to be
expected that the 2MA students feel much more confident about their bilin-
gual skills. This increased communicative confidence is possibly very
much supported by their metalinguistic skills which continue to develop as
they are taught to reflect upon their language use and assess language per-
formance in view of their future careers as teachers of English as a foreign
language, or as language experts. It is therefore possible that more solid bi-
lingual knowledge and metalinguistic skills do not tax the process of trans-
lation as much as for the 2BA students and consequently the task of trans-
lating does not seem as tiring as it did for the 2BA subjects. It has to be
borne in mind, however, that the translation experience is fresh and recent
for the 2BA subjects as they have just completed their yearly translation
course, therefore their judgement is empirically more valid. Since the 2MA
students might have lacked this fresh experience of translating having com-
pleted their translation course 3 years prior to the questionnaire, their
judgement might be more theoretical than practical.
The question which still remains open is whether translating as an activ-
ity is, indeed, a high energy consumer. Although, it is somehow admitted
that it is demanding in terms of mental effort that the translator has to ex-
pense, there is very little research done into mental effort management dur-
ing the translation process with some exceptions mentioned in chapter five
(see also Gile 1995/2009, Séguinot 2000, Whyatt 2010). The Polish charter
of translator’s rights and responsibilities issued by Stowarzyszenie Tłumaczy
Polskich [Association of Polish Translators] specifies that the number of
pages which can be expected to be translated daily should not be more than
5 standard pages where a standard page consists of 1,600 characters includ-
274 Chapter 6

ing spaces. These recommended amounts of text for translation are some-
thing clients are often unaware of. It is not uncommon that their expecta-
tions of the speed with which a translation can be produced are in sharp con-
trast to the recommended 5 pages per working day. The questionnaire results
show that for the vast majority of practicing translators, translating as an ac-
tivity is a high energy consumer. As many as 31 translators out of 40
(77.5%) said that translating is a high energy consumer whereas only 22.5%
said that they do not think so. One translator included a comment that trans-
lating is tiring but it, of course depends on the kind of the text, time avail-
able and “own well being”, another one added “tiring but highly rewarding”.
One respondent included an explanatory comment quoted below:
On a personal level, I find translation can be tiring, particularly when I
have to spend time in front of the computer for so many other things. Yet,
increasingly I find that the solution to this is physical exercise, so in some
ways one could say translation does not consume enough of my physical
energy. I may be unrepresentative here in that I do not find writing tiring
in the same way as I find speech tiring (with many people this is the other
way round). Interpreting fatigues me hugely – both physically and men-
tally – and generally I avoid doing it wherever possible.

The graph below shows the assessment of translation activity as a high

energy consumer in the three groups of respondents discussed above.
2BA
100%
87.50%
90% 2MA
77.50%
80%
70%
PT
56%
60%
50% 44%
40%
30% 22.50%
20% 12.50%
10%
0%

Yes I don’t think so

Graph 5. The assessment of translation activity as a high energy consumer.

The fact that for the majority of subjects in the 2BA and 2MA groups
translating was perceived as a difficult and tiring activity was nevertheless
insignificant for the generally positive attitude the subjects had towards
 Empirical investigation into the development of … 275

translating. As many as 90% of the 2BA subjects said that they enjoyed
doing translation tasks and as many as 75% of the 2MA subjects were still
of the same opinion. These results corroborate my previous research
(Whyatt 2008) which included 104 EFL students (57 were 2BA students
and 47 were 1MA students) in which 83% of the 2BA subjects admitted
that they enjoyed doing translation tasks and there was only a 7% drop in
the results achieved for the 1MA group (76%). The possible educational
benefits stemming from the fact that students simply enjoy a task have al-
ready been discussed in chapter three (see also Sewell 1996, 2004,
Whyatt 2008, 2009a, 2009b).
I also asked professional translators whether they enjoy translating as
an activity. The results show that the vast majority of 92.5% admitted that
they do enjoy translating. One out of the 37 translators included a com-
ment, “love-hate relationship at times”, three translators who did not en-
joy translating also left comments saying, ‘but it depends on the text’,
‘rather no but it has some good points’, ‘I enjoy it only sometimes’. This
vastly positive attitude towards translating as an activity displayed by pro-
fessional translators in terms of work psychology must have a beneficial
impact on how they carry out their work as well as on their motivation
and sense of self-fulfilment (see Graph 6).

2BA
100% 90% 92.50%
90% 2MA
80% 75%

70% PT
60%
50%
40%
30% 25%

20% 10% 7.50%

10%
0%

Yes No

Graph 6. An assessment of translation as an enjoyable activity.

As it was mentioned in chapter five the affective factors involved in the

task of translating have not been granted due attention, yet their role
seems significant in the development of translation as a human skill to
reach the level of expertise but also in the maintenance of that skill, and in
276 Chapter 6

the further evolution as professionals and experts. Again as mentioned in

chapter three and five a positive attitude to translation as an activity en-
hances the learning process in which translation skill is developed and
communication skills become refined (in case of L2 learners). It increases
the internal motivation to improve which, when combined with external
motivation, praise from teachers or financial remuneration in the case of
professionals help to carry out the task and increases the chances of self-
fulfilment and satisfaction when completing the task. A question, ‘Is
translation for you a rewarding activity?’ was directed to practicing trans-
lators and they were offered three answers to choose from: a) ‘yes’,
b)‘no’, c) ‘only sometimes’. The majority of 29 translators chose ‘yes’,
one translator added, ‘although it does not pay’, 10 translators chose ‘only
sometimes’ and one said, ‘quite rarely’. Generally, then, translating is a
rewarding activity for practicing translators. This positive assessment of
their work is comparable to the positive assessment of the EFL students
who had experienced translating in the course of their language studies
and in the vast majority said that they had a clear sense of achievement
when they had completed a translation task (Whyatt 2009b).
Another question aimed at eliciting information indicating the role of
affective factors in how professionals approach translation directed to
practicing translators was phrased in the following way: ‘How do you
treat the translation of a new text?’ and it offered three optional answers to
choose from: a) as a challenge, b) as an intellectual activity and c) just
work. The graph below shows the distribution of answers.

80%
67.50%
70%

60% 55%

50%

40%

30% 25%
20%

10%

0%

As an intellectual activity As a challenge As just w ork

Graph 7. The professional translators’ approach to a new text.

 Empirical investigation into the development of … 277

Bearing in mind that the practicing translators could mark as many an-
swers as they felt appropriate, the majority marked more than one answer.
The most favoured pair was ‘a’ and ‘b’, rarely all three. As many as
67.5% said that they treat it as an intellectual activity, 55% marked ‘chal-
lenge’ and only 25% marked ‘just work’. Out of those translators who
marked ‘just work’ only 5 chose it as the only answer. Interestingly, four
translators who said that they treat translation as just work had full time
employment as translators and only one was a freelance translator. Those
for whom translation was just work were in the clear minority, whereas
for the vast majority of translators the translation of a new text is rather an
intellectual activity and a challenge. This result confirms the role of affec-
tive factors present in creative problem solving involving knowledge in-
tegration in the search for solutions. It seems that for a career in transla-
tion to be fulfilling and satisfying there seems to be the need for interplay
between the physical, the intellectual and the emotional domains of hu-
man experience (Sofer 1997/2006). A similar conclusion arises from ana-
lysing the reasons which keep translators in the profession. When transla-
tors were asked to mark the reasons out of the following choice: a) finan-
cial (pays well), b) intellectual (I learn new things all the time) and c)
emotional (I help others to communicate) as many as 35 out of 40 marked
‘b’ (87.5%), 24 marked ‘a’ (60%) and 13 marked ‘c’ (32.5%).

100%
87.50%
90%

80%

70%
60%
60%

50%

40% 32.50%
30%

20%

10%

0%

Financial (pays w ell) Intellectual ( I learn new Emotional (I help others to

things all the time) communicate)

Graph 8. Reasons which keep practicing translators in the profession.

278 Chapter 6

Again here most respondents chose more than one reason, some felt all of
them applied in their case, only 4 out of 40 (10%) chose the financial as-
pect as the sole motivation for being in the profession. It is well evi-
denced that the intellectual aspect of translation as an activity is important
and generally acknowledged by the novices and professionals alike. The
fact that such a vast majority of professional translators appreciate the fact
that through translation they constantly learn new things and find their
work intellectually rewarding is significant in the context of KIN pro-
posed in chapter five. A comment left as advice to beginning translators
illustrates the point:

Treat translating as a challenge – it will provide you with an opportu-

nity for change, development and improvement in many ways: profes-
sionally and personally. Enjoy working with words, phrases and gener-
ally texts! The attitude is very important.

It seems that the repeated experience of knowledge integration will sig-

nificantly integrate language skills with all the knowledge structures acti-
vated and integrated for a task at hand. Since however, bilingual knowl-
edge is a dynamic function of one’s language learning history (see chapter
2), and as such is always open to improvement more insight into the bi-
lingual foundations of all the groups of subjects might reveal some inter-
esting points of relevance for the development of the human skill to use
this knowledge as a part of KINs needed to translate.

6.2.3. Bilingual foundations of translation as a human skill

It is not an overstatement to say that it is bilingual knowledge that provides

a critical link in all the knowledge structures which have to be activated and
integrated for translation to take place. In chapter two some attention was
devoted to the bilingual foundations which first have to be laid before one
can consider the emergence of translation as a human skill. In this section I
would like to present some data aimed at providing a real life illustration of
bilingual knowledge in the three groups of subjects, 2BA, 2MA and the PT
group.
In the questionnaires prepared for the three groups of subjects, a set of
questions addressed the subjects’ language acquisition history and self-
assessment of their bilingual knowledge. Since it turned out that out of 40
professional translators there were only two natural bilinguals (5%) and
 Empirical investigation into the development of … 279

as many as 28 (70%) were graduates of modern language departments

comparing the three groups of subjects seems justified, although the re-
maining 25% of practicing translators came to learn their language B
(first foreign language) through other sources. They most frequently in-
cluded, being raised in an English speaking country, living or working
abroad or self-study, passion for language, or various kinds of training
and intensive language practice. The graph below shows the different
routes to learning language B in the group of practicing translators.

80%
70%
70%

60%

50%

40%

30% 25%

20%

10% 5%

0%

Graduate of modern Natural bilingual Other sources

language departments

Graph 9. The different language acquisition histories among practicing

translators.

To provide more statistical data on their language acquisition history, all

the subjects were asked to specify when they started learning their first
foreign language which in all cases was English. Interestingly the largest
number of subjects from the group of professional translators started to
learn English at 10 to 19 years of age, there were three professional trans-
lators who started to learn their working foreign language at the age of 20,
23, 26. They would probably constitute interesting case study subjects for
the opponents of the critical age hypothesis in SLA studies (Singleton and
Lengyel 1995, Mitchell and Myles 1998). The average age of onset for
learning English in the 2BA group was 8.8, and 9.8 for the 2MA group,
and slightly later at 10.7 for the PT group.
280 Chapter 6

The youngest age was 4 (two professional translators and two 2MA
students). The oldest age for professional translators was 26, for the 2MA
group 15, and for the 2BA group 13 years of age when starting to learn
their foreign language. All students from the 2BA and 2MA groups agreed
without exception that learning a foreign language is an intellectually en-
riching experience. When asked to give reasons why they decided to
study English at university level most of them stated that they simply like
English (90% for the 2BA subjects and 97,5% for the 2BA subjects)
and/or are interested in culture of English speaking countries (35% and
47,5% respectively).
It still remains to be specified what exactly lies behind the statement ‘I
like English’. One can only surmise that the positive attitude towards the
language which one decides to study in depth and master the skill of using
it reflects the learner’s cognitive make up which allows him/her to draw
pleasure from learning a different system of communication. The relation-
ship between such affective factors and motivation as well as the ultimate
attainment of the L2 constitutes an interesting area of study for SLA
scholars as well as for TS scholars interested in translation training. Pos-
sibly some more explanation could be derived from personality research
which could correlate the love for languages with specific personality
types which are more inclined to enjoy learning languages and which can
achieve the highest level of attainment, near native-like competence (see
Paradis 2009) so much desired in translation trainees (see Snell-Hornby
1992). These issues are still open to research. For the time being, it is
clear from the data that the ELF subjects can be classified as language en-
thusiasts and their passion for language gives them motivation to aim high
and master the language to meet the standards set for students of English
as a foreign language at university level. To illustrate their progression in
language proficiency, the subjects were asked to self-assess2 their profi-
ciency in English using a scale of 1 to 5. There were significant differ-
ences in the subjects’ self-assessment. In the 2MA group the most fre-
quent rating was 4 (67.5%) whereas in the 2BA group 3 was marked by
50% of the students

–––––––––
2
Self-assessment of language proficiency is frequently used in psycholinguistic stud-
ies, although its validity as an objective measure of proficiency is questionable especially
when it comes to replicating research (chapter 2).
 Empirical investigation into the development of … 281

80% 2BA
67.50%
70% 2MA
60%
50%
50% 42.50%
40%
27.50%
30%

20%

10% 5% 5%
2.50%
0%
0%
[2] [3] [4] [5]

Graph 10. L2 proficiency self-rating for the 2BA and 2MA groups.

The graph above demonstrates that the ELF students are able to self-
assess their language skills and most likely their self-assessment shows
high awareness of their knowledge of the English Language. This ability
is in fact metalinguistic in nature as it shows that L2 learners at univer-
sity level are able to reflect on their language skills and evaluate their
level of proficiency. As suggested in chapter five, this is a feature of ex-
pert learners who develop their self-confidence on the basis of a sound
evaluation of their abilities. It is therefore possible that L2 teaching
pedagogies and Translation training pedagogies have a lot in common
(Colina 2003b) if they view the learning process from the perspective of
an expert learner. What is more, both groups of subjects were then asked
to evaluate their particular language skills. In both groups the subjects
assessed their receptive skills, reading and listening higher than their
productive skills, speaking and writing. As for the fifth skill, translating
the results of both groups differed slightly. While for the 2BA group 3
was chosen by the majority, in the 2MA group 4 was chosen by the ma-
jority of respondents.
282 Chapter 6

70% 2BA
60%
57.50%
60% 2MA
50%

40%

30% 30%
30%

20%

10% 10%
10%
2.50%
0%
0%

[2] [3] [4] [5]

Graph 11. Translation proficiency self-rating for the 2BA and 2MA group.

A closer correlation between the results in graph 10 and 11 yields some

interesting results. On the one hand, the results allow the assumption that
the increase in L2 proficiency has a positive effect on the subjects’ self-
assessment of their translation skill. On the other hand, there is a discrep-
ancy between a positive growth in L2 proficiency rating for the 2MA
group and the clear drop in translation proficiency self-assessment look-
ing at the rating at 3 (5% for L2 proficiency and 30% for translation pro-
ficiency) and 5 with 27.5% versus 10% for translation proficiency. This
result may indicate the growing awareness of other cognitive skills which
are important for translation apart from L2 proficiency. To use González
Davies’s (2004) ideas it might indicate that the 2MA students have pro-
gressed to the stage of conscious competence in their translation skill
whereas the 2BA students might still be at the stage of conscious incom-
petence. What seems significant, there is a similar tendency in the sub-
jects’ self-assessment of feeling in control of their L2 when using it for
speaking and/or writing (see Whyatt 2009b). When asked whether they
have to put conscious effort into controlling what they say when they
speak/write in English, the majority of the 2BA group said ‘yes’ whereas
in the 2MA group the majority was either unaware of it or said ‘no’.
 Empirical investigation into the development of … 283

70% 2BA
60% 57.50%

2MA
50%

40% 35% 35%

30%
30% 27.50%

20% 15%

10%

0%

Yes No I’m not aw are of it

Graph 12. The L2 control factor for the 2BA and 2MA group.

This result shows a shift from consciously controlled language process-

ing, which in itself takes a lot of time and attention, to the lesser need to
consciously control language processing reported by the 2MA group.
Such progression is consonant with the cognitive theories of L2 learning
(Mitchell and Myles 1998) according to which learning a foreign lan-
guage leads from initial awareness, through active manipulation to con-
trolled processing, and finally and ideally towards automatic processing3.
It is possible that the 2MA students feel more in control of their L2, more
confident about their language skills including the skill of translation. The
issue of language control was discussed in chapter two (see also Gile
1995/2009, De Groot and Christoffels 2006) and its relevance for the de-
velopment of translation as a human skill was elaborated in chapter five.
However, in the context of the evolution of translation as a human skill
and its role in mental effort management during the translation process it
still requires a more detailed investigation. For the time being it is impor-
tant to notice that in both groups, 2BA and 2MA the highest rating (5) for
translation proficiency was rarely marked, only by 2.5% of the 2BA sub-
jects and by 10% of the 2MA subjects, although the 2MA group did not
report a high need to consciously control their L2. This result shows that
–––––––––
3
See Paradis (2009) on the illusion of automatic processing in speeded up native-
like language performance and a different view by Kroll and Linc (2007).
284 Chapter 6

the students are aware of the need and room to develop their translation
skill which clearly is not developing parallel to their L2 proficiency.
The above correlated results including L2 proficiency, translation pro-
ficiency and language control confirm Kiraly’s (1995) claim that at some
point of translation skill development the majority of students are unlikely
to further their translation skills by practicing translation on their own,
without corrective feedback and without guidance. Although as pointed
out in chapter five it is not impossible if one takes an expert learner per-
spective. As suggested in chapter four the issue of translator training is
very complex and in need of more transparent methodologies. Interest-
ingly, a clear majority of the 1BA students (96%) said that in their opinion
excellent knowledge of English is not enough to become a professional
translator and 73% declared that other skills are also important. As many
as 60 out of 80 1BA (75%) subjects were convinced that it is necessary to
take up special training to become a professional translator. This attitude
is highly laudable and in contrast with the common misconception that
everybody who knows two languages can produce a translation up to pro-
fessional standards. The feeling of the need for training to become a pro-
fessional translator in the case of the 1BA subjects can also be the result
of their experience in translating a text for somebody which was reported
by as many as 87.5% of the 1BA subjects.
Following Malakoff and Hakuta’s (1991) call for more research into
the role of metalinguistic knowledge as far as translation proficiency is
concerned, further questions aimed at assessing the students’ metalin-
guaistic awareness as an important factor in the development of transla-
tion as a human skill. The results show progression in this aspect. Al-
though both groups showed similar results when asked whether they no-
tice that they make mistakes when performing in their L2 (‘yes’ marked
by 50% for the 2BA and 57.5% for the 2MA group) , there was a signifi-
cant difference when responding to the question, ‘Do you consciously
self-monitor yourself when you speak/write in English?’ The 2MA group
in the majority of 30 out of 40 students (75%) said ‘yes’ whereas in the
2BA group although the majority said yes, 10 subjects added ‘I try, some-
times, not always’ showing that with lower language proficiency and less
control over the L2 system, there is possibly not enough mental resources
to monitor L2 performance. Also the metalinguistic ability to provide on-
line monitoring of the ongoing performance is less available to the 2BA
group possibly due to limited L2 practice as compared with the 2MA
 Empirical investigation into the development of … 285

group richer by three years of intensive language practice, and of being

more aware of using language as a tool not as an end in itself.
Following Paradis (2009) who claims that in verbal communication
L2 learners tend to rely on metalinguistic knowledge rather than on their
L2 linguistic competence another question was phrased in the following
way: ‘Do you feel that your theoretical knowledge of the English lan-
guage is better than your practical skills?’ In both groups the subjects
were equally divided with a 50% split between those who felt that their
theoretical knowledge was better than their language skills.
The discussion of bilingual knowledge in the research subjects so far has
been very much L2 dominant. In this aspect it reflects L2 teaching pedago-
gies which as mentioned in chapter four have frequently ignored the fact
that L2 learners are L1 users (Cook 2001). Most foreign language teaching
is thus monolingual and L1 is vastly marginalized and vilified for negative
transfer and interference at all levels from phonology to discourse (Odlin
1989, Kecskes and Papp 2000). When however we discuss the bilingual
foundations of translation skill both languages have to be granted a legiti-
mate place in the mind of L2 users. They are both needed and important. As
many psycholinguists have noted there is a constant interaction between a
bilingual’s two languages in his/her mind which can take the form of sym-
biosis (Grosjean 1997, 2002) or a battle for dominance (Heredia and Brown
2006: 233) when the traditionally accepted labels L2 and L1 become insig-
nificant or even wrong as the L2 becomes more dominant than the L1. The
learners are taught to suppress L1 interference or at least to control it in their
language performance and necessarily in formal situations like exams. Are
EFL learners at university level aware of language interference? When
asked, ‘Do you feel that Polish words interfere when you speak or write in
English?’ the two groups differed in their answers: the 2MA group seemed
to have been much more in control of interference. No subjects said yes,
30% opted for sometimes, 42.5% said rarely and 27.5% said no. With the
2BA group 7.5% said yes, 45% sometimes, 35% rarely and 12.5% no.
These answers point to the growing capacity for language control (see
Green 1993) including control over L1 interference. This ability to keep
the two languages apart is a positive change vital for the evolution of
translation as a human skill. When two languages meet so close in transla-
tion the likelihood of language interference increases and for translators
the control of interference becomes a must (Nord 2005: 133). The aware-
ness of cross-linguistic influence (or interference) between the two lan-
286 Chapter 6

guage systems and being vigilant to overcome it is a part of metalinguistic

ability. This part of knowledge structures allows one to look at language
as a tool, to assess one’s success or failure in using it to achieve an in-
tended aim, whether the aim is speaking, writing or translating. Interest-
ingly, both groups of EFL students admitted that translating made them
more concerned about their L1, which in other words means that they be-
came aware of their L2 impact on their L1, or the L2 dominance over
their L1. From the 2MA group as many as 72.5% admitted that translating
made them more concerned about their Polish, from the 2BA group 70%
students were of the same opinion.

80%
72.50% 2BA
70%
70%

60% 2MA
50%

40%
30%
27.50%
30%

20%

10%

0%

Yes No

Graph 13. Growing concern about the EFL students’ L1 as a result of

translation experience.

It is possible that these feelings are due to the English dominant language
mode that students are in throughout their studies especially as far as writ-
ing is concerned. Dominant exposure to English in their receptive and
productive language skills is bound to affect the speed of lexical access
(see Levelt 2001, Kroll and Sholl 1992, Kroll and Linc 2007) from their
Polish word store as well as create uncertainty about correctness and ap-
propriateness when translating. This can be explained by the different lo-
cation of the L1 knowledge (procedural) and the L2 knowledge (declara-
tive) in the LTM. Getting used to the different location affecting routes of
lexical access possibly comes with translating experience and as it was
suggested in chapter five is related to the ability to integrate knowledge
 Empirical investigation into the development of … 287

into a KIN for the purpose of translation. In view of the above data it
seems sensible to conclude that the first step in translation skill develop-
ment is restoring a proper balance between the EFL students’ both lan-
guages which is not a matter of a quick and easy push-the-button change.
The emerging picture of bilingual foundations illustrated by the results
of the questionnaire study shows that they are still in the making. The sub-
jects are able to show their awareness of current problems and concerns re-
ferring to their L2 proficiency and language imbalance. When answering a
question, ‘When you speak or write in English what is your priority?’ The
majority of subjects in both groups (87.5% in the 2MA group and 75% in
the 2BA group) chose ‘content’ over ‘correct grammar’ (30% in the 2MA
group and 32.5% in the 2BA group) and ‘appropriate vocabulary usage’
(55% and 42.5% respectively). What might seem interesting, as many as
27.5% of the 2MA subjects and 15% of the 2BA subjects marked all three
aspects as equally important. These data show that in some EFL students
there is a growing awareness that everything is important in the overall lan-
guage performance and that one cannot convey good content without cor-
rect grammar and with inappropriate vocabulary use.
Further results showed the subjects growing confidence in the three
aspects of L2 vocabulary usage: meaning, contextual appropriateness and
style. Significantly they show the discrepancy between semantic meaning
and pragmatic aspects possibly due to the limited real life experience of
using their L2 in a wide range of authentic communicative contexts. This
also confirms that EFL students are aware that to be able to use an L2
word one needs to know more than its form and meaning (see chapter 2).
So far the analysis of bilingual foundations which are being laid to
make the development of translation expertise possible has indicated its
complex developmental nature as well as the growing metalinguistic
awareness in ELF students. Do their developing language skills give them
enough confidence to translate? Do they feel confident when using their
bilingual knowledge when they are asked to translate a text? Both groups
were asked precisely this question to see if being divided by three years of
intensive language learning the 2MA students would feel more confident
when translating than the 2BA group. The results show that there is not
much difference in rating self-confidence when translating. When asked
to mark their level of confidence on a 1 to 5 scale, in both groups the most
popular rating was 3, otherwise the distribution was very similar.
288 Chapter 6

50%
2BA
45%
45% 42.50% 2MA
40%
40%

35%
30%
30%

25%

20%
15% 15%
15%

10% 7.50%

5% 2.50%

0%
[2] [3] [4] [5]

Graph 14. Self-confidence in translation tasks for the 2BA and 2MA
groups.

The similarity of the results confirms that despite the tremendous differ-
ence in L2 proficiency which exists between 2BA students and 2MA stu-
dents and possibly the difference in their metalinguistic abilities, the 2MA
subjects did not seem more confident when translating. This again con-
firms that some major qualitative changes have to take place to further
their translation skills which in turn will give them more confidence when
actually producing a translation.
Another question which aimed at the self-assessment of the subjects’
translation performance asked ‘How pleased are you with your final ver-
sion once you finish the task?’ Just as it was with the question of confi-
dence the distribution of answers was very similar for both groups. The fa-
voured rating this time was 4 marked by 52.5% of the 2BA subjects and
57.5% of the 2MA subjects. This self-assessment shows that the EFL stu-
dents are self-critical and although they feel they did their best, they still
see room for improvement. This quality is precious for a future translator
(Chriss 2006: 140).
The above results are significant as they confirm that translation as a
human skill is built on bilingual foundations but it involves something
qualitatively and quantitatively more than communicative competence in
two languages. The issue of communicative confidence hinges upon com-
municative competence but includes other social and cognitive skills. It
remains to be seen that the confidence needed to translate can be built up
 Empirical investigation into the development of … 289

during translator training courses or alternatively it will evolve as a product

of translation practice combined with expert learner self-assessment and
corrective feedback. Another tentative conclusion from the EFL students’
L2 acquisition history and their honest self-assessment of the skills they
feel they possess is that learning a foreign language is a lifelong process
always open to further development and never completed if one is inter-
ested in high level of attainment or native-like L2 proficiency. What is
more, it seems that the more we know about our L2 the more humble we
feel as L2 users who come to experience the entire complexity of the L2.
This awareness stimulates further efforts on the part of an expert learner
who is aware of his/her abilities and deficiencies which have to be over-
come to progress further in L2 proficiency. This observation is confirmed
in the questionnaires filled out by professional translators.
When asked, ‘Do you feel you constantly learn more about your lan-
guages when you translate?’ as many as 82.5% translators chose ‘yes
about both A & B’ and only 3 translators out of 40 chose ‘no’, while 4
opted for ‘yes but only about my foreign languages B & C’.

90% 82.50%
80%

70%

60%

50%

40%

30%

20%
10% 7.50%
10%

0%

‘Yes’ about both A & B ‘Yes’ but only about foreign No

languages B/C

Graph 15. Translators as constant language learners.

These results show that the process of learning a foreign language is never
ending and the question of ultimate attainment is open to discussion (Kroll
and Linc 2007, Paradis 2009) but the idea that our L1 is also open to im-
provement throughout our life has rarely been raised and left out as self-
evident (Cook 2003). Yet, the fact that translators feel they constantly learn
290 Chapter 6

more about their working languages, both A and B shows how open the
process of linguistic development is. Their answers confirm that our as-
sumption that being native speakers of our L1 we know the language very
well is in itself imprecise since there are obviously words that we do not
know, and a variety of registers some of which although in our native lan-
guage might in fact be equally alien as if they were in a language totally for-
eign to us. Another issue is the subjectivity of our L1 knowledge which we
are frequently unaware of, and which does not have to agree with the public
and socially shared expectations imposed on language use. It is possible that
translators in the process of developing their professional translation skill
learn to objectify their L1 use, becoming at the same time aware of language
as social practice which has to comply with socially accepted norms and
standards (Toury 1995, Hermans 1991). The ongoing process of self-
improvement in L1 as a result of translation practice is frequently over-
looked in discussions of translation proficiency. Similar scant attention has
been granted to the need to a maintain balance between one’s L1 and L2
through awareness of cross-linguistic influence between the translator’s both
working languages resulting in efficient control mechanisms. Yet, as demon-
strated by the above results and hypothesized in chapter five a translator as
an expert learner is very much aware of the need to integrate all the knowl-
edge needed to translate and through self-reflection will become aware of
the inadequacies in both L1 and L2. Some comments left by practicing
translators as advice for prospective translators sum up the main points:

– A good translator does not only polish the foreign language but con-
stantly works on his/her native language. It is always more difficult to
translate from language B to language A since we tend to think that as
native speakers of Polish we know everything there is about our
mother tongue but in most (translation) cases it is not so.

– Admire languages for their richness and originality, love your own
language first (know it well).
Problems with L1 usage in translation have not received adequate atten-
tion and it is the deficits in L2 usage which usually come to attention in
translation criticism. When asked, ‘Which aspects of the foreign language
are still problematic for you?’ most of the subjects chose at least one as-
pect from the list including:
 Empirical investigation into the development of … 291

a) cultural references
b) article usage
c) figurative expressions
d) sentence structure
e) tenses
f) recognizing and reproducing register
g) punctuation
h) vocabulary use

Problems with article usage were marked most frequently (19 translators)
followed by ‘cultural references’ (18 translators) and punctuation (17
translators). Figurative expressions were problematic for 13 translators,
vocabulary use for 11, the remaining aspects were marked by fewer sub-
jects: 4 marked recognising and reproducing register, 2 marked tenses and
sentence structure, 1 subject (a natural bilingual) marked Polish case end-
ings as problematic.

50% 47.50%
45%
45% 42.50%
40%
35% 32.50%
30% 27.50%
25%
20%
15%
10%
5%
0%
article usage cultural references punctuation figurative expressions vocabulary use

Graph 16. Aspects of the English language marked as most problematic

by professional translators.

These sincere answers in Graph 16 are probably going to be comforting for

the potential translators (EFL students) whose bilingual knowledge has been
assessed in this section. They also stimulate possible explanations of the un-
derlying causes. The aspects most frequently causing problems such as cul-
tural references, article usage and punctuation are the most implicit aspects
of the English language even for native speakers (see Paradis 2009). The
rules of usage are not explicit and therefore difficult to teach as part of
292 Chapter 6

metalinguistic knowledge. Frequently, they are very much context depend-

ent and defy logical explanation acceptable for L2 users. It can be expected
that since translators are conscious of their weaknesses they become more
vigilant about these aspects of L2 usage and treat them with limited confi-
dence and apply thorough meta-cognitive control. The awareness of one’s
weaknesses might encourage translators to put a metalinguistic tag on these
aspects which they consider problematic. This brings about double checking
and making sure that the doubts are cleared at the revision stage.
Although the fact that translators remain for ever keen language learn-
ers has been mentioned in TS literature, not much research has been car-
ried out into how translation practice alters bilingual knowledge structures
as compared to those bilinguals who do not become professional transla-
tors. Yet, the results discussed so far confirm that translators are constant
language learners, i.e., expert learners who develop their systems intelli-
gence and professional self-confidence as an outcome of their work and in
preparation for future work. Many comments left by practicing translators
as advice for potential or beginning translators referred to what can most
generally be termed ‘language maintenance’. Such comments as, ‘read a
lot in L2 and L1’, ‘if you think you know a word always double check’,
‘if you think you know a lot you must always know, there is much more
to learn’, or ‘be prepared for a lifelong learning process’ are some of the
examples referring to the need for language maintenance in practicing
translators. Well maintained frequently refreshed knowledge will be more
readily available for activation and integration into a Knowledge Integra-
tion Network (KIN) needed for a new translation job.
Surprisingly however, the awareness of the need for language mainte-
nance does not mean that translators see themselves as language experts.
When asked directly, ‘Do you consider yourself a language expert?’ only
27.55 said ‘yes’ and the majority of 72.5% opted for ‘just a conscious
language user’.
 Empirical investigation into the development of … 293

80%
72.50%
70%

60%

50%

40%

27.50%
30%

20%

10%

0%

Language expert Just a conscious language user

Graph 17. Translators as language users or language experts?

It is plausible that the humble self-assessment demonstrated in the graph
above reflects years of experience with cross-cultural language transfer.
While self-confidence in the use of language is needed to translate, be-
coming overconfident can prove detrimental to communicatively ade-
quate translation as it was hinted by one of the questionnaire respondents.
To quote, “Be extremely careful with your self-confidence; things rarely
are as simple as they seem to be”.
Some other comments included in the section where practicing trans-
lators were asked in an open question to give advice to beginning transla-
tors offer further explanations of how practicing translators see bilingual
knowledge with reference to their everyday practice. I will quote a few to
illustrate the point.
– read a lot of quality press in language B: it has a positive impact on
your style in language B, it helps you to sound authentic in it.
– read original works in languages A, B and C; follow cultural, political,
economic, etc. events
– read a lot in Polish, be inquisitive about the world
– work professionally in the foreign country using the language as long
as possible, with no people around you who use your native language
– be humble
– I have spent many years teaching students what successful translation
is all about but in one sentence: The primary requirement is an ex-
tremely good knowledge of both source and target language as well as
of the cultures to which those languages belong.
294 Chapter 6

These comments left by experienced translators as a kind of their profes-

sional testimony for the novices in the profession point to the ongoing
language learning process and to the need for language maintenance as far
as both languages A and B (also C) are concerned. They also show the re-
spect with which translators learn to treat language as their tool. It is pos-
sible that experience in translation teaches them to be very much aware
that if not kept under sufficient control it will take revenge for being igno-
rant about its lively versatile and playful nature. Translations found on
restaurant menus are a good example4. What are the exact steps of learn-
ing to use the two languages to translate and escape all the traps set up for
unconsciously incompetent language users, still remains to be a fascinat-
ing area to be investigated.
This section focused on the bilingual foundations of translation as a
human skill. Looking at the language acquisition history of those who
might become translators and those who have experienced the translation
process in a structured educational environment as well as those who pro-
vide translation services on the open market has demonstrated the com-
plex dynamic nature of bilingual knowledge (see chapter two for theoreti-
cal references) which becomes a part of Knowledge Structures needed to
produce a translation. In the section which follows an attempt is made to
see how practicing translators apply their knowledge when producing
translations.

6.2.4. A translating professional at work

In order to illustrate how translators work let us follow an exemplar trans-

lator who receives a commission to translate a text from a client until
he/she produces the translation and sends it to the client. Using answers
provided by practicing translators in the questionnaire and for the sake of
clarity let us divide the work into several stages and fill each stage with
information that has been gathered in the study.

–––––––––
4
‘Walleye pollock (Theragra chalcogramma) in herb and cream sauce’ for the Pol-
ish mintaj w sosie ziołowo-śmietanowym mentioned by Kearns (2008: 198) ‘sparking
wine’ instead of ‘sparkling wine’, ‘pizza with edible fungus’ instead of ‘pizza with
mushrooms’ (for more examples see Whyatt 2006c).
 Empirical investigation into the development of … 295

Stage 1. A translation is commissioned and arrives by e-mail (today pos-

sibly the most probable scenario)

The translator opens the e-mail which probably reads, ‘we attach a text
for translation, and we need it by 11 am tomorrow morning’. Is there any
so called ‘translator’s brief’ attached as well which will specify who and
for what purposes needs the translation? The question, ‘When you receive
a new text for translation do you also receive the specification of who and
why needs it?’ gave four possible answers:

a) yes
b) rarely
c) I have to ask for it
d) I work it out myself

The data collected from the professional translators shows that only 30%
of the practicing translators receive any kind of translator’s brief, and the
majority of these translators are in full employment. Below are some
comments added by the respondents:

– yes, if not then I ask for it

– yes (in general, this is a matter of cooperation with my clients)
– yes, I mean I usually get books from a publisher – I know every-
thing about that publisher and I also know the target audience of
that book.
– no, almost never

It seems that in the majority of translation jobs there is no specific infor-

mation referring to the purpose and the target audience of the translation.
The graph below shows the distribution of answers.
296 Chapter 6

40% 37.50%

35%
30% 30%
30%

25%
20%
20%

15%

10%

5%

0%
Yes Rarely I have to ask for it I w ork it out myself

Graph 18. Do translators receive a translation brief?

These results also indicate that the translator has in fact an extra responsi-
bility to work out the translator’s brief which she/he will probably do re-
lying on experience gathered from cooperation with the clients so far and
inferring it from the situational context. One can as well say that the
translator can always ask for more information about the commissioned
job. The issue of cooperation with clients and all the social inter-personal
skills which are required needs a broader insight, although according to
the data gathered in this study only 20% admitted that they would like to
improve their cooperation with clients which shows that the majority of
the translators are satisfied with the work relations they have with their
clients. At this point in the entire translatorial action (Holz Mänttäri 1984)
we have come to the translator’s first encounter with the SL text. The
reading of the SL text has been differently termed by various scholars, for
example, reception stage, opening of the SL text, positioning, SL text
comprehension or SL text interpretation. All of these terms indicate some
kind of meaningful interaction which takes place between the SL text and
the translator (see Dancette 1992, Pym 1992, Whyatt 2003, Stolze 2004).

Stage 2. The first encounter with a new text for translation

The exemplar translator is opening the attached text for translation and as-
sessing its size. An inaccessible cascade of thoughts is set in motion in the
 Empirical investigation into the development of … 297

translating mind while the translator is scanning the SL text (cf. Holmes
1988 and his mind-map theory, also Hönig 1991). What do translators look
for during their first encounter with the text? When answering the question,
‘When you read (scan) a new text for translation do you first:

a) relate it to your previous experience (I’ve done something similar/It’s

completely new to me)?
b) assess its level of difficulty?
c) read it for gist?

the practicing translators varied in their opinions. As many as 18 transla-

tors (45%) opted for ‘a’, 19 (47.5%) for ‘b’ and 16 (40%) for ‘c’, four
translators said that they actually performed all three tasks at the same
time and one translator said, ‘I never read it before translating and of
course I know if it is new to me or not. But it makes no difference to me’.

Graph 19. The first reading of the SL text.

50%
47.50%
48%

46% 45%

44%

42%
40%
40%

38%

36%
Relate it to your previous Assess its level of difficulty Read it for gist
experience

The data in graph 19 show the individual preferences of particular transla-

tors. It might seem that only answer ‘a’, ‘I relate it to my previous experi-
ence’ would confirm the proposal of KIN, but it is plausible that all three
activities are performed with KIN as a general framework. Is it possible to
assess the level of difficulty of a text we have not seen before without plac-
ing it against the background of what in our estimation is perceived as easy
298 Chapter 6

or difficult due to our experience? When reading for gist, can we detach
ourselves from the cognitive background of our knowledge and experience
which allows us to understand it? Doesn’t the rule of inferential compre-
hension rely on integrating our background knowledge that is building a
KIN? It is therefore plausible to say that our task oriented exemplar transla-
tor sets a relevant KIN in motion with the first scanning of the text to be
translated. The Working Memory (WM) is responding to the visual recog-
nition system while the translator is scanning the text (see Bell 1991) and
KIN is searching through the Long Term Memory archives in order to re-
late the text to the translator’s previous experience, assess its level of diffi-
culty and get the gist of the conceptual content. It is only an initial warm up
but its outcome is essential in the form of assumptions and hypotheses
about the text (see Pym 2003). At first glance, the exemplar translator knows
for example that: a) I have done something similar before, b) the text is
fairly easy, c) it is an advert of the local airport. Or: a) I have not done any-
thing like this before, b) it is easy/difficult, c) an instruction manual for a
woodworking machine. Of course, other combinations of factors are possi-
ble. To sum up, during the first encounter with a text for translation, the
mind starts to prepare itself for the task with spreading activation and the
parallel processing of information from the SL text and from the translator’s
long term memory. The next question is, ‘Are translators aware that they
prepare themselves for translation, or do they simply start translating?’

Stage 3. Preparation

As suggested in chapter five KIN is hypothesised to operate implicitly

and explicitly, it can be either purely intuitive and unavailable to intro-
spection or cognitive, consciously engineered and thus subject to intro-
spection. It can as well be a combination of both, intuitive connectionist
thinking and analytical deduction (Robinson 1997). In other terms trans-
lators can be unaware of it or fully aware of its cognitive manifestation.
The question of what happens before the exemplar translator starts to
translate is thus only partially open to inspection. When asked, ‘Do you
prepare yourself before you start translating a new text?’, the respon-
dents were given four options:

a) yes, always
b) yes but only if the area/type of text is new to me
 Empirical investigation into the development of … 299

c) no but my mind does

d) I just start translating

70%
60%
60%

50%
40%

30%
20%
20%
10% 10%
10%

0%
Yes, alw ays Yes, but only if the No but my mind does I just start translating
area/type of text is
new to me

Graph 20. Do translators prepare themselves for translation?

Some of these who opted for ‘d’ included comments like for example:

– I do research as I translate via Google, personal library.

– I just start translating and find information on board at the time when I
need it ☺.
– Sometimes I do little preparations.
– I just start translating (and solve the encountered problems as I go
along).
– It also depends on the time available.
It is interesting to note that those translators who said that they start trans-
lating without devoting any time for preparation were in fact very experi-
enced and they have been providing translation services for 9 to 35 years.
This is in fact an interesting result. Apart from pointing to idiosyncratic
differences, it might also suggest that with experience KIN is more intui-
tive, implicit and when it is needed explicitely it surfaces as the translator
progresses with his/her translation. Any knowledge which is unavailable
in the translator’s LTM has to be accessed externally. In other words, if
the private memory does not hold the needed information, the public, ex-
300 Chapter 6

ternal or social memory is consulted (Donald 2001). However, as the

graph indicates the majority of translators prepare themselves before they
start translating if the text is new to them showing their awareness that an
adequate KIN has not been satisfactorily created in the translator’s mind
and some more knowledge has to be supplied or some kind of confirma-
tion or completion is needed to integrate relevant knowledge and translate
the text in question. ‘What do translators do to prepare themselves before
they start translating?’ The translators had three choices and they could
mark as many as they felt appropriate in their case:

a) I look at similar texts in the target language

b) I do some factual research
c) I try to refresh some vocabulary I might need

The answers provided by the practicing translators point to some direc-

tions. The majority of 28 translators marked ‘b’ admitting that they do
some factual research before they start translating, 22 said that they look
at similar texts in the target language and only 9 said that they try to re-
fresh some vocabulary they might need. As many as nine translators
marked all the options. The results demonstrated in the graph below show
that a KIN is frequently created with the use of external resources.
80%
70%
70%
60% 55%

50%
40%
30% 22.50%
20%
10%
0%
I look at similar texts in the I do some factual research I try to refresh some
target language vocabulary I might need

Graph 21. The pre-translation preparatory stage.

 Empirical investigation into the development of … 301

Doing some factual research aims at the acquisition of new knowledge or

refreshing the translator’s background knowledge which has not been ac-
cessed for some time (facts, specific information, data, etc.) but which is
necessary to produce a professionally adequate translation. Yet, the search
itself is intrinsically connected with acquiring or refreshing verbal means
(terminology either absent in the translator’s mental lexicon or dormant
due to low recency and frequency of use) in which this knowledge is ex-
pressed (see chapter two and five). Although clearly the lexical activation
gain seems secondary in the translator’s estimation presented in the graph,
as predicted by the mental lexicon research conceptual activation will
spread to the lexical level and therefore its effect can be expected to be
positively facilitating for the speed and fluency of lexical access which
will take place in the process of translation. Furthermore, although there
was no indication of the language in which the factual research was car-
ried out it is very possible that it was done in the target language of the
translation at hand. The fact that as many as 22 translators marked ‘a’ ad-
mitting that they looked at similar texts in the target language before they
started to translate shows that they felt the need to gather some more rele-
vant information on text type specification, discourse patterns, terminol-
ogy, level of formality, or any other clues which they can follow produc-
ing their translation. It is also possible that the role of KIN when the
translator is looking at parallel texts is in fact multilayered and performs a
conceptual and lexical warm up in the internal memory of the translator
who prepares his/her workbench gathering all the necessary tools and ma-
terials, refreshing needed algorithms and patterns. Looking at parallel
texts the translator is also performing a language switch. Having read the
SL text for translation the translator reaches into the TL repertoire prepar-
ing himself/herself for the lexical access process which will be running
perhaps more efficiently when the translation process is set in motion.
One translator working for the EU said, ‘when I go to interpret about
fisheries I think of fish, and all the fish business and the words which I
need when I do my job just come to my mind’5. The subjects in this study
seemed to produce the same memory activation effect. It is significant
that some translators marked all three options as valid in their case. In fact
all three ways of activating memory are essential components of an ade-

–––––––––
5
Maciej Machniewski – presentation on the EU translation services given at
A. Mickiewicz University, Poznań 2009.
302 Chapter 6

quate KIN and stimulate one another and feed on each other with their
joint purpose to activate the translator’s LTM and prepare the ground for
smooth cooperation between the working memory and the long term
memory archives. In the process of translating a text the working memory
will be calling for specific conceptual content and its adequate verbal
cladding in the TL (see Bell 1991 on the central executive and slave sys-
tems of human memory). A comment left by one of the respondents points
again to the stimulating effect of the preparation stage. When answering
the question, ‘What do you do to prepare yourself for translation?’ the
translator said, ‘Yes, to all three. My biggest problem is ‘b’ which I usu-
ally find more interesting than the actual translation’.
This shows that the intellectual curiosity required for translation is
stimulating and involves the perks of the job in the form of acquiring
more knowledge quite often in areas which are outside every day experi-
ences. The need to do research and acquire needed knowledge is appreci-
ated by the translators (see graph 8 above with 87.5% claiming that the
intellectual benefits of learning new things are the highest motivation to
remain in the profession) and frequently underestimated by clients. As ob-
served by Vourikoski (private communication) when asked to perform
telephone interpreting for a medical company selling a multitude of medi-
cal equipment she received a voluminous manual of how the equipment
operates, yet the client offered her a standard hourly rate for the interpret-
ing job and was not willing to increase the rate taking into account all the
time that she would have to spend preparing for the actual telephone in-
terpreting job. When the interpreter asked for a more adequate rate, the
client did not accept the terms and most probably found somebody else to
do the job. The need to do research when preparing for written translation
or oral translation in the case of interpreters is a part of translation as a
profession although it obviously takes different forms and has different
requirements depending on the modality in which the translation is per-
formed. In both cases research prior to translating helps to build an ade-
quate KIN and is bound to create affectively a positive attitude by remov-
ing the feeling of being in an unknown domain. Again together with fac-
tual knowledge we gain the verbal means to express it, to analyze it and to
pass it on to our potential readers. To sum up, the preparation stage is an
explicit manifestation of KIN which shows that translators need to feel
that they have activated and integrated enough knowledge structures for
the process of translation to physically begin. While the majority of trans-
 Empirical investigation into the development of … 303

lators perform this task prior to the onset of translation others prefer to
have it as if running on-line and they claim that they do not devote any
time for explicit pre-translation preparation. Some, however are aware
that although they do not literally prepare themselves their mind neverthe-
less does prepare itself for the task ahead. Following these considerations,
it seems justified to expect that the process of translation will be
smoother, perhaps faster and more satisfactory in the translator’s opinion.
What might seem interesting, as many as 70% of the translators said
that they feel that the process of translation is slow to start with and then
speeds up. The clear minority said that they were not aware of it (15%) and
the remaining 15% said that they did not think that the process of transla-
tion is slower at the beginning. What is significant here is the fact that 5 out
of 6 translators who said that they did not feel that the process of translation
is initially slower admitted that they did not devote any time for explicit
preparation before they started to translate a new text. Two out of the 6
translators who were not aware of the different speed of their translation
process also did not do any research before they embarked on the transla-
tion task. Still, it would be premature to claim that there is a close correla-
tion between devoting some time for explicit knowledge integration (KIN)
and the efficiency of translating. On the other hand, translation as a creative
task does not have to differ with reference to an initial slow pace from other
creative tasks, like for example creative writing. Even in creative writing
some warm up techniques are frequently used to put oneself ‘in the right
frame of mind’. In translation which involves language switch such warm
up might be even more productive and could suggest the facilitating effect
of KIN on the translation process. This claim however requires a more con-
trolled experimental validation. In the present analysis of the translating in-
dividual at work, we have come to the next stage, writing down the trans-
lated text that is the physical articulation of the translation process which
has been given many names by different scholars. The most frequent ones
include re-expression or re-writing.

Stage 4. Writing down the translation

As mentioned before the difficulty in analysing the translation process lies

in its hidden nature, in the metaphorical ‘black box’ which makes all the at-
tempts mere speculations and approximations. The analysis which follows
here does not make any claims to be better in its epistemological limita-
304 Chapter 6

tions. The only and essential difference here is that the voice is given to
practicing translators who will speak from their experience. In a set of
questions prepared for the PT group I tried also to tap into the procedures
which the translators apply when writing down the translation. I assumed
that the process of translating and writing down the translation will run par-
allel and that the problems with smooth transfer of the informative content
expressed in the SL into its TL version will be manifested by disruptions in
the writing process or to be more precise in the typing of the emerging TL
text. My questions referred to how translators deal with the disruptions to
the smooth flow of the typing process. Relying on my own experience as a
freelance translator and a teacher of translation courses I asked how transla-
tors deal with the problems of lexical retrieval (e.g., either due to memory
blocks or lack of needed lexical items in their own mental lexicon) that are
known to be experienced by translators (cf. Rose 1991). I wanted to analyse
their behaviour when uncertain about how to solve problems and gain in-
sight into the tools they use in the process.
Problems with recall or retrieval of the desired lexical means are a part
of the translation process and are very well documented by TAP studies
(see chapter five). I asked my respondents, ‘When you can’t recall a word
that you need when translating do you?

a) use a dictionary
b) use any similar word that comes to your mind and mark it as to be
checked later
c) leave a gap

Graph 22 below shows that as many as 82.5% of the translators said that
they use a dictionary when having problems with recalling a word they
need, 27.5% opted for using any similar word that comes to their mind
and marking it as to be checked later. Only 10% would leave a gap and 2
translators said that they resorted to all three procedures depending on the
task at hand.
 Empirical investigation into the development of … 305

90% 82.50%
80%
70%
60%
50%
40%
27.50%
30%
20% 10%
10%
0%
Use a dictionary Use any similar word that comes to Leave a gap
mind and mark it as to be checked
later

Graph 22. Coping with retrieval problems during the translation process.

The majority of the translators used dictionaries whenever they felt the
need as they progressed with their translation. Interestingly, 20% of the
PT subjects admitted that they consciously postponed using a dictionary,
possibly until the revision stage. These translators seem to rely on their
internal bilingual lexicon to start with and possibly in this way they can
develop a steady tempo of translating. Although the clear majority uses
dictionaries even during the first draft, the 20% of those translators who
do not do so shows the individual preferences among translators in their
favoured procedures during the translation process. Contrary to my ex-
pectations, there is no clear correlation between the number of years in
the profession and the tendency to delay dictionary use, as among those
who delayed using a dictionary there were translators who have been in
the profession for 8 up to 30+ years. This finding confirms Fraser’s
(1999: 26) suggestions that for an efficient translation process profes-
sional translators most likely combine the use of dictionaries with what
she calls ‘intelligent guessing’ to combat problems with lexical access
and not to let themselves despair as illustrated by a quote from New-
mark (after Fraser 1999: 25),

Sometimes when I translate, I am hoping to release better words from

my unconscious, my memory hold. I go on repeating the start of a sen-
tence and stop, expecting something fresh to emerge, to pop out. Some-
times it does; it’s a relief, and I start my smirk. More often it doesn’t,
and I despair (Newmark 1991: 32).
306 Chapter 6

Fraser (1999: 32) having studied 21 professional translators concluded that

learning to ‘release words’ from memory is essential to translation compe-
tence and expertise of self-sufficient translators where self-sufficiency
means also knowing where the limits of one’s internal mental resources.
The question of tools in the translation process is vital to take into ac-
count as it participates in the process and is essential for the speed of the
process and for the quality of the final product. The ability to use the tools
was taken as a separate subcompetence in the componential model of
translation competence (PACTE 2003 see chapter 4). More so with the
development of IT, skilled use of available tools, including electronic dic-
tionaries, the Internet, data bases and other CAT tools has gained a
broader perspective. As predicted by Wilss (1999: 231) the ‘data highway’
creates the need for a new skill of being able to know how to effectively
and efficiently search for the needed knowledge and how to cope with the
information overload. I asked my respondents which tools they use in
their every day practice giving them the following choices:

a) paper dictionaries
b) electronic dictionaries
c) the Internet
d) Translation Memories
e) Other CAT tools
f) Translator’s internet forums

The results demonstrated in graph 23 below show which external memory

tools are used by the translators analysed in this study.
95%
100% 87.50%
90% 77.50%
80%
70% 62.50%
60%
50%
40% 30%
30%
20% 12.50%
10%
0%

Paper Electronic The Internet Translation Other CAT Translator’s

dictionaries dictionaries Memories tools internet
forums

Graph 23. The external memory tools used by translators.

 Empirical investigation into the development of … 307

The Internet is used by 95%, e-dictionaries by 87.5%, paper dictionaries

are still used by as many as 77.5%. Apart from the three major tools,
62.5% use translator’s internet forums, but only 30% use translation
memories and only 12.5% use other CAT tools. The Internet is also used
by the majority of the translators when in need of factual knowledge. As
many as 37 out of 40 translators use the Internet when lacking domain
specific knowledge whereas 23 out of 40 use experts. Although 49 out
of 50 questions were closed multiple choice questions many respondents
included some comments. Here some practicing translators said that ex-
perts are used very rarely now, but previously they were consulted very
often. One translator said that the primary source of factual information
is ‘my own vast library’. Although there is not much recent empirical
data on the use of translation tools, it is obvious that the electronic turn
has resulted in rapid changes. Kuliniak (2009) in her project on the use
of dictionaries by EFL students during a translation task which was
monitored by Camtasia software (the screen spy) reported that the
Google search engine is used more frequently than electronic dictionar-
ies for purely language specific queries. The EFL students who were the
subjects in Kuliniak’s (2009) study demonstrated the time consuming
aspect of the inefficient use of the tools discussed in chapter five and the
enormous popularity of Google as a language concordancer used to look
for language specific information with the frequently mentioned rule, ‘If
you don’t know it, google it’. The results gathered from practicing trans-
lators confirm that on-line access to information both factual and lan-
guage specific is an essential part of the translation process for profes-
sional translators. It is important to remember however that the external
aids and tools are helpful if used under adequate metacognitve supervi-
sion which as hypothesized in chapter five relies on the ability to build
relevant knowledge integration networks (KIN). Although dictionaries
and the Internet help the translator, the multilayered information proc-
essing (including language processing) is conducted in its entirety in the
translating mind.
Other questions arise. Are translators aware of the temporary imper-
fections in the way they render the meaning expressed in the SL text when
typing their TL version? Do they aim at finding the best possible solutions
straightaway or do they know from their experience that it is important to
keep the inter-lingual communication going between the SL text, its con-
tent comprehended by their mind and the emerging TL text? The question,
308 Chapter 6

‘Do you mark stretches you are not happy with for further revisions?’
aimed at eliciting some information in this respect and gave a ‘yes’ or ‘no’
option.
90%
80%
80%

70%

60%

50%

40%

30%
20%
20%

10%

0%

Yes No

Graph 24. Marking stretches of the translated text for further revision.

Asking such a question was aimed at eliciting data which could confirm that
the self-monitoring device is constantly switched on during the translation
process and whatever comes into existence in the TL text is immediately
screened by the translator’s internal quality control running under the meta-
cognitive supervision based on the KIN which was created prior to the actual
translation process, and which in itself is dynamically readjusted throughout
the task. As suggested in chapter five this self-assessment of the emerging
outcome demonstrates that there is a close integration between the concep-
tual base and linguistic knowledge of both languages as systems, pragmatic
knowledge of both languages as social practice and the meta-linguistic
knowledge which assesses the produced language in terms of correctness,
communicative quality and appropriateness as belonging to the adequate dis-
course the TL text will be a part of. The answers provided showed that as
many as 80% of the practicing translators marked stretches of the TL text
they were now happy with. This of course confirms the multitasking which
takes place during translation but above all it confirms the on-line network-
ing in the translating mind which constantly integrates knowledge (linguistic,
metalinguistic, pragmatic, conceptual) in order to fulfil the task.
 Empirical investigation into the development of … 309

Another question asked whether in case of not knowing which word

to choose translators leave other options in brackets for further revision.
The whole group was equally divided between those who do and those
who do not showing again that the way translators carry out some pro-
cedures is highly individual and possibly in agreement with their per-
sonality dictating one’s favourite way of doing things. Yet, personality
research has not been sufficiently applied in Translation Studies (see
Henderson 1987, Schweda-Nicholson 2005) and it is possible that it will
be able to provide a significant explanation of translation aptitude and
prospects for a successful career in translation. Leaving the research
area open to future investigation, let us follow the translator into the
stage when the first draft is finished and undergoes further revision.

Stage 5. Revision

As pointed out in chapter five the question of revision is complex and in-
cludes different procedures reflecting the different cognitive styles of
translators. Graph 25 shows that the majority of translators revise while
translating or at least are conscious of the need for further revision. In this
section I will focus on the post-drafting revision. The question directed to
professional translators was phrased in the following way, ‘When you fin-
ish the first draft do you?

a) read through it as soon as you finish

b) leave it and have a break before reading through it

The question was aimed to confirm or reject the common assumption that
it is better to leave a translation aside before revising it in order to look at
it with a fresh eye some time later.
310 Chapter 6

100%
87.50%
90%
80%
70%
60%
50%
40%
30%
20% 12.50%
10%
0%
Read through it as soon as you finish Leave it and have a break before reading
through it

Graph 25. Preferences for immediate or postponed revision among prac-

ticing translators.

The answers provided in graph 25 show that the majority of 87.5% prefer
to leave the translation aside after the first draft is ready. Some translators
left comments like, – time permitting, if possible, it depends on the dead-
line, unless it’s a very tight deadline. Three translators said that they do
both run through it as soon as they finish and leave it for further revision.
One comment was more extensive ‘as much I'd have liked to mark b),
also knowing from school this is the best thing to do, it was almost always
impossible due to deadlines/ time constraints’.
This comment shows the procedural awareness and the clash with real-
ity which frequently makes translators skip a more relaxed mode of work
because of time pressure. This, however, does not have to tax the quality of
translation especially for experienced translators as it is possible that they
have themselves worked out some procedures which safeguard the quality
of their translations. Still, the vastly preferred approach is letting some time
pass before revising the translation. This most likely indicates something
significant about the translation process from the point of the translator who
has been so much engrossed in it that he/she becomes, on the one hand,
mentally and perhaps also physically tired of the high concentration re-
quired by the language switch and the need to integrate knowledge all the
time (KIN). More importantly however, it is possible that at this stage the
 Empirical investigation into the development of … 311

experienced translator feels internally convinced that the valid so far frame
of mind (KIN) has to be re-set for the revision process to be productive in
terms of catching out mistakes of various degree both incidental, minor and
major. In other words, the mind has to be re-framed to carry out the task of
revising efficiently in order to guarantee the intended high standard profes-
sional translation. How translators revise their texts or what they actually
do is the next question to ask. The question, ‘When you revise your first
draft do you?’ gave three options:

a) check it with the original sentence by sentence, then read it as a new

text
b) read it as a new text only occasionally checking with the original
c) check it and correct without looking at the original

The answers provided by the respondents showed individual differences

in the favoured procedures chosen by translators during the stage of post-
drafting revision. The majority of 55% chose ‘b’, 30% went for ‘a’ and
the remaining 15% marked ‘c’.

60% 55%

50%
40%
30%
30%

20% 15%

10%

0%
Check it w ith the original Read it as a new text only Check it and correct w ithout
sentence by sentence, then occasionally checking w ith looking at the original
read it as a new text the original

Graph 26. Revision procedures applied by professional translators.

The results demonstrated in graph 26 are significant as they show the

point of detachment from the SL text during the revision stage, detach-
ment which is essential and necessary in the view of translation as a func-
tional purpose oriented activity. As discussed in chapter five the problems
312 Chapter 6

with detachment from the SL text are common for novice inexperienced
translators. The fact that the majority of translators analysed here said that
during their first revision they already read the TL text as a new text and
only occasionally check it with the SL text shows that they have reached
the level of self-confidence which does not make them feel the need to
check the new text sentence by sentence with the original although still as
many as 30% of the respondents check their translation for meaning by a
close sentence by sentence comparison of the SL text and its TL represen-
tative. Since for the majority of the practicing translators studied here re-
vision is marked by a complete or partial (55%+15%=70%) physical de-
tachment from the SL text some questions arise. Is the ability to abandon
the SL text only the matter of self-trust stemming from experience? Does
it mean that the abandonment is conscious and done to free the translating
mind from the SL text frame and the SL as a system of communication? Is
it possible that the translator with experience becomes aware of his/her
capacious working memory which holds the ‘mental imprint’ of the SL
text very fresh in the memory while reading the new TL text? Is it a sim-
ple rule, ‘out of sight out of mind’ and the SL text is consulted only if the
TL text shows problems while being read by the translator? All these
questions remain to be investigated. In terms of the hypothesized KIN,
however, it can be tentatively concluded that the preferred time lapse be-
fore post-draft revision indicates the need to re-frame the KIN used so far
and locate oneself only within the TL as a system and as a social practice,
suppressing at the same time the impact of the SL activation and possible
interference (see chapter 2 on language control in a bilingual mind). It
seems that the strong need for efficient meta-cognitive control makes the
abandonment of the SL text essential. A question arises, whether it is pos-
sible to develop the ability to re-frame and look at the translation draft
with a fresh eye without the time lapse which ensures a mental break and
the re-adjustment of the KIN before revising and creating a final version.
Would doing a different although simple cognitive task, like for example
solving a mathematical formula or making a shopping list help the trans-
lator to re-frame in a short time to perform an adequate revision? To con-
firm such speculations experimental research is needed conducted in co-
operation with experimental and cognitive psychologists.
Comments left by the translators in the study suggest that most likely
the revision stage is rarely a onetime activity but constitutes a process in
itself with a series of reading through the TL text and correcting to
 Empirical investigation into the development of … 313

maximize the communicative effect it is supposed to have on the reader

(see Hansen 2008). Some translators included comments pointing to a
series of revisions which they usually perform. The example below
shows a detailed procedure described in the comments to the multiple
choice options provided for the question, ‘When you revise your first
draft do you…?

a) check it with the original sentence by sentence, then read it as a new

text [TO CREATE FIRST REVISED DRAFT = 2nd draft]
b) read it as a new text only occasionally checking with the original [3rd
draft]
c) check it and correct without looking at the original [final draft, though
I try to do this twice at two different times]’

These detailed comments show the serial nature of revising carried out by
professionals who most likely have developed the ability to self-correct
relying on their own internal quality assessment procedures (the inner
critic).
The revision stage also has to be considered against the changing
working environment of the translator who is less and less a solitary
worker and more and more a member of a networking group where the
responsibility for the final version of the translation is spread over several
people, including apart from the translator, a reviewer and a proof-reader
as well as an editor (see Tabakowska 2003). How frequently translators
use other people as revisers or proof-readers was my next question.
When answering the question, ‘Do you have you work proof-read by
somebody else before you send it to a client?’, the translators had the fol-
lowing choice:

a) yes, always
b) only if I translate into a foreign language
c) occasionally
d) no
314 Chapter 6

The preferred answer was ‘c’ (67.5%) meaning that translators use proof-
readers only occasionally. The comments included explained that it is so
depending on the text and because it means extra cost. There were only 3
translators (7.5%) who have always had their work proof-read and 3
translators who marked ‘b’ admitting that they use proof-readers only
when they translate into their foreign language. Similarly a minor number
of 7 translators (17.5%) said that they do not have their work proof-read.
Graph 27 below shows the distribution of answers.

80%
67.50%
70%

60%

50%

40%

30%
17.50%
20%
7.50% 7.50%
10%

0%

Yes, alw ays Only if I translate into Occasionally No

a foreign language

Graph 27. Involving others in the revision process as editors and proof-
readers.

Some translators added that the translation agency they work for takes
care of it. One translator who said ‘no’ explained that it was for economic
reasons, though he considered it advisable or even recommendable. These
results show growing self-confidence which perhaps does not make trans-
lators feel the need to have their work checked before sending it to their
clients. It is also possible that they themselves have developed significant
skills in self-editing when they can objectively assess the conceptual and
formal quality of their translations. These skills then become essential
components of their human expertise in translation. Bearing in mind the
fact that as many as 72.5% admitted that they translate from their lan-
guage A into their language B though, there is some room for concern
about the quality of translations into the translator’s foreign language
(Korzeniowska and Kuhiwczak 1994, Korzeniowska 1998, Pokorn 2005).
To sum up, the analysis of the exemplar translator at work showed the
multitasking involved in producing a written translation. The work start-
 Empirical investigation into the development of … 315

ing with receiving a commission includes integrating conceptual knowl-

edge and activating lexical access which means that the skill to navigate
one’s internal memory is an essential part of the human ability to translate
at the level of expertise. As pointed out by the subjects knowledge inte-
gration is needed also to work out the translation brief which is not al-
ways supplied by the client. In this enormous knowledge integration task,
at least in written translation as opposed to interpreting, the translator’s
internal library (LTM) is aided by tools including dictionaries and data
bases, translation memories and spell-checkers. The decisive factor how-
ever even in the case of using CAT tools is the human factor and the hu-
man skill to chose relevant information and assess its adequacy to the task
at hand. Translating as any type of writing also includes revision and edit-
ing to prepare the TL text to function as a part of the TL reality while at
the same time meeting professional quality standards. All the six stages
analyzed take time and in the translating business time is money. The
question of time management is the next issue to consider.
Since the majority of my subjects were freelance translators and in
consequence their own bosses, they are, among other things, responsible
for their own time management. It turns out that it is not an easy task and
the majority of the practicing translators admitted that they would like to
improve their time management skills. When answering the question,
‘Are there any aspects of your work in which you would like to improve?’
the participants had a choice of four aspects:

a) time management
b) keeping deadlines
c) estimating time needed for translation
d) cooperating with clients

The majority of 60% said that they would like to improve their time man-
agement and a further 35% said that they felt the need to improve as far as
keeping deadlines and estimating the time needed for translation are con-
cerned. All the three aspects mentioned so far refer to time management
skills which include being able to estimate the time needed for translation so
that deadlines are realistic and allow for a sufficient amount of time needed
to translate a commissioned text without the need to overwork oneself. It
seems that this particular skill requires the same psychological qualities as
any other self-employed profession in which it is vital to strike a balance be-
316 Chapter 6

tween intensive mental effort when full concentration is required and the
mind is mobilising all cognitive resources, and the mind’s need to rest and re-
lax to replenish the resources needed for efficient work. Although as men-
tioned in section 6.2.2 professional translators are aware of the fact that trans-
lation is a high energy consumer (77.5%), they still are not fully satisfied
with how they manage time in their work and this indicates that there is room
for improvement in terms of self-control. One translator included advice on
time management quoted below:

focus on time management from the very start, try to specialize in se-
lected fields, use the Internet wisely, accept advice from professionals,
have your work proof-read at random, read original works in languages
A, B and C; follow cultural, political, economic events, etc.

So far I have analysed potential and real translators in terms of their ideas
about the profession, the activity, the bilingual foundations which they use
and the actually applied procedures by practicing professionals. Fre-
quently, I pointed to individual, highly idiosyncratic preferences to go
round certain tasks intrinsic to translation as a mental process. To make
the picture more complete in terms of the human factor which is involved
I also looked at the actual person performing the activity, the translator as
an individual with the overriding aim of finding confirmation for the abil-
ity to build KINs which according to my hypothesis put forward in chap-
ter five develops in the evolution of translation as a human skill to reach
the level of expertise. The final section of the present analysis is devoted
to the translator as an individual to make the invisible person (see Venuti
1995) for a change visible (see Tymoczko 2007).

6.2.5. The translator as an individual

Since every translator is a bilingual but not every bilingual will become a
translator the first question that needs to be asked is the motivating factor
which makes some graduates of language departments want to choose a
career in translation. The question directed to practicing translators gave
five options to the multiple choice question, ‘What made you choose
translation as your career (tick what you find appropriate)?’:

a) love for languages

b) financial aspect
 Empirical investigation into the development of … 317

c) I simply like to translate

d) I like to help others to communicate
e) I like to work with people

The answers given are represented by percentages in graph 28.

70%
60% 57.50%
60%
50% 42.50%
40% 30% 30%
30%
20%
10%
0%
I simply like to Love for Like to help Financial aspect I like to w ork
translate languages others to w ith people
communicate

Graph 28. The motivating factors behind choosing a career in translation.

The favourite reasons, ‘love for languages’ and ‘I simply like to translate’
were dominant in the choices marked by the translators. They point to the
relevance of affective factors which make a passion (love for languages) and
pleasure (I simply like translating) a significant reward for the work put into
developing professional competence and expertise in translation (cf. the
questionnaire study reported by Płusa 2007: 94). As I had already mentioned
the information hidden behind choosing, ‘I simply like to translate’ still re-
mains to be uncovered and explained possibly via personality research
(Henderson 1987, Schweda-Nicholson 2005, Kurz 2000, Zawada 2009) in
which attempts could be made to see which personality types are attracted to
the profession and which are not following the psychological principle that
people instinctively choose activities which agree with their naturally fa-
voured ways of doing things (Jung 1923). This could help to pinpoint the
necessary qualities which not only predispose people to become good trans-
lators but which make them happy to be translators. Although there is a long
list of requirements that translators should meet which was partially dis-
318 Chapter 6

cussed in chapter four, there has been a dearth of empirical studies which
would provide at least descriptive observational data to confirm how the
prescriptive qualities expected of translators match the reality. A translator’s
skills profile issued by the European Commission and discussed in chapter
four specifies the intellectual and psychological requirements. There is no
doubt that the demands are extremely high and not many graduates of for-
eign language departments will be able to meet them.
What image or profile of the real not ideal translator emerges from the
study conducted for the purpose of this analysis? Generally the image is
very positive which is straightforwardly reflected in the fact that as many
as 40 professionals responded to the questionnaire without any promise of
financial remuneration. They did so in response to an appeal by an aca-
demic who can only express her appreciation and gratitude for their pre-
cious time spared to provide answers to 50 questions. What is more the
majority of the respondents did not stop at ticking adequate answers from
the multiple choices provided but they frequently supplemented the cho-
sen answers with valuable and insightful comments which allow me to
draw some generalizations on their behalf.
Professional translators are undoubtedly people with high intellectual po-
tential. They see themselves as inquisitive individuals (82.5%), some of the
respondents added exclamation marks when marking ‘yes’ to the question,
‘Are you an inquisitive person?’, other comments included, ‘to a certain ex-
tent’, ‘sometimes’. Only one respondent marked ‘no’ and 5 marked ‘I’m not
aware of it’, and one translator added ‘it depends on the subject’. This means
that generally speaking translators are intellectually curious and keen knowl-
edge seekers, like to learn new things and are perseverant in their hunger for
making sense. These qualities are certainly what they need when translating.
Another question aimed at testing whether they consider themselves knowl-
edgeable people. As many as 29 translators (72.5%) said ‘yes’ only 2 said
‘no’ and 9 marked ‘I don’t know’ as the most relevant answer in their estima-
tion. There were three comments added by the respondents: ‘yes, to some ex-
tent of course’, ‘it depends on the area’, ‘I don’t know but I definitely know
where to look for the info I need it’. The 10% drop between seeing them-
selves as inquisitive and seeing themselves as knowledgeable is also interest-
ing and can be interpreted as a sign that the more one knows the more one
realizes how much more there is to learn and thus becomes more humble in
perceiving oneself as knowledgeable. Still the fact that as many as 72.5% of
the translators analyzed in this study consider themselves as knowledgeable
 Empirical investigation into the development of … 319

is possibly a result of their professional career in translation where as sug-

gested in chapter five a wide knowledge base is needed and it is constantly
extended and updated due to the on-line need to activate, integrate or acquire
new knowledge currently unavailable but urgently needed for the task at
hand. It was therefore not surprising to see as many as 90% of practicing
translators admitting in the questionnaire that they have special interests and
that the majority of them in fact have a wide range of interests.
Individual special interests and hobbies entertained by the respondents
included: buses, international transport, needle craft, fantasy literature,
linguistics, psychology, history, finance, politics, literature, forensic medi-
cine, Italian Renaissance art, climbing, flamenco, belly dance, horse rid-
ing, learning, water sports, motorcycling, tourism, anything connected to
J.R.R. Tolkien’s The Lord of the Rings, medieval chivalric culture, recon-
struction of historical events, tabletop games, role-playing games, sci-
ence-fiction and fantasy books, craft, dancing, traveling, philosophy,
spirituality, art, evolution (biology), astronomy, geology, mathematics,
choir singing, diving, cave diving and related science, photography, sci-
ence, ballet, philosophy, architecture, jogging.
The answers provided by the practicing translators are summarized in
graph 29 below showing a cognitive profile of a professional translator.

100% 90%
90% 82.50%
80% 72.50%
70%
60%
50%
40%
30%
20%
10%
0%
Inquisitive Know ledgeable Wide range of
interests/hobbies

Graph 29. Cognitive profile of the professional translator based on self-

assessment.
320 Chapter 6

Focusing more on language related activities as many as 87.5% of transla-

tors admitted that they are keen readers. The vast majority of translators
(82.5%) said that they were considered good writers at school, only 15%
felt average and one translator admitted to having some problems with
writing at school. Trying to probe their analytical skills which are men-
tioned as important in translation I also asked the professionals whether
they were good at mathematics during their schooling. The results showed
that only 52.5% of the subjects were good at mathematics, 35% said that
they were average and 12.5% admitted to having some problems.
This possibly points to a strong language aptitude (Carroll and Sapon
2002, Pimsleur et al. 2004) bias in the group studied in this analysis there-
fore digging further into the intellectual qualities of professional practic-
ing translators I asked about their language related behaviour and more
precisely about their metalinguistic awareness. The question, ‘Do you pay
attention to how other people use language?’ was aimed to test whether
being conscious users of language/languages themselves they are obser-
vant and perhaps more sensitive to how others use language in communi-
cation. The results showed that 100% of the professionals admitted pay-
ing attention to how others use language, two respondents added a com-
ment, ‘very much’, ‘very much so’. The graph below shows the summa-
rized language related profile of the professional translator.

120%
100%
100% 87.50%
82.50%
80%

60%

40%

20%

0%
a keen reader a good w riter pays attention to how others
use language

Graph 30. The language-related profile of the professional translator.

 Empirical investigation into the development of … 321

The 100% result is highly significant and supportive of the role metalinguis-
tic knowledge, ability and awareness plays in the development of translation
as a human skill to reach the level of expertise. Such a unanimous response
is the first one noted in all the answers analyzed so far in this study. A possi-
ble explanation includes the need to see language not as knowledge but as
social practice, as it is currently used by its many speakers/writers. Observ-
ing and paying attention to how others use language is perhaps a means to
objectify the way a translator as an individual uses language and to verify
one’s own manner of using it bearing in mind that translations are always
performed for other people, other language users whose communicative ex-
pectations may differ from those subjectively held by the translator. This
awareness of the translation addressee is something which is difficult for
EFL students who start to practice their translation skill as well as for novice
translators who enter translator training. Many translation teachers would
probably agree that novices in translation tend to treat their own idiolect as
common social practice and frequently argue that a given word can be used
in such and such a way. These multiple voices in the translation classroom
(see González Davies 2004 for more ideas) sometimes develop into amusing
and revealing discussions about language use, issues of register and regional
differences surface as students usually come from all over the country and
argue that for example, ‘mashed potatoes’ can be translated into Polish as,
tłuczone ziemniaki, ubite ziemniaki, śturchane ziemniaki, puree, etc. Such
discussions and metalinguistic debates stimulated by translation practice en-
courage students to question their choices even in their native language and
they frequently start to ask themselves and others, ‘Can you say that in Pol-
ish?’ It seems that the metalinguistic inquisitiveness confirmed by the practi-
tioners shows that indeed translating makes translators get to know in depth
not only their foreign language but their native language as well. This ob-
servation echoes the views quoted by Kesckes and Papp (2000) when stu-
dents frequently observe that the best way to get to know your native lan-
guage is to learn a foreign language. By analogy it can be said that the best
way to get to know your two languages is to translate. Consequently the
most likely relationship between bilingual knowledge and translation skill is
that of mutual benefit. The fact that translators so unanimously admitted
paying attention to how others use language might show that in this way
they update their internal data base of possible expressions, word combina-
tions which are also a function of the changing times, affected by current
events, technological advances and linguistic fashion. Since translating
322 Chapter 6

makes translators be more language observant when it comes to other

speakers/writers, a related question seemed whether translating as their ca-
reer has changed the way they use language themselves. Indeed, as many as
82.5% said that translating has changed they way they use language which
as compared to only 10% of respondents who said ‘no’ and 7.5% who said
that they were not aware of it is a statistically valid result.

90% 82.50%
80%
70%
60%
50%
40%
30%
20%
10% 7.50%
10%
0%
Yes No I’m not aw are of it

Graph 31. The impact of translation as a profession on the way translators

use language.

In a way to find out that professional translators are highly sensitive lan-
guage users is not surprising as it is what is professionally expected from
them (Nida 2002). Since, however the professional translators who took
part in the study formed a group with varied professional backgrounds,
diverse kinds of translation in which they specialise as well as the fact
that the length of time for which they have provided translation services
on the market stretched from 1 year to 40 years, the finding is significant
and it confirms the importance of metalinguistic awareness and ability in
the process of the evolution of translation as a human skill.
So far the emerging image of the translator as an individual has fo-
cused on their general cognitive make up and language-related skills.
Translators appear to be knowledgeable people, intellectually curios, with
versatile interests and hobbies able to skilfully use language and are con-
scious of its impact as well as observant on how it is used by others. It is
virtually impossible to say whether these qualities had been granted to
 Empirical investigation into the development of … 323

them prior to their translation experience or whether they appeared as a

consequence of their experience in translating. One can only assume that
probably both kinds of aetiology are involved. The descriptive profile
sketched out on the basis of the questionnaire study discussed so far
matches the requirements specified for the European Master’s Degree in
Translation discussed in chapter 46.
As it can be gathered from the study translation is not a career for eve-
rybody (Hejwowski 2004), but as the data so far show those who have cho-
sen it seem to enjoy it. It is definitely a career for versatile personalities,
people with a wide range of interests, intellectually curios who like the
multitasking involved in activating and integrating all kinds of knowledge
needed for the task at hand. The question how to train or educate (De
Beaugrande et al.1998, Klaudy 2003) such professionals is continuously
discussed and debated (Cronin 2005, Kiraly 2006, Gile 1995/2009). What
might seem interesting is the fact that as many as 21 of the translators who
participated in this study have not had any formal training as translators.
What is more as many as 22 (55%) have not attended any workshops for
translators. Indeed, when I asked the translators to mark their answers to
the question, ‘How did you acquire your professional expertise?

a) translation training programme

b) self-development only
c) mentor who shared his/her professional experience with you
d) through professional work and self-reflection

the distribution of answers showed that as many as 38 translators (95%)

marked d) admitting in this way that their translation expertise has been
acquired through professional work and self-reflection. Other choices
were selected by fewer translators, 10 marked ‘a’, 12 marked ‘b’ and 13
marked ‘c’. The results are shown in Graph 32 below.

–––––––––
6
For more details visit the website: http://ec.europa.eu/emt.
324 Chapter 6

100% 95%
90%
80%
70%
60%
50%
40% 30% 32.50%
25%
30%
20%
10%
0%
translation training self-development only mentor who shared through professional
programme his/her professional work and self-reflection
experience

Graph 32. Different paths to translation expertise.

The choices marked by practicing translators show something significant

about translation as a human skill and its developmental potential. First of
all, it has to be acknowledged that translator training institutions in Po-
land are a fairly recent phenomenon (Płusa 2000). However, this high
percentage of self-trained successful professionals confirms the validity
of the concept of the translator as an expert learner discussed in chapter
five. Those individuals who are able to follow this path of professional
development can indeed become good translators without formal training
in accordance with the definition of translation competence as experience-
derived knowledge (Toury 1995, Kiraly 1995, 2005, Chriss 2006,
Gouadec 2007, Sofer 1997/2006) internalized and made productive
through practice and conscious self-reflection (Donald 2007b)
These results confirm that the psychological epistemology of translation
skill is best located in constructivism7 according to which humans acquire
expertise through experience (Kiraly 2000, Klir 2001). Today’s translator
training programmes are acutely aware of the need to provide trainees with
real life experiences (Kiraly 2005, Kearns 2008) and the role of self-
–––––––––
7
According to constructivism, all systems are artificial abstractions and they are not
made by nature to be discovered by us, but we construct them by our perceptual and mental
capabilities with the domain of our experiences (Klir 2001). This view is compatible with
the perception of a translator as a systems intelligent person (see chapter five).
 Empirical investigation into the development of … 325

reflection is stressed by key scholars in TS (Gile 1995/2009, Cronin 2005,

González Davies 2004, Kelly 2005). Self-reflection undoubtedly enhances
self-learning but it also amplifies the experience of KIN performed for a
translation task, strengthens memory traces which will be left by every
translation experience and used in future when the translator assesses
his/her performance in terms of quality and economy. These traces will be-
come stored in the personal memory archives and if needed in future will
be recalled to make the KIN perhaps more efficient. Despite the different
paths to translation expertise, the majority (67.5%) said that they would be
interested in attending workshops for translators and a further 30% said that
they would have been interested in such workshops when they were start-
ing their career but not now. This gives a decisive majority who are inter-
ested in expanding their expertise through training with only 1 out of 40
translators who was against it. A quote from the recently revised edition of
Daniel Gile’s (2009) seminal book, sums up the discussion on different
paths to translation expertise and translator training:

Formal training is not mandatory, but it can perform at least two impor-
tant functions. One is to help individuals who wish to become profes-
sional interpreters and translators enhance their performance to the full
realization of their potential. The other is to help them develop their
translation skills more rapidly than through field experience and self-
instruction, which may involve much groping in the dark and learning
by trial-and-error (Gile 2009: 7).

Following the translator as a developing individual irrespective of their

career path I asked the respondents to reflect upon their professional ca-
reers and self-assess which aspects have positively changed with their ex-
perience. The possible answers included the following:

a) the time it takes to translate

b) your network of clients
c) your attitude to language
d) the size of your mental dictionary
e) your self-confidence
f) the way you use dictionaries
326 Chapter 6

Each translator could choose as many aspects as they felt valid in their
case and most marked at least two or three aspects while several marked
all of them. Graph 33 below shows the distribution of these aspects which
were marked by translators as having changed positively with time.

85%
90%
80% 72.50%
70% 62.50% 62.50%
57.50%
60%
50% 42.50%
40%
30%
20%
10%
0%
the time it the netw ork your attitude the size of your self- the w ay you
takes to of your to language your mental confidence use
translate clients dictionary dictionaries

Graph 33. Aspects which positively changed with experience.

As the results confirm acquiring expertise in translation has an impact

on the translator as an individual. As many as 85% felt that the size of
their mental dictionary has grown with their experience, and as many as
72.5% marked a significant increase in self-confidence. The self-
reported awareness of the effect of translation practice on the size of the
translator’s bilingual mental lexicon stimulates further questions refer-
ring to the functional organization of the internal vocabulary store in
people who provide translation services. Unfortunately, the effect of
translation practice on bilingual memory remains an under researched
domain although quite recently some new studies have addressed this is-
sue in conference interpreters (see Tymczyńska 20118). Wilss (1999:

–––––––––
8
Tymczyńska (2011) conducted a study of lexical access in trilingual interpreters,
interpreting trainees and trilinguals without interpreting practice and reported that
against her expectations the group of practicing interpreters (employed by the EU) did
not perform better in terms of reaction times but in terms of accuracy, lack of omissions
and errors the professionals outperformed the trainees and trilinguals. Trainee interpret-
ers did not perform better than trilinguals in all three categories which led to the conclu-
 Empirical investigation into the development of … 327

220) as early as in 1999 observed that in general memory research had

not attracted much attention from TS scholars. Yet, it is the translator’s
and interpreter’s most precious tool. As it was discussed in chapter two
and hypothesized in chapter five it is plausible that decisive changes in
the functional organization of bilingual memory take place (and have to
take place) so that the translator can navigate his/her memory to allow
for efficient knowledge integration (KIN) and fluent interlingual lexical
access (De Groot 2002, Presas 2000). The need for language control (see
Green 1993, De Groot and Christoffels 2006) driven by the demand to
have both languages simultaneously active yet kept clearly apart in the
actual articulation of the translation calls for efficient meta-cognitive
supervision to filter out unintended language interference (Whyatt 2010,
Tymczyńska 2011).
Apart from the size of the mental dictionary, the highest scoring aspect
which has changed with years of practice was the feeling of increased self-
confidence (72.5%). In chapter five I discussed the postulate that the proc-
ess of acquiring expertise in translation perceived from the perspective of a
translator as an expert learner can be viewed as a process of developing
self-confidence. Obviously the same principle applies to other complex
skills which are developed to the level of expertise. With time and practice
and conscious self-reflection many constituent actions become procedural-
ised (see Squire and Kandel 2009), adapted to the individual cognitive
make-up of the user and encoded in the form of algorithms later used by
memory to trigger off activation and integration of knowledge needed for
complex tasks like translation, or creative writing, or playing an instrument.
Bartlett (1932/1967) would say that the repertoire of schemas is growing
and since they are frequently used it is much easier to activate tasks sche-
mas (Price and Green 1999), and the activation and integration is done with
more self-confidence and possibly also faster. I asked the question about
self-confidence more directly in the questionnaire to further see whether the
progress in the level of self-confidence is something translators are aware
of. As many as 95% of the practicing translators noted that they felt more
confident when translating at present than when they were beginning their
professional careers. Only 2 out of 40 translators said that they were not
aware of feeling more confident in comparison to how they felt at the be-

sion that the two years of training are not enough to result in changes in the functional
organization of their mental lexicons.
328 Chapter 6

ginning of their professional careers. One translator added a comment, ‘yes

(incomparably; should I ever knew what I was going to undertake…)’.
There are still a lot of questions that one would like to ask and which are
frequently asked by translation students including the most popular one,
‘How long does it take to become a self-confident, competent translator?’.
What do translators gain by becoming more self-confident? Does the fact
that translators have more self-confidence mean that they translate faster?
The correlation between the time needed for translation and experi-
ence was marked by as many as 62.5% which allows to conclude that for
the majority of translators the translation process takes less time as a con-
sequence of better memory navigation, faster lexical access and more
self-confidence about how to solve problems and make appropriate deci-
sions. It is not without reason that translation has been compared to a
game of skill (Gorlee 1994). The more you play it the better prepared you
feel for the next one, but each next one is a new challenge. One translator
included an interesting comment, ‘if necessary, I can translate faster than
before, but under normal circumstances translating is just as time consum-
ing as it was’. The temporal aspect of translation performance has rarely
been studied in controlled experiments (Jensen 1999, Jensen and Ja-
kobsen 2000) although it is commonly assumed that time constraints and
time pressures will tax the translators’ performance in both written and
oral translation (see De Groot 1997).
Among other aspects which have been marked as positively affected
by experience in translation 62.5% of the respondents marked that their
attitude to language has positively changed with experience in translating
which allows to assume that all the components of their communicative
competence have become integrated and they know more about both lan-
guages and become more in control of them, more aware of themselves as
conscious language users. The network of clients was also marked by a
slim majority of respondents as positively affected by translation practice.
The way translators use dictionaries was the least positively assessed as
changing parallel to experience.
Having elicited the changes that translators see in themselves as a
product of their career in translation I asked the professionals to mark
these aspects of translation as a profession that they would like to see
change. When answering the question, ‘Are there any aspects of the pro-
fession in which you would like to see improvement?’ the translators had
the following choice:
 Empirical investigation into the development of … 329

a) more understanding from clients

b) more social awareness of what is involved in translation
c) more concern for the quality of translation
d) more respect for the job
e) better financial conditions

The answers marked by those who provide translation services are dem-
onstrated in graph 34 below.

80% 70%
65%
70%
60% 50%
45%
50%
40% 27.50%
30%
20%
10%
0%

more more social more concern for more respect for better financial
understanding aw areness of the quality of the job conditions
from clients w hat is involved translation
in translation

Graph 34. Aspects of the profession which need improvement.

The results showing the need for improvement in the social concern about
the quality of translation reported by 70% of the translators are significant
and reveal a lack of understanding and respect for the quality of translation
as a final product. The need for more concern about the quality of transla-
tions is generally admitted by TS scholars, but the policy of controlling the
quality is virtually non-existent especially in countries where no accredita-
tion is required to provide translation services. As pointed out by Chriss
(2006: 141) there is a lot of asymmetric information about “which transla-
tors are worth hiring and working with and which should be avoided at all
possible cost”. On the other hand, translators have to prove to their clients
that they are professionals who can do their work reliably and responsibly.
Considering the growing need for translation services of high quality much
is left to market forces. Gouadec (2007), for example is of the opinion that
competition will get rid of the ‘outlaws’ and survival on the market is the
best indicator of the quality that is provided by individual translators. If
330 Chapter 6

however, we accept that it does not matter if translation is a craft or an art

when in today’s world it is primarily a business we have to say that it is a
jungle out there (Wilss 1996, Gouadec 2007). Competition means that fre-
quently the clients are not interested in high quality products (Nida 2002)
because they cannot themselves tell a bad translation from a good one since
they do not know the foreign language involved (cf. Durban 20039). They
are mostly interested in how quickly they can get it and of course at the
lowest possible price they will have to pay for the services. A concern for
quality is expressed in the guidelines and recommendations issued by asso-
ciations of translators and interpreters on a national and international level
(FIT, STP, TEPIS). These organizations stress that concern for quality
shows respect for business ethics and the translators’ awareness that they
bear responsibility acting as “the communication conduit for a product or
service, for information or opinion, and so must consider the consequences
of their linguistic decisions” (Chriss 2006: 140). The issues of translation
quality and professional ethics are complex and require a broad discussion
which exceeds the intentions set behind this volume. The fact that as many
as 70% of the practicing translators wish that the concern for the quality is
improved points to the need to educate society about translation quality and
its relationship with adequate skill and expertise.
Indeed, the need for more social awareness of what translation involves
is the next high scoring aspect of the profession after the quality issue. As
many as 65% of the translators expressing the need for more understanding
confirm that translation as a skill is socially poorly understood (see chapter
1). This lack of understanding is reflected in frequently unrealistic demands
from clients as far as the time needed for translation is concerned. Further-
more, those who seek translation services are unaware of the expertise
which is needed to produce an adequate translation and as a result they fre-
quently approach language students or anybody with some knowledge of
the foreign language and expect an adequate translation. As some EFL stu-
dents have reported if those who seek translation services are told by the
students that it is too difficult for them to translate a given text, the L2 stu-
dents get an impression that their refusal to translate a given text is inter-
preted as ‘obviously their English is not good enough’. In effect many
translations available on the open market are performed by unskilled trans-

–––––––––
9
Durban’s (2003) brochure, Translation: Getting it right – A guide to buying trans-
lations is available on the website of FIT – The International Federation of Translators.
 Empirical investigation into the development of … 331

lators who rely on their natural ability to translate which as discussed in

chapter three is characterized by formal flaws (linguistic and/or conceptual)
which may affect the interpretation of meaning by TL receivers. It seems
that more social awareness of what is involved in translation is a very much
desired improvement and a challenge for Translation Studies. Some schol-
ars observe that the so much needed profession is still undervalued (Chriss
2006, Snell-Hornby 2006, Gouadec 2007). Educating society and commu-
nity on a global and local level about the developmental nature of transla-
tion as a human skill might ultimately be hoped to build due respect for the
profession. The question how to do it requires a complex insight, at this
point in the discussion presented so far, I am inclined to claim that the edu-
cation should start with foreign language students at university level as they
are the most likely potential translators. During their language studies they
should have a chance to receive at least a taste of the skills10 which are
needed to produce a translation of adequate quality and with appropriate
feedback on how they are coping with the task of translating as well as
some guidance as to what they have to learn if they want to pursue a career
in translation. Exposure to translation tasks during language studies at uni-
versity level is needed not only to raise awareness of translation as a proc-
ess but to inform students (potential translators) about professional ethics
and to entail a process of self-selection of those who will in future take up a
career in language translation. Having acquired some insight into the pro-
fession then as graduates they might be more informed and better able to
judge whether or not meeting the requirements of a translation commission
is within the scope of their abilities. They might be more likely to send the
client seeking translation services to a professional translator rather than
take up a task for which they do not have sufficient expertise.
To sum up, professional translators who shared their experience and
self-reflected on their own process of development come across as inter-
esting individuals who are very much aware of their progression from
language learners, language users to professionals who provide translation
services for those who are divided by language barriers. The advice they
shared shows the centrality of language related skills and the respectful
attitude towards language as a tool. A professional approach to translating
highlights the importance of self- management referring to the individual

–––––––––
10
see Witte et al. 2009 on the popular policy to drop translation courses in modern
language departments.
332 Chapter 6

translation tasks and the entire business of earning a living through trans-
lating. The most significant, however, is the awareness of the ongoing
process of development in their human ability to translate which necessi-
tates constant self-learning to maintain and enrich the available knowl-
edge base. Considering the above analysis of the translator as an individ-
ual it seems plausible to assume that the development of the human skill
to translate relies on the developing ability to integrate knowledge and use
it for the efficient functioning of the human translator.

6.3. Conclusions

The purpose of this chapter was to provide descriptive data about the real
agents involved in the journey along the continuum in the process of trans-
lation skill development, the potential translators and the real practicing
translators. The data discussed here confirm that translation as an activity is
idealised by those who lack experience, but this is possibly true of any
other complex human skill which when watched by an observer gives the
impression of ease, smoothness and harmony. The clash between these un-
informed ideas about translation as an activity and the transformation in
these ideas as a result of experience was visible in the students growing
awareness of the complex nature of translation as a skill which involves in-
tegrating knowledge from different domains and from different memory
storages (declarative and procedural). A drop in the percentage of the EFL
students who see themselves as future translators when they enter univer-
sity and those who still entertain this possibility when they are just about to
leave seems significant. It is possible that choosing a career in translation is
open to self-selection and with time and experience EFL students discover
for themselves whether or not the requirements of the work involved in
translation match their individual cognitive make-up and their personalities.
Personalities which contain such qualities like intellectual curiosity, inquisi-
tiveness, high level of self-control, a wide range of interests, quick think-
ing, patience and perseverance as well as readiness to learn (reported by
1BA students and professional translators) might be more inclined to be-
come translators. Others opt for different professional careers. It is possible
that in terms of personality research (Schweda-Nicholson 2005: 124) there
is much to be discovered about the cognitive prerequisites which are impor-
tant and might predict the success or failure in developing the human skill
to translate to reach the level of expertise. These qualities might be relevant
 Empirical investigation into the development of … 333

for designing translation aptitude11 tests which could help to select the most
promising candidates for translation training programmes more reliably
than the frequently used L2 proficiency tests. Interestingly, more EFL stu-
dents seem to prefer written translation and only few see interpreting as
their first choice. Although practicing translators also seem to favour writ-
ten translation, nearly half of the PT subjects said that they are ready to pro-
vide oral as well as written translation. This shows that the pronounced dif-
ferences between the processes involved in written translation and inter-
preting are not so important for practicing professionals. This result can
send a message to translator training methodologies that both modalities
share a common store of knowledge about translation, and practice in both
modalities can prove beneficial for written translators and interpreters alike.
The analysis of ideas about translation as an activity provided a lot of
insights into the nature of the processing involved in translation perform-
ance. The similarities between all the groups of subjects were significant
in their assessment of translation as an activity and they included the per-
ception of translation as a high energy consumer. These observational data
confirm that the task of translating uses a significant amount of mental re-
sources and its tiring effect is experienced by the more and less experi-
enced translators. Although translation is attested to be intellectually
highly demanding, it is also perceived as enjoyable as if the translator was
indeed involved in an intellectual game when overcoming major and mi-
nor hurdles is a part of the translation process. This highly affective atti-
tude is confirmed by the practicing translators who in their majority ad-
mitted that they treat every new text as an intellectual activity and a chal-
lenge. The intellectual benefit of learning new things is the primary rea-
son which keeps translators in the profession. The generally positive atti-
tude to translation as an activity is an asset not only for translator trainers
but also for L2 teaching and learning methodologies, an asset that should
no longer be ignored as translation is socially expected of bilinguals and it
can be used as a “hook to amplify pride in bilingualism” (Malakoff and
Hakuta 1991: 163) on the one hand, and make L2 learners and users
aware of their intercultural position.

–––––––––
11
Research on language aptitude tests could be of help here although there are still
disagreements among SLA scholars on those features which are the most relevant to
predict success in L2 learning (see Carroll 2002, Pimsleur et al. 2004, Rysiewicz 2008).
334 Chapter 6

A significant amount of attention in this chapter has been devoted to

the bilingual foundations of the subjects who represent potential and prac-
ticing translators. Since in translator training TC is built on these founda-
tions it is my belief that as much as possible should be known about their
structure, organization and functioning before a pedagogical effort is
made to build professional translation expertise upon them. The results
show that even when the L2 proficiency is very high (e.g., among the
2MA subjects) there is still a fair developmental route to cover in order to
master the human skill to translate with confidence to meet professional
standards of quality. This probably indicates that L2 learners and users
have to develop high awareness of themselves as bilinguals who can use
their L2 and, what has been taken for granted, their L1 knowledge to
translate. The lack of self-confidence as translators reported by the 2BA
and 2MA groups points to the monolingual bias of L2 teaching method-
ologies which aim at L2 communicative competence without taking into
account the learners’ L1 as the language of primary enculturation. The
2BA and 2MA subjects admitting their concern for their L1 as a result of
translation experience show that some qualitative changes have to take
place in the organization of bilingual knowledge to allow for adequate
and later on efficient translation. This awareness can be taken as evidence
for the process of knowledge integration that is necessary to translate with
competence and confidence. At the same time being aware of the lan-
guage imbalance is an expression of one’s metalinguistic awareness.
The role of meatlinguistic knowledge which develops parallel to
growing L2 proficiency is possibly very important for the development of
the human skill to translate. Here as in the case of Roehr’s (2008 quoted
in Paradis 2009: 7) study the relationship between metalinguistic knowl-
edge and the human skill to translate is that of reciprocal benefit. In fact
the results confirm that the experience of translation has more than a rip-
ple effect on bilingual knowledge including L2 proficiency. The results
reported by the group of practicing translators decisively confirm that
translators are expert language learners in their lifelong learning process
seen as the evolution of their self-confidence as conscious language users
and translators. The tacit assumption that translators are or should be
equally proficient in both working languages is challenged by the data
showing that certain aspects of L2 use remain problematic for well ex-
perienced practicing translators. Translation by its integrative nature and
the requirement to activate and integrate knowledge (KIN) provides a
 Empirical investigation into the development of … 335

window on how bilingual minds specialise in interlingual and intercul-

tural communication. The proposal of KIN discussed in chapter five
seems to provide the framework in which the relationship between lin-
guistic, metalinguistic and extralinguistic knowledge can be investigated
in all those who perform translation irrespective of their stage at the de-
velopmental continuum of translation as a human skill.
The integrative nature of translation as an activity in professional prac-
tice has been extensively analysed in section 6.2.4. The results seem to con-
firm that every new translation task sets off a process of knowledge integra-
tion that was hypothesized in chapter five. The process starts off at the first
encounter with the SL text, and it is not as wished by translation peda-
gogues always facilitated by a translation brief. The first reading of the SL
text seems to activate the translator’s prior experience which allows assess-
ing the level of difficulty of the task at hand as well as getting the gist of the
content and positioning it against previous experience in translating similar
texts. This initially created knowledge integration network (KIN) is further
developed by some factual research which was frequently reported by
translators to be prior to the onset of the physical production of the TL text.
Looking at similar TL texts and refreshing TL vocabulary was also reported
as a part of the preparatory activities before the actual translation process.
These procedures strongly indicate the need for knowledge activation and
integration before the translator feels ready to start typing the TL text. The
fact that the favoured preparatory activity was doing some factual research
points to the primacy of conceptual activation and integration which as hy-
pothesized in chapter five facilitates lexical access when the TL text comes
into existence. The intellectual curiosity and the frequent need to acquire
new knowledge in order to create an adequate knowledge integration net-
work (KIN) is perceived as a challenge and a reward of the profession. The
ability to search for required information or missing words becomes an im-
portant subskill which relieves the internal memory of the translator who
can turn to the knowledge stored in the external memory archives including
dictionaries and electronically available data bases. The use of the Internet
as a ‘reference shelf’ of collective knowledge, both factual and linguistic, is
symptomatic and exceeded the use of electronic dictionaries in case of the
practicing translators studied in this analysis. The possibilities of exchang-
ing ideas and sharing translation problems with others include translators’
internet forums which were reported to be used by the majority of transla-
tors. Translation memories and other CAT tools were used by the minority
336 Chapter 6

of the respondents which can suggest that they are not judged as signifi-
cantly facilitating the translation process in case of the professionals stud-
ied in this analysis. This can be interpreted in two ways, the generation of
today’s practitioners is not fully aware of their potential or the potential is
not as good in practice as it may seem to be in theory in terms of facilitat-
ing the knowledge integration processes in the translating mind. These is-
sues will have to be accounted for in further research. A similar matter of
concern is revision as an integral part of the translation process which
might require some additional skills. The data reported by practicing trans-
lators point to the need to re-frame from the initially created KIN to ensure
adequate revision leading to the final version of the TL text. Whether or not
the re-framing always requires a temporal (temporary) break from the al-
ready produced translation is a matter which needs further investigation
with reference to the quality of translation as a final product. It seems that
ensuring adequate revision and involving others as revisers and proof-
readers also requires attention from translator trainers especially in the case
of translating into the B language (see Pokorn 2005). The fact that transla-
tors see themselves as open to improvement is undeniable.
The final section devoted to the translator as an individual aimed to pro-
vide an empirically valid personal profile of those who have chosen to pursue
a career in translation. The results show that the translator is best perceived
as a developing self. When asked about the reasons to become translators the
vast majority pointed to affective factors (I like to translate, the love for lan-
guages, I like to help others to communicate) which show that the profession
is driven by passion and devotion to the profession which is bound to raise
morale and job ethics. The cognitive profile of the practicing translator con-
firms care and attention to maintain and up-date one’s knowledge base
through intellectual curiosity and a wide range of interests/domains in which
one can specialize. The language-related profile shows the primacy of lan-
guage maintenance needed for the job. The role of metalinguistic and meta-
cognitive abilities is yet again confirmed by the subjects admitting that their
experience in translation has affected the way they themselves use language.
However, the fact that they do not see themselves as language experts but
just as conscious language users is symptomatic as it allows the assumption
that they also become more aware of the magnitude of knowledge involved
in language as social practice. The answers and comments provided by prac-
ticing translators seem to confirm that the development of translation as a
human skill can be viewed as a process of developing self-confidence (sys-
 Empirical investigation into the development of … 337

tems intelligence) in the way bilingual knowledge can be used for cross-
cultural communication. The evolution of translation as a human skill in-
volves learning to integrate all the knowledge needed for the task at hand.
This ability cannot be expected of untrained bilinguals who perform transla-
tion but develops through practice and experience coupled with self reflec-
tion. Yet, this integrative effort is however socially under-estimated and as
pointed out by practicing translators raising the social awareness of what is
involved in translation and more concern for the quality of translation is
needed to grant the professionals a socially respected position they most cer-
tainly deserve. The fact that 95% pointed to experience and self-reflection as
the primary source of their professional expertise is significant and sends a
clear message to translation pedagogy to provide an appropriate learning en-
vironment and encourage self-reflection in translation trainees. The question
how to teach knowledge integration needed for translation tasks is open to
suggestions. It is possible that the time consuming translation process could
be dismantled and individual stages could be practiced before students can be
expected to perform entire professional translation projects following the
principle that complex skills consist of simple subskills (see De Groot 2000:
53 for a componential approach)12.
The major question remaining is whether the ability to integrate knowl-
edge develops as a slow process enhanced by professional experience or
whether it is something we can explicitly teach speeding up the otherwise
slow process. This question is taken up in chapter 7 which describes 2 ex-
periments in which the hypothesized ability to integrate knowledge as a de-
cisive constituent of translation expertise is put to the test.

–––––––––
12
A book by Gonzaléz Davies (2004) is a rare but an excellent attempt to provide
translation trainers with a pool of practical tasks which can be used to help trainees to
practice knowledge integration involving the conceptual and verbal level.
Chapter 7

Empirical investigation into the development

of a Knowledge Integration Network in
translators – part two

The purpose of this chapter is to search for further empirical validation of

the hypothesis that the development of translation expertise relies on the
growing ability to create Knowledge Integration Networks relevant to the
translation task at hand proposed in chapter 5 and confirmed by descrip-
tive research data discussed in chapter 6. The twofold analysis presented
in this chapter has different aims. The first study aims to investigate the
performance of translators who are at different points on the developmen-
tal continuum in the process of translation skill development and who
were asked to translate the same text. The second study describes an ex-
periment designed to test a hypothesis that creating a KIN before the on-
set of the translation process will have a positive effect on the perform-
ance of inexperienced translators. Both studies investigate the perform-
ance of translators during the process of written translation production
and they share the same research method. The chapter will open with the
description of the research method called Translog which was used in
both studies. Then both studies will be described and the results will be
discussed in order to further validate the hypothesis that the human ability
to translate to reach the level of expertise involves the ability to integrate
knowledge in the translation process. The chapter will close with conclu-
sions and implications for translation pedagogy.

7.1. Research method

Translog is a computer software program which was designed by Arnt

Lykke Jakobsen, his son Lasse Schou (Jakobsen and Schou 1999, see Ja-
kobsen 2006 for the program’s history) and his colleagues from the Co-
penhagen Business School. The program records all the keyboard activity
performed during a TL text production. In essence it is similar to key log-
340 Chapter 7

ging (or keystroke logging) software programs used in the study of text
production in creative writing research (see Sullivan & Lindgren 2006 for
an overview of the research method and its applications). Lauffer (2002)
mentions that before Translog was available cameras were used to ob-
serve the computer, mouse and keyboard activity of translators while they
performed a task. All the data was later transcribed into detailed logs. As
expressed by Lauffer (2002: 63), “Translog was developed to obtain
quantitative reinforcement of assumptions about translation”. This quanti-
tative reinforcement was very much needed to externalize translation per-
formance in a less invasive method than Think Aloud Protocols. Using
Translog to investigate the translation process the translator is not asked
to do anything but translate a text. The SL text is visible in the top part of
the computer screen and the translator types the TL text in the bottom part
of the screen which means that the SL text is all the time available to the
translator. While being totally unobtrusive to the translator the program
records all the keyboard activity into a file (Translog file) which can be
later played back and analyzed in detail. As expressed by Jakobsen (2006:
95), “Using the computer to record a translator’s keystrokes provides a
window onto the process” with the main assumption that “the temporal
course of the typing process is a recordable and observable reflection of
underlying cognitive processes” (Jakobsen 2006: 96). The research
method inspired by Schilperoord’s 1996 publication entitled, It’s about
time. Temporal aspects of cognitive processes in text production can be
applied to study the interaction of the cognitive processes by analyzing
the temporal patterns of text production and pauses in the flow of typing
which reveal the dynamic processing involved in text comprehension and
text production not only during the translation process but also in other
monolingual or multilingual tasks. In a way it is possible to call Translog
a keyboard spy (cf. Camtasia as a screen spy) as it records the file and al-
lows to see something that is otherwise never visible in a finished transla-
tion as a final product without any intrusion into the translation process as
such1. The translator who is translating a text in Translog is not disturbed
in any way in his/her creative translation process.

–––––––––
1
It is possible to compare Translog to eye-tracking programs which record eye fixa-
tion and movement to infer the focus of attention and information processing in the ob-
server or reader of a text (Rayner 1998, Wright and Ward 2008). Recent developments
include a combination of eye tracking with Translog (Jakobsen 2011).
 Empirical investigation into the development of … 341

The Translog program has two independent but interconnected com-

ponents, Translog User and Translog Supervisor. Translog User is the
component designed for data collection in the form of log files based on
projects created in the Translog Supervisor. The experimenter when pre-
paring a project in the Supervisor component prepares an input text
(Source Text) and specifies a number of parameters which will determine
how the text will be displayed in the User component for a subject who
participates in the experiment. The subject works in the User component
on a split screen. As soon as the ‘start logging’ command is chosen the
Source Text is displayed in the top half of the screen and the Target Text
window opens at the bottom half of the screen where all the typing and
editing of the translation takes place. The actual key logging does not in-
terfere with the subject’s text production and text editing activity which
makes Translog an overt research tool with high ecological validity al-
though the subjects are aware of taking part in an experiment, and when
they finish their translation and choose the command ‘stop logging’, the
program will ask for the file name so that it can be successfully saved and
later retrieved for analysis. The log files saved in the Translog User com-
ponent can be accessed in the Translog Supervisor component. When a
log file is opened the SL text and the TL text are displayed on the left (just
as it was the case in the User component) and the log file is displayed on
the right hand side of the screen. The log file consists of keystrokes, some
symbols and time indications which allow the observation of all the activ-
ity performed on the keyboard by a person translating a text. As explained
by Jakobsen (2006),

This behaviour may seem far removed from the thinking that underlies
the text production, but the advantage of such focusing is that hard, ma-
chine-recorded data are made available about an aspect of the total be-
haviour in text production, which our theorizing about text production
in translation cannot ignore. It may not be possible to make detailed in-
ferences about mental language processes from these data alone, but
any attempt to account for these processes should take evidence from
every aspect of the total situation into account (Jakobsen 2006: 99).

The statistics available through Translog include: the duration of the TL

text production, total number of user events (TUE) which include a key-
stroke or a cursor/mouse movement, number of characters noted as text
production (TP), all the text which was eliminated or erased by the trans-
342 Chapter 7

lator (TE) that is elements otherwise invisible in the finished product yet
very indicative of the problem solving and decision making in the transla-
tion process. Further data include all the instances of cursor navigation
(CN) and mouse events (ME) which show how the translator moves
within the TL text. These aspects are revealing for the process of self-
revision and on-line editing of the emerging TL text. Most importantly,
however the program records pauses and time delays which can be used
to interpret the information processing and reaction times needed to make
a decision on how to translate a given stretch of text or a particular word.
Further data calculate the ratio of TUE per minute and text production per
minute for each individual Translog user.
Using reaction times to infer language processing has a long history in
the studies of bilingual lexicon (Dijkstra and Van Heuven 2002, Kroll and
De Groot 2005) and can therefore be used to infer language processing in
the study of the translation process. While mental lexicon research uses
reaction times to infer the speed of lexical access (see chapter 2) by
measuring reaction time for individual words (usually nouns) taken out of
context, the lexical retrieval which can be observed in Translog files is
contextualized and, similar to psycholinguistic experiments using e-
prime2 software, calculated with the precision of milliseconds. Therefore,
analyzing reaction times during the translation process to infer the ease or
difficulty of lexical access seems justified. Short pauses will indicate fast
access, quick decision making possibly using automatized memory traces
to use cross-linguistic associations, whereas longer time delays will indi-
cate either problems with fluent retrieval followed by problem solving
strategies including frequently conscious decision making supported by
dictionary searches or consulting the Internet for necessary factual infor-
mation to solve a particular translation problem. All the data available
through Translog can be analyzed quantitatively which makes Translog a
valuable contribution to the translation process research which so far has
lacked a tool which would allow for quantitative analyses with the quali-
tative case studies as the most frequent research design in empirical
Translation Studies. Furthermore, it is possible to replay the actual typing
process of the TL text production offering a dynamic visualization (see
Jakobsen 2006: 100) of the translator’s efforts involving problem spotting

–––––––––
2
E-prime is a popular software program for measuring reaction times following the
presentation of stimuli in psychological and psycholinguistic experiments.
 Empirical investigation into the development of … 343

and problem solving with sometimes numerous trials and errors. The re-
play function has been used by some scholars (Hansen 2002, Buchweitz
and Alves 2006) to collect retrospective data by asking the subject to
comment on the translation process while watching it being replayed in
actual time.
A sample of the Translog file from Study 1 will provide an illustration
of the data set collected by Translog. The section below shows a sample of
the Translog file which was created while the subject (G1 AZ) was working
on the translation of the following sentence from the SL text (in Polish):

SL text (Polish): Film oparty na opowiadaniach Jarosława Iwaszkiewi-

cza, wydaje się być subtelną i wzruszającą historią o niemożliwej miłości.

While the subject produced the TL version in English below:

“The movie based on Jaroslaw Iwaszkiewcz’s short story seems to be a

subtle and poignant story about an impossible love”.

The Translog program recorded the following file:

[Start][ 30.445]The movie is based on the sto

ru ⌫⌫y b⌫ of Jaroslaw Iwaszkiewicz[ 15.275][ ] ⌫⌫⌫
[ ] ⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫
⌫⌫⌫⌫⌫⌫⌫⌫⌫Jari⌫oslaw Iwa szkiewcz 's story
seems to be a subtel and poignant s tory about
an ⌫⌫⌫ an impos sible love[ 28.908][ ] short
[ ] .

The symbols explained:

Pause = (1 second)
Pauses (time delays) longer than 9 seconds are counted
[ 31.436]
Eliminated text = ⌫⌫⌫
Cursor navigation =
Mouse event = [ ]

Looking at the final product, the TL sentence it is impossible to guess

where the most time consuming translation problems occurred. The
Translog file which has been recorded during the translation process
makes the problematic time consuming stretches visible. The subject took
344 Chapter 7

30.445 seconds before starting to type the translation. There were two
time delays (15.275 seconds and 28.908 seconds). The eliminated text
shows that the subject changed her mind as to the word order and erased
the initial “based on the story by Jarosław Iwaszkiewicz” to “based on
Jarosław Iwaszkiewicz’s story”. The time delay of 28.908 seconds was
most probably used for revising the sentence and possibly comparing it
with the SL sentence which made the subject realize that the English
equivalent of the Polish word opowiadanie is not ‘story’, but ‘short story’
as it is indicated by the mouse event and inserting the word ‘short’ before
‘story’. The subject did not consult any external resources while produc-
ing the translation of this sentence.
The translation of the same SL sentence by another subject (G2 MP)
resulted in the following TL version:

“The movie, based on a short story by Jarosław Iwaszkiewicz, seems to

be a subtle and moving story of impossible love”.

The Translog file created in the process is as follows:

[Start][ 02:13.421]"Sweet flag" The movie [ ] B,

I ⌫E [ ] is based o n Jarosław Iwaszkiewicz ⌫⌫
⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫a story by Jarosław
Iwaszkiewicz [ ]⌫⌫⌫⌫ [ ] seems to be a sy
⌫ ⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫ seems to
be a subtle and moving [ ] short [ ] story
of [ 37.881]impossible love I .

The differences are visible in terms of time spent to read the text, which is
2 minutes and 13.421 seconds. Also, in contrast to the first Translog sam-
ple the subject used electronic dictionaries indicated by the letters ‘B’ (bi-
lingual dictionary) and ‘I’ (internet). As it has been demonstrated by the
Translog samples, the research method offers a promising insight into the
translation process which is quite unmatched by such methods as error
analysis or Think Aloud Protocols.
Although Translog records the actual TL text production understand-
ingly, some part of the translation process is still outside the scope offered
by Translog as a research method, including the stage of the SL text reading
and comprehension. However, since the SL text is displayed only when the
user clicks ‘start logging’, it is possible to infer how much time was spent
 Empirical investigation into the development of … 345

on reading the SL text prior to the actual typing of the TL translation. Re-
cent developments in the application of the method include combining
Translog software with eye-tracking (Jakobsen 2011) which offers yet fur-
ther insight into mental effort management and distribution of cognitive at-
tention (Wright and Ward 2008) during the translation process.
Unfortunately, it is as yet not possible for Translog to record dictionary
look-ups or Internet searches performed during the translation process using
the Translog program alone. Yet, it is possible to supplement the data on text
production available through Translog with a screen capture program such
as Screencam or Camtasia which will record the translator’s activity outside
Translog when the external sources including electronic dictionaries and
data bases are consulted during the translation process. Translog will record
the time pauses between typing down the translation and Camtasia will al-
low seeing what the translator was looking at to find assistance in making
the final decision. Combining such methods offers a comprehensive insight
into problem solving during the translation process and allows to understand
why and how translators are looking for external information (Lauffer 2002:
69). However, as it is to be expected and as it has been frequently mentioned
by those who use the combined methods such experiments require infra-
structure and are extremely labour intensive. Lauffer (2002: 60) for exam-
ple, studied only three translators, a junior and a senior translator at Toyota
Canada and a translator (no indication of the length of practice) who has
worked with the Ontario Government’s Translation Service. To study a large
number of subjects requires a lot of time and effort not only in terms of data
collection but also in terms of data analysis, and therefore it would be feasi-
ble in large-scale national or international projects.
In the studies which I carried out for my post-doctoral project described
in this chapter I used Translog following a pilot study which confirmed that
the method is sufficient to test the research hypotheses. The texts chosen for
the experiment were simple in terms of language and translating them did
not require any factual knowledge that would be outside the reach of the
subjects who participated in the experiment. What is more, using a combina-
tion of methods (Translog + Camtasia) would further complicate the analy-
sis which would run the risk of shifting the research focus from the transla-
tion skill development within a translator as an individual to other issues,
e.g., the use of electronic resources in the translation process. This aspect of
the translator’s performance deserves a separate and thorough investigation
which is not the focus of the present studies discussed in this chapter.
346 Chapter 7

To sum up, generally speaking the software program Translog seems

user friendly although obviously it is not the program in which translators
normally translate and using it will constitute a novel experience for the
translators participating in the experiment. In the study described by Lauf-
fer (2002) which used Translog among other research methods, a potential
drawback of Translog was indicated. Although most translators said that
they liked using Translog, some of them were short of certain features that
they are used to, and which are not available in Translog including: spell-
checkers, various fonts, underlining, highlighting and pasting function.
The subjects who participated in my experiments, both Study 1 and Study
2 which will be described in detail below, generally found the program
unobtrusive. Although the Translog program is not ideal and open to im-
provement as confirmed by the continuous efforts of its designer (see Ja-
kobsen 2006, 2011 on subsequent versions of the program), it is possibly
the most unobtrusive method of investigating the translator’s performance
available so far the use of which does not compromise the ecological va-
lidity of the translator’s work. Just as it is the case in real life the transla-
tor works on a computer and is free to use any external resources which
are available to him/her.

7.2. Study 1 – objectives, participants, procedure

The purpose of Study 1 was to conduct a comparative analysis of three

groups of translators with different experience in translation practice:
Group 1 (G1) included 10 2BA students of English as a foreign language
at the School of English, Adam Mickiewicz University, Group 2 (G2) in-
cluded 10 translation trainees who were in their 2MA year and who spe-
cialized in Conference Interpreting. Group 3 (G3) included 10 practicing
translators with various length of practice, ranging from 1 year to 30 years
of experience in providing translation services. The comparative analysis
of the translation performance in the three groups of subjects was aimed
to test the following research hypotheses:

1) The ability to integrate knowledge for the purpose of a translation

task develops with experience.
2) The least experienced translators (G1) will integrate only part of
the knowledge required for the translation task (e.g., only their bi-
lingual knowledge). They will translate without the support of an
 Empirical investigation into the development of … 347

adequate Knowledge Integration Network (KIN) which will slow

down the translation process and result in frequent interruptions
because of low L2 proficiency.
3) The more experienced translators (G2), will make a conscious ef-
fort to build an adequate KIN which will extend the translation
process and result in conscious metacognitive supervision with
frequent checking for confirmation to ensure adequate transla-
tion.
4) Practicing translators with many years of experience automatically
integrate knowledge needed for the task at hand which speeds up
the translation process and makes it more fluent.

The three groups of subjects were asked to translate a short text from Pol-
ish into English. The direction, language A to language B was chosen so
that in terms of comprehension the SL text is equally accessible to all the
subjects since they are all native speakers of Polish. A further reason was
that the Translog program is able to gather detailed data on TL text pro-
duction and not on SL text comprehension. The text was carefully chosen
according to the following criteria:

1. It has to be a complete text, not an extract of a larger text to elimi-

nate the possibility that not knowing the entire text could hinder
understanding of the purpose and function of the SL text.
2. It has to be fairly short so that it is possible to translate it in one
sitting and translating it does not cause fatigue and tiredness
which could cause cognitive overload or make it necessary for the
subjects to have a break.
3. It has to be within the reach of the subjects’ comprehension and
vocabulary and understanding it is not dependant on prior factual
knowledge.
4. It has to be ecologically valid in terms of the potential need to have
it translated into English in real life.
5. It has to be fairly topical in terms of current events in which the
subjects participate.

The choice of such texts is fairly limited and the text type (Reiss 1981)
which was initially chosen and pre-tested in a pilot study was a film re-
view. The film entitled, Tatarak by one of Poland’s greatest directors,
348 Chapter 7

Andrzej Wajda was just coming into Polish cinemas3. It received a lot of
media attention and since most of Wajda’s films have been shown in for-
eign cinemas, translating its review seemed a realistic commission that a
translator can receive. The text consisted of 172 words (1, 033 signs). It
was written in clear language and it seemed to match the requirements of
such text types in terms of language and function. It was placed on an ad-
vertising leaflet which was available free of charge in Polish cinemas.
All the participants received a translation brief: Translate the follow-
ing film review for an on-line English cinema magazine. The subjects
were informed that they are participating in a translation process aware-
ness project which aimed at gathering data on the process of translation
with the help of the Translog program. They were not told how the pro-
gram records the process and which information it makes available. This
precaution was made not to influence the subjects’ performance prior to
the actual translation task. It might have happened that if the subjects
knew what particular data are recorded (e.g., duration) they might have
performed differently than in real life. Instead they were asked to behave
as if they were commissioned to produce the translation, taking as much
time as necessary. They were also allowed to use electronic dictionaries
and the Internet as they normally use them when they translate at home.
Since Translog does not record activities outside the program such as con-
sulting e-sources the subjects were asked to mark the use of external re-
sources typing a letter ‘b’ for bilingual dictionary, ‘m’ for monolingual
dictionary and ‘I’ for the Internet sites directly before switching on to use
them. This was done to make it later possible to accurately interpret
longer pauses and time delays as spent on dictionary/Internet consultation
and not on something else.
All the translations were performed in a computer lab with five work
stations in several sessions. The researcher, myself was always present
and all the participants were instructed on how to proceed. When every-
thing was clear the subjects clicked ‘start logging’ and worked on their
translation until they finished and then they pressed ‘stop logging’ and the
file was saved. Then the subjects were shown their log file and most of
them were very much surprised by the detailed record of their translation

–––––––––
3
Tatarak (translated as Sweet Rush) by Andrzej Wajda won the Alfred Bauer
Award at 59th Berlin International Film Festival for “opening new perspectives in the art
of cinema” (http://www.pisf.pl/en/film-production-guide-1/film-industry).
 Empirical investigation into the development of … 349

process. The majority were impressed by the program and expressed their
interest in the data which are made observable by the program.
From the 30 files recorded four were rejected as faulty. In one case the
program switched itself off during the actual translation. Two subjects
were inconsistent with marking the use of dictionaries and during one ses-
sion the Internet dictionary become inaccessible. In total an equal number
of 8 Translog files for each group were chosen for analysis.

7.3. Study 1 – data analysis

The focus of the analysis was on the TL text processing data available
through the Translog program. The primary research question was: Are
the effects of translation experience observable not only and as it can be
expected in the quality of the translation as the end product (see Shreve
1997 quoted in chapter 5) but in the actual performance, to be more spe-
cific in the way translators type their translations? If there are pronounced
differences between the three groups, the ultimate question was: Is it pos-
sible to infer from the results that with translation experience translators
learn to integrate all the necessary knowledge for a particular translation
task and their ability to create KINs has a positive impact on both, the
process and the product?
The analysis focused on those aspects of the translation process which
are believed to differ the most according to the translator’s experience
(see chapter 5) and they included:

1) duration, the total length of time taken by each subject to translate the
same text
2) the number of dictionary/internet look ups
3) typing speed
4) distribution of the remaining time in the translation process including:
– time spent before the onset of translation (time devoted to ‘open-
ing’ the SL text’)
– time spent on post-draft revision
– the amount of text elimination
– the amount of navigation in the text (movement within the text,
including making on-line changes to the emerging translation)
350 Chapter 7

Each of the above aspects will be discussed with reference to the research
hypotheses to see if the hypothesized KIN in the process of translation skill
development can be supported by the empirical data collected in Study 1.

7.3.1. Duration

There were pronounced differences between the representatives of the three

groups with respect to the amount of time which was devoted to translate
the text. Since there were no time restrictions, it can be assumed that all the
subjects worked on their translations at their usual pace. They were never
hurried and were told to take as much time as they felt necessary. This in-
formation was hoped to remove any unnecessary time pressure which would
test rather how translators cope with stress induced by time constraints. Al-
though it is of course likely that in real practice translators sometimes do
work under time pressure, it was not used as a variable in this study in
which a single variable of translation experience was tested. The time spent
on the translation of the SL text ranged from 7 minutes 49 seconds 390 mil-
liseconds to 49 minutes 23 seconds and 292 milliseconds. The table below
shows the duration for each individual subject in the three groups

Table 1. Duration of the translation task for individual subjects in the

three groups.

Subjects Duration G1 Duration G2 Duration G3

S1 11:15.341 12:47.821 07:49.390
S2 20: 49.129 17:03.000 10:26.474
S3 22:58.577 18:49.966 11:53.145
S4 24:45.742 18:58.082 23:29.172
S5 25:28.701 21:59.100 28:13.043
S6 28:47.928 24:27.551 29:14.935
S7 36:57.039 28:19.123 39:32.513
S8 36:58.285 31:24.057 49:23.292

As it can be seen in the table, both the fastest and the slowest translators
came from the group G3, which is the group of practicing translators. The
result which might seem initially surprising has its explanation in the
length of practice marked by each translator. The fastest translator has
been providing translation services for 30 years whereas the slowest
 Empirical investigation into the development of … 351

translator has been working as a freelance translator for 1 year. The ques-
tion, ‘What do translators learn from experience?’, put forward in chapter
five and hypothetically answered by the proposal of KIN, becomes even
more justified in view of the above results. Looking at the individual raw
data for duration is important because of the significant idiosyncratic dif-
ferences within each of the three groups which amount to nearly 25 min-
utes for the least experienced group (G1), 19 minutes for the translation
trainees (G2) and the surprising 42 minutes for the group of practicing
translators (G3) when the slowest and the fastest translators are taken into
account. These individual differences following an inferential statistical
analysis using ANOVA disappear. The graph below shows the mean aver-
age for the duration of the task in the three groups.

00:28:48 00:26:00 00:25:00 00:24:14

00:21:43
00:21:36

00:14:24

00:07:12

00:00:00
beginners translation professional Total
trainees translators

Graph 1. Statistics for the duration data.

Needless to say, the number of subjects is small to expect statistical signifi-

cance and the wide range of variation in each of the groups reduces the
likelihood of statistically significant differences. The results offer more in
the form of raw data and point to the differences with which all the subjects
dealt with the same task. Since groups G1 and G2 are fairly homogenous in
terms of age, language proficiency and translation experience (group G1 is
equivalent to group 2BA described in detail and analyzed in chapter 6, the
section on participants, whereas group G2 is equivalent to group 2MA in
chapter 6 but with a structured exposure and practice in translation offered
352 Chapter 7

in their Conference Interpreting programme), it is justified to assume that

these differences originate rather in their general cognitive capacities than
in the language proficiency domain. The variety in the duration data as such
could, although much indirectly confirm the proposal of KIN, suggesting
that the skill to translate relies on the individual’s ability to integrate all the
necessary knowledge and not only on the ability to activate one’s bilingual
knowledge. The group of practicing translators is the least homogenous in
terms of age and length of practice, subject S1 has had 30 years of profes-
sional experience, S2 – 20 years of practice, S3 – 9, S4 – 4, S5 and S6 both
have worked as translators for 4 years and S7 with S8 have been working
as freelance translators for 1 year only. All the practicing translators were
free lancers for whom translation is not the only source of income. Subjects
S1 and S2 were self-taught translators and the remaining translators were
graduates of the Conference Interpreting programme who also provide
written translation services. Looking at the correlation between years of ex-
perience and the duration of the translation task for all individual translators
reveals results which can be interpreted as supporting hypothesis 4.
00:52:00
 

00:44:00

00:36:00

00:28:00

duration
00:20:00

00:12:00

00:04:00

0 5 10 15 20 25 30

PTE in years

Graph 2. Correlation between years of experience for practicing transla-

tors (PTE) and duration rounded to minutes.
 Empirical investigation into the development of … 353

The strength of the correlation is convincing and shows that with time and
professional experience the process of translation takes less time which al-
lows to infer that problem solving and decision making is less time con-
suming4. As the graph below shows if the two slowest translators with only
one year as practicing translators are eliminated from group G3 the differ-
ences in terms of duration become more pronounced between the groups.

00:28:48 00:26:00
00:21:43 00:22:24
00:21:36 00:18:31

00:14:24

00:07:12

00:00:00
beginners translation professional Total
trainees translators

Graph 3. Statistics for the duration data without the two translators with 1
year of experience in group G3.

Clearly, the differences visible in graph 3 reflect the expectations about

the role of experience and its impact on the speed of the actual translation
process. The immediate question which arises is, ‘On what aspects of the
translation process do translators with longer experience save time?’
Drawing from the comparative analysis of inexperienced and experienced
translators presented in detail in chapter 5, the use of a dictionary seems
the first candidate for extending the duration of the task at hand.

–––––––––
4
This conclusion however is reached when the SL text was very simple. In the case
of more complex texts experience does not necessarily mean a significantly faster trans-
lation process (see chapter 6).
354 Chapter 7

7.3.2 Duration and dictionary use

Table 2 below shows the number of dictionary/Internet searches for all the
participants of the study.

Table 2. The use of external sources for individual subjects in the three
groups.

Subject G1 G2 G3
1 5 6 1
2 3 6 1
3 8 3 6
4 10 2 8
5 5 3 1
6 8 9 13
7 10 17 17
8 6 12 14

Similar to the data on duration the insight into raw data is essential as in
terms of inferential statistics all the individual differences become invisi-
ble. What is more, graph 4 demonstrates that on average it is the group of
practicing translators which scored the highest in terms of consulting dic-
tionaries or using the Internet.

9.00
7.3 7.6 7.3
8.00 6.9
7.00
6.00
5.00
4.00
3.00
2.00
1.00
0.00
beginners translation practising Total
trainees translators

Garph 4. Statistics for consulting external sources.

 Empirical investigation into the development of … 355

The group of translation trainees (G2) on average used dictionaries also

more frequently than the least experienced group (G1). It seems that such
results show no correlation between language proficiency which obvi-
ously is much higher in groups G2 and G3 and translation skill. On the
other hand, the results could demonstrate that with experience there is a
greater awareness of the translation process and more concern for the
quality of the final product. These issues are more important for those
subjects who have been trained (or who self-developed their expertise) as
translators. It is possible that the least experienced translators (group G1)
are at the stage of unconscious incompetence (González Davies 2004) and
their decisions are easier to make and more intuitive even if they run a
high risk of making transfer errors or choosing contextually inappropriate
TL words in their translations. Translation trainees have probably entered
the stage of conscious competence and they feel a strong need to confirm
their solutions and take every possible precaution to choose the best pos-
sible word which would match the text type, the context and the function
the TL text is supposed to fulfill. It is also possible and significant for
supporting hypothesis 3 that translation trainees and practicing translators
with only one year of professional experience consciously create adequate
KINs and therefore search for missing ‘pieces’ of their integrated net-
works in external resources, e.g., dictionaries or data bases. As it was
pointed out in chapter 6 conscious language processing consumes a lot of
time and a substantial amount of mental energy which will extend the
translation process and create many distractions and disruptions when ex-
ternal sources are used. Two subjects S7 from groups G2 and G3 used ex-
ternal sources 17 times which considering the fact that the SL text was in
their native language and its vocabulary was fairly simple is striking and
points to the subjects’ limited self-confidence. This is most likely not the
result that would be expected although it can be justified by the trainees
dedicated effort to do their best taking as much time as they feel needed
on consulting external sources. The time and effort devoted to consulting
external sources can be interpreted as their conscious attempt to create an
adequate KIN (see chapter 5). Activating all the necessary knowledge is
not possible without supplementing what is needed from outside their
own memory store (cf. Tabakowska 2003, Stolze 2004) possibly com-
bined with the need to find confirmation for their hunches and hypotheses
due to low self-confidence on how to best solve the subsequent translation
problems.
356 Chapter 7

In contrast the most experienced translators S1 and S2 as well as S5

from group G3 used a dictionary only once to find the English equivalent
of the low frequency word tatarak, in the title of Wajda’s film. Clearly,
there can be a range of reasons explaining the drop of dictionary use with
growing experience in the profession. As it was suggested by practicing
translators in the questionnaire study (see chapter 6), their mental lexicons
continue to grow and collect detailed information about words (e.g., their
grammatical behaviour, style, etc.) throughout their working life. They
can also resort to automatic interlingual associations which have left
memory traces which speed up lexical access and might give the impres-
sion of by-passing the conceptual base (see Paradis 2009). Possibly,
though their activated and integrated knowledge (bilingual, interlingual,
metalinguistic, pragmatic, intercultural, textual, domain specific, etc.) in
the form of a specific KIN primes lexical access to the desired TL words.
Thinking globally (in a top down manner), rather than locally (in a bottom
up manner) individual words cease to cause problems since the same in-
formation can be frequently expressed in many different ways, of course
with some exceptions, like in the case of tatarak, where external memory
sources become indispensable. Clearly, however, there is more self-
confidence and self-reliance in more experienced translators and the need
to consult dictionaries and confirm solutions decreases with years of prac-
tice at least when it comes to the most frequently used store of vocabulary
(cf. Bell’s (1991) frequent lexis store or Gile’s (2009) gravitational
model). The proportion between reliance on internal memory and external
memory use changes with practice possibly similar to what happens in the
development of any other kind of expertise (see Hoffman 1997). Self-
reliance is also a feature of systems intelligence discussed in chapter 5
which develops through interaction and feedback, which are consciously
used by an expert learner (a developing translator) to improve future per-
formance. In the case of the practicing translators who participated in the
Translog study correlating the years of experience with the number of dic-
tionary look ups shows a significant correlation depicted in Graph 5 be-
low.
 Empirical investigation into the development of … 357

 
20  

15
Sources

10

0 5 10 15 20 25 30

PTE in years

Graph 5. Correlation between years of experience for practicing transla-

tors (PTE) and the number of dictionary look-ups (sources).

When interpreting the above results, however it has to be remembered

that the text used for the study was fairly simple and did not require the
knowledge of specific terminology. The KIN which was created in the
mind of an experienced translator could safely rely on the internal cogni-
tive resources including the bilingual mental lexicon. If however, the text
is more complex, more in the range described as ESP (English for Special
Purposes), the translator would probably have to devote more time to cre-
ate an adequate KIN and the use of external memory sources would be
necessary, and it would consequently extend the time spent on the transla-
tion task. This was indicated by the practicing translators who responded
to the questionnaire analyzed in chapter 6.
Looking at the data on dictionary use in the three groups of subjects it
is interesting to see how the analysis changes if the two practicing transla-
tors with 1 year of work experience are removed from the statistical
analysis. Similar to the manipulation done on the duration data, the differ-
358 Chapter 7

ences between the three groups shown in Graph 6 are again more pro-
nounced and more in line with expectations about the educating role of
experience.

8.00 7.3
6.9
7.00 6.5
6.00 5.0
5.00
4.00
3.00
2.00
1.00
0.00
beginners translation professional Total
trainees translators

Graph 6. Statistics for consulting external sources without the two transla-
tors with 1 year of experience in G3.

The results demonstrated in Graph 6 show that with time the need to con-
sult dictionaries drops with immediate effect on the duration of the trans-
lation task. It is interesting, however that the most frequent users of dic-
tionaries are now the translation trainees. This result probably reflects the
habit installed during training which aims to develop in future translators
the need for checking and double checking (Tabakowska 2003). There-
fore, the above results can be interpreted as positive and as such they only
confirm that the trainees are consulting external resources with the right
intention. For them it is not the speed that is the aim but the quality of the
translation. They are most likely at the stage of conscious competence
with limited self-confidence. Although, following the manipulation of
data the differences are more visible, and more in line with expectations,
they are still not significant in terms of statistics.
 Empirical investigation into the development of … 359

To sum up, the use of dictionaries and other external memory sources
is responsible for extending the translation process. Each search takes
time which as calculated by Translog can range from several seconds to
two or three minutes. As it was observed in chapter 5, consulting external
sources may also have a disruptive effect on the flow of the translation
process for less experienced translators. Asadi and Séguinot (2005) noted
that even their experienced subjects tended to make a typing error when
after consulting a dictionary they returned to their translation pointing to
the fact that the flow of language processing is disrupted by dictionary
searches not only for less experienced translators. Nevertheless, the use of
dictionaries or other sources is not the only factor responsible for the in-
dividual differences in the time taken to translate the SL text.

7.3.3. Duration and typing speed

The data available through Translog revealed another interesting relation-

ship between the duration of the translation task and the overall keyboard
activity marked as events per minute (EPM) and text production per min-
ute (TPM), in other words the typing speed. Table 3 below shows the cor-
relations between the duration and EPM as well as between the duration
and TPM calculated for the slowest and the fastest translators from each
group.

Table 3. The correlations between the duration and EPM and TPM calcu-
lated for the slowest and the fastest translators from groups G1, G2, G3.

Correlations
duration
Pearson Correlation -,941(**)
EPM Sig. (2-tailed) 0,005
N 6
Pearson Correlation -,922(**)
TPM Sig. (2-tailed) 0,009
N 6
**. Correlation is significant at the 0.01 level (2-tailed).

As it can be inferred from the strength of the correlations typing speed is

an important factor influencing the duration of the entire translation
task. This correlation might seem so blatantly obvious that it has, at least
to my knowledge, been left out by investigators of the translation proc-
360 Chapter 7

ess. Yet, its significance deserves attention and raises several questions:
‘Is the slow typing speed determined by poor typing skills?’, or: ‘Is it
determined by slow language processing in the translator’s mind? These
questions are not easy to answer. Do people generally type faster in their
native language than in their foreign language? Are typing skills taught
at schools or self-acquired? Is it only typing skills or general computer
skills that had an impact on the speed of the translation performed by
my subjects? Or ultimately, has the typing speed been determined by the
task of translating to such an extent that even those subjects who have
decent typing skills took much longer to solve problems and make deci-
sions? A section of two Translog files produced by the slowest and the
fastest translator from group G3 might illustrate the point. Subject S8
while translating the following sentence from Polish: W ten sposób obie
kobiety – Krystyna i Marta – zlewają sie w jedną głęboko zranioną is-
totę, która musi znaleźć w sobie siłę i poradzić sobie z perspektywą nie-
uniknionej śmierci took a significant amount of time with marked time
delays, two instances of dictionary use (indicated by letter ‘B/I’) and a
certain amount of erased text.

S8 (G3) Translog file

This way both women merge into one [ 23.434]⌫⌫⌫⌫⌫⌫

⌫⌫⌫⌫⌫⌫⌫⌫⌫blend into [ ] B [ ] one, de e
⌫⌫⌫⌫⌫ deeply wounded person , who must
[ 25.152]pluck up the strength [ 59.305]⌫⌫⌫⌫
⌫⌫⌫⌫⌫⌫⌫⌫⌫ strenght I[ 10.064] to battle the
[ 25.556]prospect of [ ] B[ 15.939][ ]inescapab
le death.

The same sentence when translated by a practicing translator with 30

years of experience left the following record in the Translog file:

S1 (G3) Translog file

In this way both women, Krysyt⌫⌫tyna and Marta, bond into

one deeply hurt being , which must find strength
in itself and cope with the prospect of i⌫ uh⌫nav
oidable death.
 Empirical investigation into the development of … 361

Clearly there is no marked time delay apart from 1 to 3 seconds between

typing single words of the TL text. Such pauses then are only an indica-
tor of motor processes, e.g., typing ability (see Rothe-Neves 2003), there
is no time taken for consulting dictionaries and there is a scant amount
of erased text. In the last line there is a five second pause following let-
ter ‘i’ which was quickly erased. In psycholinguistic terms one can
speculate that the translator possibly made a quick decision and rejected
the word ‘inevitable’ as a competing equivalent to ‘unavoidable’ or real-
ized a potential spelling/grammatical mistake, or suppressed language
interference. This possible explanation confirms Jakobsen’s (2006)
comment about the extent to which one can infer mental language proc-
essing in the translation process using Translog data. Obviously at the
fine-grained level many questions will still remain open as to what actu-
ally goes on in the translating mind. At the coarse-grain level, however,
it seems that the differences recorded in the Translog files of subject S1
and S8 point to the differences in the language processing of both trans-
lators. For the translation trainee the duration is extended by the pauses
in between the actual typing of the TL words. Buchweitz and Alves
(2006) noted that this difference points to differences in the segmenta-
tion of the SL text into translation units and they reported that with more
experience there is significantly less segmentation and the translation
units become larger in size (see also Jakobsen 2002, Dimitrova 2005).
Buchwietz and Alves (2006: 248) came up with an interesting idea of
‘cognitive rhythm’ of the translator. The individual cognitive rhythm can
range from a word to a phrase or a full sentence which is processed as a
single translation unit uninterrupted by pauses longer than 3-5 seconds.
The example above seems to lend support to the above suggestion. It is
therefore justified to infer that with growing experience in interlingual
transfer of meaning the time to make decisions is reduced possibly be-
cause of the ability to integrate all the knowledge including procedural-
ised knowledge underlying a range of skills (e.g., typing/computer skills
are then a part of the translation skill as well) needed to perform the task
efficiently in terms of time. The fairly smooth, virtually uninterrupted
fluent typing of the TL text by the most experienced translator is most
probably carried out under metacognitve supervision which allows to
process larger stretches of the SL text as single translation units. Less
experienced translators work on smaller units and frequently do not syn-
chronize their choices (i.e., lexical searches) with metacognitve supervi-
362 Chapter 7

sion and consequently they lack the guiding direction in their translation
process and face too much uncertainty which they try to dispel by con-
sulting dictionaries. Holmes (1988) would say that they lack a macro-
strategy, their mental map of the TL text which is sketched out while
they read the SL text for translation. I would like to add that they could
have envisaged the mental map of their TL text if they had created an
adequate KIN. Hönig (1991) further explained that inexperienced trans-
lators try to compensate for their lack of macro-strategy employing mi-
cro-strategies which frequently lead them into rule-governed use of lan-
guage, which sometimes makes them oblivious to the rules of discourse
grammar and the choices they make seem unnatural and out of place,
like ‘to pluck up the strength’ or ‘to battle the prospect’ (see the S8 (G3)
Translog file sample). Kussmaul (1991) noted a similar tendency when
investigating creative thinking in translation. His subjects who were
translation students seemed very creative initially but when it came to
writing down their translations they frequently abandoned their creative
choices (possibly an outcome of their unstable KIN) for the rule-
governed linguistic choices as if they were afraid of taking liberty and
preferred to play safe. Most likely, there is a problem of confidence, a
sign of still feeling inadequately in control of the L2 grammar and lexis
which slows down the fluency of thinking and limits the flexibility of
strategic choices. Guy Cook (1991: 128) pointed out that inexperienced
translators who are frequently also inexperienced L2 users are deter-
mined to search for fixed meaning and are devoted to find exact words
which in their opinion are perfect equivalents. Experienced L2 users
(e.g., practicing translators) are on the other hand more tolerant of the
indeterminacy of meaning and consequently have a wider choice of TL
means. If they had created an adequate KIN their choices are predeter-
mined and the priming effect facilitates lexical access and the risk of in-
appropriate choice either semantically or pragmatically is much lower
(see chapter 2 on lexical access). The tempo and fluency of decisions
made by the translator with 30 years of experience not only points to
more self-trust but it also demonstrates the division of labour according
to which many actions have been automatized (e.g., typing skill, lexical
access, self-monitoring, on-line revision, self-correction) while others, if
need be, might be carried out in a consciously controlled manner with-
out the danger of cognitive overload. All the actions though are run un-
der metacognitve supervision which in turn, as it was demonstrated in
 Empirical investigation into the development of … 363

chapter 5 relies on the KIN created for the task at hand. It is possible to
assume that an adequate KIN will decide about the ratio between auto-
matic and controlled processing. The more automatic and fluent the
translation process appears to be the more cognitive capacity remains
available for metacognitive supervision. In effect the translation task
takes less time and runs a higher chance of meeting professional stan-
dards of quality.
In simple terms the translator knows if the SL text is more or less
demanding and consequently if it will require extra effort to create an
adequate KIN. This assumption was confirmed by the questionnaire
study as well (see chapter 6). In the case of the SL text used in Study 1,
the professional translator with 30 years of experience was probably re-
lying on fairly automatic interlingual associations. However, as it is
demonstrated by the longest pause of 5 seconds in the last line of the
Translog file (subject S1 G3 showed above), the metacognitve supervi-
sion is switched on all the time keeping all the knowledge integrated in
the KIN and ready to supply feedback when the need arises. In this case
the translator protably spotted a potential spelling mistake and erased
the potential erroneous ‘inavoidable’ changing it to ‘unavoidable’ after a
5 second pause. The time was probably used to recall the proper spelling
from memory or for rejecting a competing equivalent (inevitable) as less
adequate. It seems that the metacognitve supervision is scrutinizing
other automatic or semi-automatic actions which are performed at the
same time, e.g., reading the SL text, accessing TL vocabulary and gram-
matical patterns, typing the TL text and revising the emerging TL text as
adequately representing the SL text . In this way the professional experi-
enced translator is like Donald’s (2007b: 75) piano player who although
focused on his/her interpretation of the entire piece will nevertheless no-
tice a problem at the lower level of the actual production and eradicate it
at once, if possible.
A further insight into the relationship between TPM and the duration
of the task is offered by correlating the TPM and the years of experience
in translation calculated for the subjects from group G3. The graph below
shows that the more experience in providing translation services the high-
est the TPM for each subject.
364 Chapter 7

 
200,00  

150,00

100,00

TPM

50,00

0,00

0 5 10 15 20 25 30

PTE in years

Graph 7. Correlation between TPM and years of experience (PTE).

The strength of the correlation is significant (see Table 7 for significant

correlations) and points to the benefits of the length of practice. The
time needed for performing the same translation task decreased with ex-
perience not only because of not feeling the need to consult dictionaries
but because of the internal self-confidence in making the right decisions
which allows for smooth language processing, which in turn is reflected
in the fast typing of the TL text. It is plausible that the fast language
processing is achieved due to the growing ability to integrate knowledge
(and skills) into an adequate knowledge integration network (KIN)
needed for the task at hand.
So far I have focused on the duration and two factors which affect the
duration with a considerable strength. Bearing in mind, however that the
translation process is multilayered and involves multi-tasking including
both comprehension and production processes, the next question refers to
how the remaining time (excluding dictionary consultation discussed ear-
lier) is distributed throughout the task.
 Empirical investigation into the development of … 365

7.3.4. Distribution of the remaining time in the translation process

Having assumed in chapter 5 that the KIN for a specific translation task
is created during the first reading of the SL text and it is further dy-
namically readjusted as the translation proceeds, I wanted to measure
how much time was taken by each subject before the actual onset of the
TL text production. As it is demonstrated by graph 8, the beginners
(G1) took the least amount of time before they started to type their
translations.

00:01:44 00:01:25 00:01:28

00:01:26 00:01:15
00:01:09 00:00:51
00:00:52
00:00:35
00:00:17
00:00:00
beginners translation professional Total
trainees translators

Graph 8. Time prior to the onset of translation for the three groups of subjects.

On average 51 seconds were devoted by the beginners in translation to

read the SL text, and acting on the hypothetical assumptions laid out in
chapter 5, to create a KIN for the task at hand. The translation trainees
group spent on average 1 minute 25 seconds and the most experienced
group took only slightly longer 1 minute 28 seconds. The average scores
however, just as it was with the previous aspects discussed above blur
individual differences which are important because translation is always
an individually performed activity. The table below shows raw data on
the time taken by individual subjects before the onset of the TL text pro-
duction.
366 Chapter 7

Table 4. Time prior to the onset of translation for each subject.

Subject G1 G2 G3
1 00:06.066 00:25.442 01:19.910
2 00:30.803 00:30.707 01:12.550
3 01:08.376 00:11.763 00:25.442
4 01:20.854 01:59.112 04:01.829
5 00:58.853 02:52.461 00:36.385
6 00:51.829 01:53.547 00:53.685
7 01:00.323 01:40.746 02:21.836
8 00:59.371 01:50.326 00:54.189

Subject S1 from group G1 took the least time, about 6 seconds which is
clearly enough to glance at the SL text but probably not enough to read it
with comprehension which will necessarily activate relevant knowledge
structures (Sperber and Wilson 1986, Stolze 2004, Donald 2007b) to create
an adequate KIN for the translation task ahead. Subject S1 from group G3
took nearly 1 minute 20 seconds (about 16% of the entire duration of the
translation process which lasted 7:49.390) before the onset of translation
and although from that time 47 seconds 668 milliseconds were used to find
the equivalent of the Polish word tatarak, the time was probably sufficient
to position the SL text (Stolze 2004) and create an adequate KIN. It seems
possible that the time invested in creating a KIN will pay off later on in
smooth and fluent translation. The results confirm the differences between
inexperienced and experienced translators with respect to how they ap-
proach the SL text (see chapter 5), although as the table above shows there
are significant idiosyncratic differences between the subjects within each
group.
Since another pronounced difference between the inexperienced and
experienced translators is the time taken for post-draft revision, I also
measured the time subjects took to revise their first drafts. The graph be-
low shows that indeed the beginners spent the least time revising their
first draft.
 Empirical investigation into the development of … 367

00:07:12 00:06:25 00:06:22 00:05:42

00:05:46 00:04:19
00:04:19
00:02:53
00:01:26
00:00:00
beginners translation professional Total
trainees translators

Graph 9. Time for post-draft revision in the three groups.

The group G2, the trainees took the longest and the practicing translators
were only 3 milliseconds behind. Again, it seems essential to look at raw data
not to allow generalizations to obscure the actual issue under discussion, the
time for revision. The table below shows surprising individual differences.
The two most experienced translators who have been in the profession for 20
(S2) and 30 (S1) years took the least time to revise their translations. Both of
them, however, made a comment that they usually leave the first draft and
revise it later. This approach to post-draft revision was confirmed by the
practicing translators discussed in the questionnaire study in chapter 6. The
raw data confirm the results obtained from the questionnaire study in which a
clear majority of 87.5% practicing translators said that they leave the transla-
tion before revising it. The raw empirical data obtained from the Translog
files lend further support to the hypothesized need to re-frame the KIN be-
fore the final revision of the TL text (Table 5).

Table 5. Time spent on post-draft revision for individual subjects.

Subject G1 G2 G3
1 02:18.505 02:41.864 00:06.397
2 03:11.639 03:04.807 00:12.549
3 05:42.210 06:34.091 02:41.864
4 06:20.172 04:37.207 03:08.447
5 02:26.727 05:24.281 09:25.971
6 05:41.723 11:28.934 13:56.215
7 04:23.490 06:20.279 03:27.396
8 04:35.201 11:14.901 17:59.194
368 Chapter 7

It is therefore plausible that translators feel the need to re-frame themselves,

refresh their mind before they can comfortably check and improve their
translation. It is however, also possible that the need to abandon the KIN
created for the purpose of translating the SL text before they can effi-
ciently revise their translations has important psycholinguistic conse-
quences for the revision process. Time delays before revising a TL text al-
low to inhibit the SL (see chapter 2 on language control) at least by elimi-
nating the recency effect on lexical access as if in the common saying ‘out
of sight, out of mind’. Following this line of reasoning and focusing on
self-revision, I am more inclined to suggest that the two tasks, translating
and revising require some changes in the Knowledge Integration Network
(KIN). It is likely that re-framing from the KIN which was created for the
purpose of translating a given text to a KIN needed to revise the TL text is
most efficient with a time break between the two tasks. The time delay
usually filled with different experiences erases the KIN created for the ac-
tual translation task and allows to gain an objective distance to both the
SL text and the TL text produced in the translation process. This possibil-
ity is confirmed by Hansen (2008) who suggested that translation compe-
tence and revision competence are in fact different sets of competences as
they require different skills.
The individual differences within each group are the most striking in
group G3 where they range from 6 seconds to nearly 18 minutes taken by
the translator with only one year of professional experience. The most
natural question which follows is, ‘What is the time used for when trans-
lators revise their work?’. The Translog program allows for analysing not
only the whole TL text production process, but it also allows to analyse a
chosen section of the process. It therefore allows to measure not only the
time spent on the post-draft revision but it shows the actual text produc-
tion during that time which in turn indicates how many changes were in-
troduced during revision. Graph 10 shows that group G2, the translation
trainees introduced the most changes which are indicated by the number
of keystrokes during the time spent on revising the draft translation.
 Empirical investigation into the development of … 369

100.00 92.4
76.4
80.00 69.9 67.0
60.00
40.00
20.00
0.00
beginners translation professional Total
trainees translators

Graph 10. Text production during post-draft revision.

With respect to the number of changes introduced during the revision be-
ginners were similar to the practicing translators and much less likely to
change their initial solutions. This result is consistent with Jakobsen (2002)
and Alves (2005) who concluded their studies by claiming that the more
experienced translators produce ‘more durable texts’. The raw data in Table
6 however explains how superficial the statistical similarity in fact is.

Table 6. Text production during post-draft revision for individual subjects.

Subject G1 G2 G3
S1 4 84 0
S2 27 42 0
S3 79 126 84
S4 83 64 7
S5 256 24 124
S6 27 145 118
S7 43 36 85
S8 40 218 118
370 Chapter 7

 
00:18:00  

00:15:00

00:12:00
revision

00:09:00

00:06:00

00:03:00

00:00:00

0 5 10 15 20 25 30

PTE in years

Graph 11. The correlation between years of experience as practicing

translators (PTE) and revision.

In each group there are pronounced differences between individual sub-

jects which disappear when average numbers are calculated by inferen-
tial statistics. These individual differences in Group G3 where the two
most experienced translators did not carry out their revision at all, post-
poning it as it is their usual procedure until later affected the strength of
the correlation between revision and years of practice. See graph 11.
As it is demonstrated in table 7 which summarizes the correlation
between years of experience as practicing translators (PTE) and text
processing features discussed so far, the correlation between the time
taken to revise the text and years of experience is not as strong as the
correlations between the duration, dictionary use and text production
per minute.
 Empirical investigation into the development of … 371

Table 7. Significant correlations between years of experience (PTE) and

text processing features.

PTE in years
Pearson Correlation -,803(*)
duration Sig. (2-tailed) 0,016
N 8
Pearson Correlation ,950(**)
TPM Sig. (2-tailed) 0,000
N 8
Pearson Correlation -,716(*)
Sources Sig. (2-tailed) 0,046
N 8
Pearson Correlation -0,613
revision Sig. (2-tailed) 0,106
N 8
*. Correlation is significant at the 0.05 level (2-tailed).
**. Correlation is significant at the 0.01 level (2-tailed).

There is, however another interpretation which is possible here. As ob-

served by Buchweitz and Alves (2006: 263), the amount of revision can
be used to infer to what extent the translators are “satisfied with their
work once they get to the last word”. In this sense it is possible that both
most experienced practicing translators were fully satisfied with their
work so far whereas the less experienced kept revising taking a consid-
erable amount of time and introducing a lot of changes in the hope that
through their conscious effort they will produce a more adequate transla-
tion. With all respect to their conscientious approach it seems that the
trainees are still in the process of developing their self-confidence not
only as translators but also as L2 users. In other words, another possible
explanation comes from the declarative procedural divide between their
L1 linguistic competence and their L2 use which has to be surmounted
in the translation process as a consciously controlled metalinguistic op-
eration involving transporting meaning across language barriers. Care
has to be put into not losing something essential on the way (see Whyatt
2007a) as well as to deliver the contents in an appropriate presentable
form for TL readers. This requires applying metalinguistic knowledge,
pragmatic knowledge and intercultural awareness as well as linguistic
competence. In short, revising also requires integrating knowledge but
372 Chapter 7

with different weightings attached to bring the receiver of the translation

in focus and to make sure that the translation will fit in with the TL real-
ity it will enter as soon as the finished product is sent to the client.
Emma Wagner (see Chesterman and Wagner 2002) gives some insight
into the strategies applied by professional translators working for the EU
translation services to distance themselves from their translations and
perform an efficient post-draft revision. Some of them include, for ex-
ample sitting in a different chair when revising, or looking at a print out
of the translation rather than revising on the computer screen.
Having devoted some attention to the distribution of time following
the actual end of the TL draft translation the revision aspect has not as
yet been exhausted. Bearing in mind the discussion on different revising
strategies (see chapter 5) it is important to analyze the so-called on-line
revision which will take into account all the changes during the TL text
production calculated by text elimination data as well as, what can be re-
ferred to as navigation within the TL text. Navigation within the text is
indicated by cursor and mouse movement during the actual typing of the
TL text. Buchweitz and Alves (2006) referred to what I call navigation
within the text using the term recursiveness which they explained as the
on-line revision of the text. All the features indicating movement within
the TL text show that translation is a non-linear process. Many practic-
ing translators would probably agree with Séguinot (1996) who ob-
served that even if the translator finds a solution his/her mind might
continue the search for a better solution and the changes are immedi-
ately introduced within the emerging TL text. Indeed, the study of the
translation process frequently shows signs of parallel processing when
the translator works on more aspects or items at a time (see Lauffer
2002: 70) and will move up and down within the TL text to introduce
changes which are aimed to improve the TL text under production. This
can be interpreted as yet another sign of the on-line metacognitive su-
pervision over the translation process which draws an insight from the
Knowledge Integration Network created for the translation task. Graph
12 below shows the data for text elimination in all the groups of sub-
jects.
 Empirical investigation into the development of … 373

180.00 162.4
152.6 146.2
160.00
140.00 123.6
120.00
100.00
80.00
60.00
40.00
20.00
0.00
beginners translation professional Total
trainees translators

Graph 12. Text elimination in the three groups of subjects.

It is the group of beginners (G1) which was most active in terms of text
elimination whereas the group of translation trainees was the least active
in terms of text elimination during the TL text production. On the other
hand, it was the group G2 which devoted the largest amount of time on
post-draft revision as well as introduced the most changes at the post-draft
revision stage. It is interesting to note the similarity between the begin-
ners and practicing translators in terms of the eliminated text and text
production during revision. It seems possible that the significant amount
of time devoted to changes in the post-draft revision by trainees is the
transfer of training effect when students are repeatedly reminded about
revising and improving their translations. Consequently, it seems that they
perform less on-line revision focusing on the completion of the first draft
which they will later revise taking as much time as they feel necessary.
Beginners in translation frequently change their mind while typing their
first draft possibly for two reasons; the first one being that they don’t have
enough communicative confidence in L2 (see Whyatt 2009a) or they aim
at the best possible solution while drafting their translation. The second
possibility is confirmed by the fact that they rarely revise their finished
translation showing low level awareness of translation as a purposeful
process and a functional product.
374 Chapter 7

The further data on cursor and mouse movement show that the groups
G1 and G2 move within the text more than the group G3 which seems fo-
cused on progression towards the end of the task. Graph 13 below indicates
that both the beginners and the translation trainees exhibit a lot of recur-
siveness which Buchwietz and Alves (2006: 246) described as, “a process
that at times, goes around in circles, without moving linearly forward”.

284.6 277.6
300.00 254.6
250.00
201.5
200.00
150.00
100.00
50.00
0.00
beginners translation professional Total
trainees translators

Graph 13. Cursor and mouse movement for the three groups of subjects.

The data showing that the more experienced translators exhibit the least
movement within the text point to the amount of self-confidence in their
decisions which as it was theoretically assumed stems from their ability to
integrate all the knowledge relevant for their metacognitve supervision
which guides their subsequent decisions. As suggested in chapter 5 having
created an adequate KIN the semantic priming effect facilitates lexical ac-
cess to the point that well experienced translators seem set on word
choices while typing the TL text and they rarely change their mind. Be-
low, there are samples of Translog files from representatives of the three
groups when the subjects were translating the same stretch of the SL text:

Marta (Krystyna Janda) jest żoną doktora (Jan Englert) z małego mias-
teczka. Pewnego dnia Marta poznaje znacznie młodszego od siebie męż-
czyznę, prostego robotnika Bogusia (Paweł Szajda), który oczarowuje ją
 Empirical investigation into the development of … 375

swoją młodością i niewinnością. Ich spotkania na brzegu rzeki porośnię-

tej tatarakiem są naznaczone wzajemną fascynacją. Jedno z nich zmierza
do przedwczesnego końca, drugie zaś wkracza dopiero w dojrzałość. Jed-
nak los obchodzi się z nimi okrutnie.

Translog file S8 (G1)

Marta ( Krystyna Janda ) [ 12.228]id ==⌫⌫⌫⌫s a

doctor's ( Jan Englert ) wife from the
provin ce ( M) [ 01:23.344][ ] ⌫
( I) [ 34.244]provincial town. Once ⌫⌫⌫e day
Marta meets [ 18.289]a (b)[ 01:07.
078] remar kably younger man[ 11.582] who is a
simple phisical worker named Boguś (Paweł
Szajda ) and who charms her with his
youthfull ⌫ness and inocence . [ 14.670]Ther⌫ir
m eetings at the river ba nk (b) [ 47.
854][ ] sawmill- co⌫⌫⌫ covered [ ][ ][ ][ ][ ]
[ ][ ][ ] [ ][ ] ⌫⌫⌫⌫
(b) [ ] are marked with [ 14.3
12]a (b)[ 12.381] mutual fas cination .
One of them is heading towe ⌫ards the premature
end , the second ⌫, on the other hand, [ 11.004]i
s just stepping into maturity[ 52.566], but the
faith is not [ 01:08.119]⌫⌫⌫⌫ cruel to them

Translog file S8 (G2)

Marta (Krystyna Janda) is doctor's wife[Ctrl ]

( Jan Engl ert) .[Ctrl ][Ctrl ][Ctrl ]
[Ctrl ][Ctrl ][Ctrl ][Ctrl ] a small town One
day Marta meets a man a lot yp⌫ounger than her
self [Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ]who is
⌫⌫⌫⌫⌫ [Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ct
rl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ] Bogu ś ⌫ ,
⌦⌦⌦worker [Ctrl ][Ctrl ][ShftCtrl ][ShftCtrl
]⌦far i[ 56.281][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl
][Ctrl ][Ctrl ][Ctrl ]simple [ 10.530]

, that ⌦ charmes her

376 Chapter 7

with his innocence and youth. [ 10.748]They

meet on a river bank covered with swee t flag[ 11.9
63] and they are fascinated about each other, ⌫⌫
. One is heading ⌫⌫⌫⌫⌫⌫⌫⌫appro-
aching as ⌫⌫ premature end, the other is stepping int
o maturi ty but the fate is cruel to the,m⌫
⌫m...

Translog file S1 (G3)

Marta (Kr⌫⌫starring Krystyna Janda) is the wife of a

doctor (starring Jan Englert) from a small town. One day
Marta meets a much younger man , a simple worker Buguś
(sta⌫⌫⌫Paeł⌫⌫we S⌫⌫ł Szajda_⌫), who enchants
her with his youth and innocence. Their meei⌫ting at
the bank of a river grown with se⌫week flag is mutuall
y facinating. One of them [ 13.236]is on the road to
pre⌫⌫⌫a premature end, the other ju⌫⌫is just
entering t⌫maturity . But their future
together is very painful...

To sum up, Translog as a new computational tool allows to demonstrate

pronounced differences in the TL text production performed in the proc-
ess of translating a SL text from the subjects’ native language (Polish)
into their foreign language (English). Since all the subjects were native
speakers of Polish and the SL text was fairly simple it is possible to as-
sume that all the problems with translating the text can be attributed to the
differences in the level of their translation expertise. As discussed in chap-
ter 4 translation expertise is a cluster concept and bilingual knowledge is
one of its fundamental components. The three groups of subjects differed
significantly in terms of their L2 proficiency which had consequences on
their translation performance. The most experienced translators are thus
also the most advanced L2 users. Unfortunately, there is a dearth of re-
search data available on how developing translation expertise affects the
organization of bilingual knowledge in the translating mind (see
Tymczyńska 2011). Similar, there has been not much investigation into
the role of metalinguistic abilities in the process of translation skill devel-
opment although Malakoff and Hakuta indicated in their 1991 study that
metalinguistic abilities might be decisive factors in developing translation
proficiency. It is quite likely that the metalinguistic abilities play a crucial
 Empirical investigation into the development of … 377

role in the revision process. The ability to distance oneself from one’s
own translation and observe with the fresh eyes of a reviser, editor and
proof-reader one’s own creation (TL text) seems a difficult skill to grasp
yet an essential part of translation expertise. The reason for this difficulty
lies probably in the parallel development of translation expertise and bi-
lingual communicative competence. Inexperienced translators who are
frequent research subjects are also inexperienced L2 users who focus on
controlling their L2 rather than the translation process with no capacity
left for the metalinguistic distance to the emerging TL text. As indicated
before, translation experience raises the metalinguistic awareness and by
putting one’s bilingual knowledge to the test enhances the communicative
competence in both L1 and L2. Consequently, when some L2 learn-
ers/users choose to pursue a career in language translation and embark on
a translator training programme they frequently become overcautious and
devoted to lengthy revising to ensure the adequate rendition of the SL
text. Their translation behaviour is strategic and consciously controlled
and so is their language use. Although the attitude is correct and appropri-
ate for the stage on the developmental continuum of their translation ex-
pertise their choices and final decisions frequently lack creativity in lan-
guage use. Practicing translators with many years of experience have
automatized a lot of language processing freeing in this way sufficient
mental resources to provide efficient metacognitve supervision over the
translation process, which as suggested in chapter 5 relies on the inte-
grated knowledge network (KIN) created for the translation task. Their
metalinguistic abilities play an important part in how the supervision con-
tributes to the speed and fluency of the TL text production. The results
observed in this study referring to on-line and post-draft revision and
navigation within the text further confirm that the role of metalinguistic
and metacognitive abilities requires further investigation.
Having discussed so far all the aspects of the translation process in
which beginners, trainees and practitioners differ, it might seem disap-
pointing to report that the differences between the three groups do not
reach statistical significance. Graph 14 (see Appendix 1) shows the results
of significant tests carried out with the analysis of variance (ANOVA).
The lack of significant differences between the groups of beginners
(G1), translation trainees (G2) and practicing translators (G3) can possi-
bly be explained by two factors: a small number of subjects in each group
(N8) and lack of homogeneity in group G3. However, the results showing
378 Chapter 7

no significant differences are by no means disappointing but on the con-

trary they point to some important findings. First of all, it is possible to
confirm that the relationship between the human skill to translate and the
level of L2 proficiency is not strong enough to decide about the transla-
tion performance features which were analyzed in this study. This view is
supported by the wide range of different results within each group of sub-
jects, also those which were fairly homogenous in terms of translation ex-
perience and L2 proficiency, G1 and G2. The individual differences were
the most pronounced within the group of practicing translators which was
the least homogenous in terms of translation experience and consequently
in terms of their bilingual knowledge including L2 proficiency, lexical ac-
cess and bilingual language control. For example, in terms of duration the
range was from 07:49.390 to 49:23.292, in terms of time prior to the onset
of translation the range was from 00:06.066 to 04:01.829, in terms of
post-draft revision 00:06.397 to 17:59.194 and in terms of changes intro-
duced during post-draft revision the range was from 0 – 256 keystrokes.
Further, in terms of text elimination calculated in signs the range was
from 47 to – 345 signs and in terms of navigation within the TL text the
subjects performed movements ranging from 20 – 1,442. In view of the
above wealth of idiosyncratic differences between the subjects the deci-
sive factor in the development of translation expertise might be the sub-
jects’ ability to build an adequate KIN in which bilingual knowledge
plays an essential but not a decisive part.
Considering the significant correlations between years of experience
and duration and the use of external sources and text production per min-
ute, the role of experience is confirmed in terms of the speed of language
processing in the translation process. What do translators learn from ex-
perience? It is possible to suggest that the experience of translation brings
about a reorganization of the translators’ ability to use (activate and inte-
grate) their knowledge structures (including their bilingual knowledge) to
create adequate knowledge integration networks (KINs). In consequence
of the repeated translation experience it is likely that a process of trans-
formation from a bilingual into an interlingual individual takes place (see
Grosjean 2002) within a developing translator. Perceived as an expert
learner and as a systems intelligent individual the translator learns to ef-
fectively and efficiently interact with the two linguistic systems (see chap-
ter 5). Growing translation expertise while relying on knowledge integra-
tion results in the development of the interlingual capacity (Toury 1995,
 Empirical investigation into the development of … 379

Presas 2000), the ability to assess similarities and differences between the
two working languages perceived not only as systems of knowledge but
as social practice shared by the language users.
To sum up, the results of Study 1 tentatively confirm the hypothesized
role of the translator’s ability to integrate knowledge for the task at hand.
The ability to integrate knowledge in a dynamic Knowledge Integration
Network (KIN) plays an important role in the evolution of translation as a
human skill. The key issue here lies in the activation of the translator’s in-
ternal mental library so that all the knowledge recognized by KIN as kin
to the task at hand can be ready for use when translating (see Paradis
2004, Heredia and Brown 2004, Jakobsen 2005). KIN establishes parame-
ters for the metacognitive control of actions when translators plan – exe-
cute – and rehearse their performance (Bell, 1991, Donald 2007b), KIN
operates on the conceptual level (CUCB Paradis 2007, Kesckes and Papp
2000, Pavlenko 2005), and results in a temporary functional linkage be-
tween the two languages at all levels of language use. Building a KIN in
the act of translating has a positive effect on the speed of the translation
process and most likely on the quality of translation as a final product, al-
though the quality of the product was not an object of analysis in this
study. Novice translators integrate only parts of knowledge (i.e., the lexi-
cal level) and the speed and quality suffer. Translators beginning their
professional life make a conscious effort to integrate knowledge which
extends the time of translation and calls for the extensive use of external
sources. To conclude, Study 1 provided evidence for the 4 hypotheses
formulated for the purpose of this study.

1) The ability to integrate knowledge for the purpose of a translation task

develops with experience as confirmed by the correlation between years
of experience and text processing features (duration, consulting external
memory sources and text production per minute).

2) The least experienced translators (G1) seem to have integrated only

part of the knowledge required for the translation task (i.e., only their bi-
lingual knowledge and more specifically their L2 lexical knowledge).
They possibly translated without the support of an adequate Knowledge
Integration Network (KIN) which slowed down the translation process
and resulted in frequent interruptions because of low L2 proficiency.
380 Chapter 7

3) The more experienced translators (G2) seemed to have made a con-

scious effort to build an adequate KIN which extended the translation
process and resulted in conscious metacognitive supervision with fre-
quent checking for confirmation to ensure adequate translation. Their
extensive revision is of particular interest and shows their growing
awareness of the translation process and care for the quality of their
translation product.

4) Practicing translators with many years of experience seem to auto-

matically integrate knowledge needed for the task at hand which speeds
up the translation process and makes it more fluent. Significant differ-
ences in the distribution of the text processing features between the trans-
lators with only one year in the profession and those with longer working
experience point to the developmental nature of the ability to integrate
knowledge and apply a Knowledge Integration Network in the translation
process.

Although the evidence is by no means strong and was not confirmed by

inferential statistical analysis due to the small number of subjects and a
wide range of idiosyncratic differences in the performance of subjects
from the three groups, it nevertheless provides empirical evidence for the
developmental nature of the human ability to translate. It can be tenta-
tively concluded that translation expertise is based on the growing ability
to integrate knowledge into active knowledge networks used for meta-
cognitve supervision of the translation process leading to translation as a
functional product. However, the development of translation expertise has
to be viewed as parallel to and interconnected with the development of
expertise in L2 communicative competence including also the metalin-
guistic abilities. More questions arise. If the development of translation
competence is basically a process of learning to integrate knowledge (bi-
lingual, intercultural, conceptual) then one could expect that inexperi-
enced translators could perform better if they were told to make a con-
scious effort to build a KIN before they translate a given text.
A further step to take is then to experimentally test if leading the trans-
lator to create a KIN for a specific translation task would influence the
translation performance to a significant extent. This purpose was pursued
in Study 2.
 Empirical investigation into the development of … 381

7.4. Study 2 – objectives, participants, procedure

The major objective of Study 2 was to further test the hypothesis that creat-
ing an adequate KIN will have a positive impact on the translation per-
formance of translators with short experience in translation. Following the
tentative conclusions from chapter 6 and Study 1 that the ability to integrate
knowledge for the purpose of translation constitutes an important part of
translation expertise an experiment was designed to test if an experimen-
tally induced KIN in the least experienced group of translators will have a
significant effect on the text processing features in their translation per-
formance. The subjects chosen for the experiment were 2BA students of
English as a foreign language who represented the same level of L2 profi-
ciency and translation experience as the group G1 in Study 1 and group
2BA in the questionnaire study discussed in chapter 6 (see chapter 6 for
more details about the 2BA subjects). The choice of participants was made
not only for the pragmatic ease of recruiting such subjects, but also because
they are the least likely to automatically create complete KINs as it was
demonstrated by Study 1. Translation trainees (G2) as well as the newly
employed translators make a conscious effort to integrate knowledge as
they have been taught to do so (cf. Levy’s metaphor of text as a picture of
reality, macro-level text analysis (Nord 1988/1991), etc.). Well experienced
practitioners create KINs automatically and possibly with the first encoun-
ter with the SL text and the translation brief. This time a much larger sam-
ple was chosen to heighten the chances for statistically significant results.
Altogether 48 subjects took part in the experiment.
The group of 48 subjects who were fairly matched in terms of age, trans-
lation experience and L2 proficiency was randomly divided into two experi-
mental groups: the KIN group and the NO KIN. Both groups consisting of
24 subjects were asked to translate the same text, which following the same
text type as the one used in Study 1 was a film review, of The Legend of
Zorro. The text of the review consisting of 131 words was taken from a
popular TV magazine and the translation brief instructed the subjects to
translate the review for an electronic English language version of the TV pa-
per. Both groups translated the SL text in the Translog user program which
recorded the translation performance just as it was the case in Study 1.
The major difference between the two groups was an artificially induced
KIN for the KIN group (N24) and lack of this manipulation for the NO KIN
group of subjects (N24). The KIN group was told that after they click start
382 Chapter 7

logging they will see the SL text which they should read but before they start
translating it, they will be asked to answer 10 questions referring to the text.
The questions were typed on a piece of paper and the subjects were asked to
look at them after they have read the SL text and provide brief answers. The
following questions were prepared for the subjects in the KIN group:

1. Is the story of Zorro familiar to you?

2. Have you seen the film?
3. Where is the film set?
4. What category does the film belong to?
5. Give a few words to describe the story:
6. Is the story of Zorro universal?
7. Give two names of similar fictitious characters:
8. In the most general terms who is fighting against whom?
9. Give three words to describe Zorro:
10. Finish the sentence in 3-4 words: ‘When I think about Zorro I can see…

The 10 questions were very simple and answering them the subjects had
to refer to their own prior experiences, judgments and ideas about the film
and its hero, Zorro. What is important however, is the fact that the ques-
tions were in English, the target language (the language into which the
text will be translated) and the subjects’ answers were also expected to be
in English. None of the subjects seemed to be surprised that the SL text
was in Polish and the questions referring to the text were in English which
shows that switching languages is a natural occurrence for EFL students.
The choice of language in which the questions were prepared was not ac-
cidental but intentional and aimed at helping the subjects to activate, inte-
grate and initially recode their non-verbal knowledge about the film and
the hero into English with the hope that through this an adequate Knowl-
edge Integration Network will be initially created. This as hypothesized in
chapter 5 should result in semantic priming by activating adequate con-
cepts and sending spreading activation down to the TL lexical level.
7.5. Study 2 – data analysis
The analysis of the data focused on the text processing features available
through Translog. The comparative analysis aimed at finding confirmation to
the hypothesis that creating a KIN prior to the onset of translation will have a
positive effect for the translation process. Following the questionnaire study
 Empirical investigation into the development of … 383

and the results of Study 1, it was expected that the translation process for the
subjects belonging to the KIN group will be faster and more fluent with
fewer instances of dictionary consultations. The analysis of variance
(ANOVA) mercilessly showed that the results were not what was expected.
7.5.1. KIN as a single variable

Most of the answers to the questions for the KIN group were very brief and
can be summarized as follows. The story of Zorro was familiar to all the
subjects, not all the subjects have seen the film, the majority have placed
the film as being set in the USA but there were some subjects who thought
it was Spain, the film category was named as ‘adventure’ or ‘adventurous’
film. The story was described as funny, action packed, interesting, surpris-
ing, uncomplicated, dynamic, etc. All the subjects agreed that the story is
universal. Among other fictitious characters judged similar to Zorro the
names which were frequently mentioned included: Batman, Superman,
Spiderman, Robin Hood, or the Polish hero Janosik. All the subjects gave a
brief account of the two opponents in various ways, for example: the good
against the bad, Zorro against Armand. The words given to describe Zorro
included a list of adjectives: brave, handsome, clever, noble, smart, fit, cou-
rageous, fearless, and reckless. The sentence: ‘When I think about Zorro I
can see…was finished in a multitude of ways, for example, ‘When I think
about Zorro I see a hero in disguise, his letter ‘Z’ or Antonio Banderas.
Although a lot of answers were similar, they included personal judg-
ments about the film, the story and the main hero. Consequently, it can be
assumed that all the subjects have activated in their memory their concep-
tual knowledge relevant for the SL text they were just about to translate.
The cognitive components kin to the content of the SL text must have
been activated in the subjects’ memory but whether or not they have been
adequately integrated for the task at hand remains unclear. However, the
further task to encode their answers in English which is the language into
which they will translate provided favourable conditions for the process
of knowledge integration including conceptual and bilingual knowledge
structures. Following this line of reasoning I assumed that the subjects in
the KIN group had created a relevant Knowledge Integration Network
which as it was hypothesized in chapter 5 should facilitate the translation
process, speed up language processing and lexical access. The KIN would
provide essential information for the metacognitve supervision that will
be switched on until the translation task is successfully completed.
384 Chapter 7

To test this hypothetical assumption the translation performance of all

the subjects from the KIN group was recorded in Translog files and
matched against the translation performance of an equal number of subjects
from the NO KIN group. The No KIN group translated the same text with
the same translation brief but they did not receive the list of questions that
is, there was no induced KIN involved in the task. All the translations were
performed on five computer stations in the computer lab which was re-
served for that purpose. The conditions were similar to those described in
Study 1. The subjects were asked to work at their usual pace and they were
told that they are participating in a translation process awareness project.
During single sessions the subjects belonged to the same group, either the
KIN or NO KIN group so that they would not be aware of the differences
in the experiment design. The researcher was tactfully present all the time
in case the subjects needed assistance but every care was taken not to inter-
rupt the privacy of work carried out by the subjects. All the subjects worked
very conscientiously, they were extremely involved, focused and devoted to
do their best and they did not interrupt each other in any way. None of the
subjects requested a break and they all worked until they finished and
clicked ‘stop logging’ and saved their Translog file.

7.5.2. Duration and other text production features

Since the two groups KIN and NO KIN differed in one respect (single vari-
able), namely the KIN group had to answer 10 questions in writing and an-
swering them required consulting their knowledge base, thinking, making
judgments and formulating opinions prior to translating the film review, it
was expected that the KIN group will on average take much longer to per-
form both tasks than the NO KIN group which was asked only to translate
the text. Looking at the raw data, answering the questions took from
03:08.070 to 10:03.116. At the very start the KIN subjects were then from 3
minutes to 10 minutes behind the NO KIN subjects. If they managed to catch
up with the KIN subjects anyway it would mean that the time spent on build-
ing the KIN paid off. Graph 15 shows that answering questions did not make
the KIN subjects lag behind the NO KIN group. The time taken to read and
write down answers to the 10 questions in a way paid off and the KIN group
was slightly faster, although the difference is statistically insignificant. Like it
was in Study 1, the reasons might lie in still statistically small samples (data
gathered from 50 subjects in each group might show significant differences)
 Empirical investigation into the development of … 385

and in major idiosyncratic differences between the subjects in both groups.

As Graph 15 (see Appendix 1) shows indeed there was no significant differ-
ence not only between the duration of the task in both groups but also with
respect to other features available through Translog.
The time taken to perform the entire task was slightly longer for the
KIN group which shows that the extra cognitive tasks of activating and
integrating relevant knowledge prior to the onset of translation introduced
only slight differences (under 3 minutes) between both groups. The slight
difference in duration is lower than it was predicted taking into account
the range of time the KIN subjects took to answer questions. Statistically
the difference is insignificant. It is the other features of the translation
process that were more disappointing when they showed no significant
differences between the two groups. Significant correlations though con-
firmed the mutual relationship between duration and text production fea-
tures and are visible in Graph 16 (see Appendix 1).
The strength of the correlation between duration and TPM (r =
−0,929) clearly shows that the lower TPM the longer the time spent on
translation. The correlation between duration and dictionary use (DU) is
also very strong (r = 0.581), has the opposite direction and confirms that
the more use of dictionaries, the longer time was needed to complete the
task. Another significant correlation was revealed between dictionary use
(DU) and typing speed (TPM). The negative direction of the correlation
confirms that the more frequent use of dictionaries affects the TPM to a
moderate strength (r = −0,535). The remaining features, text elimination
(TE) and navigation in the text (CN+ME) did not reach a significant level.
However, there was a strong correlation between navigation in the text
and text elimination (r = 0,371) pointing to the tendency of beginners in
translation tasks to frequently change their minds when drafting their
translation (cf. group G1 Study 1).
If the effect of KIN was as predicted able to facilitate the translation
process making it more fluent the KIN subjects should have consulted
dictionaries less and should have shown less uncertainty reflected in text
elimination, and a considerably higher rate of Text Production per Minute
(TPM) than the NO KIN subjects. The next step to search for significant
differences was taken. The time used for reading and answering questions
(induced KIN) by the KIN subjects was then subtracted for each individ-
ual translator and the results are demonstrated in graph 17.
386 Chapter 7

0:36:00 0:32:17 0:30:42

0:29:07
0:28:48
0:21:36
0:14:24
0:07:12
0:00:00
KIN no KIN Total

Graph 17. Duration for both groups after the time for inducing KIN was
subtracted.
The graph shows that the KIN group was on average a bit faster in their
TL text production although the difference does not reach statistical sig-
nificance in the analysis of variance (ANOVA). Similar to the results ob-
tained in Study 1, a look at the raw data shows vast idiosyncratic differ-
ences between the subjects in both groups which similar to Study 1 are
obliterated in the results delivered by inferential statistics. To illustrate
this highly personalized translation performance Graph 18 shows the du-
ration for the slowest and the fastest translators in both groups.

1:04:48 00:59:37 KIN

0:57:36 00:51:38 no KIN
0:50:24
0:43:12
0:36:00
0:28:48
00:19:57
0:21:36 00:16:58
0:14:24
0:07:12
0:00:00
The slowest The fastest

Graph 18. Duration for the slowest and the fastest translator.

As it could be expected the idiosyncratic differences visible in duration

were also reflected in other text production features. In line with the re-
sults obtained in Study 1 the most significantly correlated features in-
 Empirical investigation into the development of … 387

cluded the use of dictionaries and text production per minute (TPM). Ta-
ble 7 presents raw data for the slowest and the fastest translators in both
groups with the data for a practicing translator with 30 years of experi-
ence in providing translation services (subject S1 G3 from Study 1) who
also translated the SL text used in Study 2.
The tavle shows the features of TL text production which are the most
time consuming for participants in Study 2 and the most time saving for
an experienced professional translator.
Table 7. Differences in text production features between the slowest and
the fastest translators in both groups matched with the data for a practic-
ing translator with 30 years of experience.
Translation performance Dictionary use Text elimi- Navigation TPM
nation in the text
KIN the slowest 21 277 166 28.45
KIN the fastest 4 44 93 58.72
NO KIN the slowest 14 65 193 25.52
NO KIN the fastest 7 46 135 69.74
PT 0 71 20 115.77

Consequently, the differences in duration (graph 19) are impressive and

again the correlations with other important text production features affert
the amount of time needed to train translate the SL text.

1:04:48 0:59:37
0:57:36 0:51:38
0:50:24
0:43:12
KIN
0:36:00
no KIN
0:28:48
0:19:57 PT
0:21:36 0:16:58
0:14:24 0:09:59
0:07:12
0:00:00
The slow est The f astest PT

Graph 19. Duration for the slowest and the fastest translator matched with
the duration for a practicing translator [PT] with 30 years of experience.
388 Chapter 7

The differences between the text production features visible in the raw data
are quite striking even between the fastest translators in the KIN and NO
KIN groups and a well experienced practicing translator. It is therefore con-
firmed that experienced professionals save time on such translation per-
formance features as dictionary consultation, excessive navigation within
the TL text and faster typing speed. These time saving factors can be a con-
sequence of the proposed ability to integrate knowledge for the task at
hand, which develops in practicing translators parallel to the professional
experience and due to their conscious effort to reach the level of translation
expertise. As it was tentatively concluded in chapter 6, building a KIN
which is a natural and fairly automatic goal oriented process for a well ex-
perienced translator facilitates the translation process in terms of the time it
takes to translate a text and the fluency of the TL text production process.
As suggested in chapter 5 the ability to integrate relevant knowledge devel-
ops with translation experience and it should not come as a surprise that it
cannot be artificially created in novice translators. The disappointing results
in Study 2 confirm this suggestion. One more attempt was made to find
significant results between the KIN and NO KIN groups. The two slowest
and two fastest translators were removed from each group which meant that
the number of subjects was reduced to 20 (N20) per group for further
analysis. Having removed the most peripheral subjects no significant dif-
ferences became apparent in all other features apart from two. As it is visi-
ble in table 8 the two groups now differed significantly in terms of events
per minute that is in terms of the speed of the overall keyboard activity and
in terms of TPM.

Table 8. Significant differences between the groups.

Independent Samples Test

t-test for Equality of Means
Mean Std. Error
t df Sig. (2-tailed)
Difference Difference
Text elimination 0,784 38 0,438 19,850 25,322
Total User Events 0,458 38 0,650 62,850 137,238
Text production 0,019 38 0,985 0,600 31,502
Cursor navigation 0,231 38 0,819 25,750 111,535
Mouse Event 1,286 38 0,206 16,550 12,871
Navigation in text 0,387 38 0,701 42,300 109,265
 Empirical investigation into the development of … 389

Dictionary Use -0,551 38 0,585 -1,250 2,269

Events per minute 2,183 38 0,035 12,83650 5,87906
Text Production per minute 2,471 33,105 0,019 8,52250 3,44903
Duration 0,170 38 0,866 00:00:30 00:02:53

It was the NO KIN group which performed significantly more actions per
minute than the KIN group. Could this be taken as evidence lending sup-
port that the KIN group knew better what they were looking for following
the knowledge integration effort which was induced in the experiment de-
sign? Even if it could, it is evidence which follows the manipulation of
raw data as the previous samples (N24) did not produce significant differ-
ences. More keyboard activity for the NO KIN group can mean that their
translating was more dynamic and included more hit and miss choices.
To sum up, the results in Study 2 did not support the hypothesis that
an artificially created KIN will facilitate the translation process in inexpe-
rienced translators in terms of speed and fluency. This result although ini-
tially disappointing can be explained in terms of the developmental con-
tinuum alongside which the human skill to translate develops. The evolu-
tion of translation as a human skill from a natural ability to expertise is a
developmental process. Essentially, it is a learning process in which the
translator as an expert learner (see chapter 5) self-discovers (either sup-
ported by structured training or self-learning) how to solve problems and
make decisions in the most effective and efficient way. Every new com-
mission and every new text for translation is the source of new experi-
ences which teach the reflective translator (expert learner) to make in-
formed choices relying on all the knowledge that can be made available
and integrated into a responsive KIN. The results of Study 2 confirmed
that the inexperienced and experienced translators look at the translation
task from a different vantage point and therefore have different priorities.
The novices are devoted to making the best most ideal choices at the level
of words therefore for them the words matter and since their bilingual
knowledge is still in the making they search for confirmation of their
choices, change their mind and make an admirable mental effort to bridge
the language gap and complete the translation task. It is clear that the 2BA
subjects do not feel at home in their L2 and integrating only their bilin-
gual knowledge takes, if not entire then a substantial part of their mental
space. The lack of communicative competence and communicative confi-
dence in L2 use meant that the subjects who were induced to perform
390 Chapter 7

considerably more information processing including their cognitive and

linguistic level were neither faster nor more fluent in the TL text produc-
tion process than the subjects who went straight into the translation proc-
ess. A glance at the Translog files of randomly chosen translators from the
KIN and NO KIN group illustrates this point (see Appendix 3).
In contrast, the well experienced translators, among other things feel
more at home with their L2 and they have discovered that often there is
no single ideal solution to a translation problem but there is a range of
choices. It is possible that their ability to integrate all the relevant knowl-
edge into an adequate KIN activates only those TL lexical items which
match their conceptual intentions. With time and conscious effort to use
their both languages as efficient experienced translators they become
more confident about their choices. Having crossed the language barriers
many times they develop trust in their decisions and discover that it is not
the particular words which are selected that matter but the function they
perform in the conceptually driven translation process. In a way the lexi-
cal choices and other text processing features are subject to higher order
thinking which guides metacognitve supervision over the entire transla-
tion process. As hypothesized in chapter 5 and confirmed in chapter 6 the
metacognitve supervision relies on an adequate Knowledge Integration
Network created in the translating mind. The results obtained in Study 2
show that the ability to integrate knowledge for the purpose of translating
is developmental in nature and cannot be artificially induced in translators
with low levels of translation proficiency. It is most likely that the evolu-
tion of the ability to integrate knowledge is a slow process which requires
conscious involvement of the translator as an expert learner who through
repeated translation experience, corrective feedback (external in formal
instruction or internal in self-development) and self-reflection learns to
build Knowledge Integration Networks for the purpose of translation.
Then just as an expert piano player he or she can focus on the entire out-
come, the overall impression while at the same time being sure that
his/her metacognitve supervision will spot and see to all the problems at a
lower level of lexical choices during the physical execution of the transla-
tion process. Inexperienced translators still spend a lot of time looking for
an appropriate key (e.g., word) to offer the most accurate lexical match
between the SL text word and the TL text word rather than create a de-
sired effect using other linguistic means.
 Empirical investigation into the development of … 391

7.6. Conclusions

The purpose of this chapter was to provide empirical evidence to confirm

that the development of the human ability to translate to reach the level of
translation expertise relies on the ability to integrate knowledge for the
purpose of translating. Consequently, the less experienced translators
were assumed to be less skilled at building adequate knowledge integra-
tion networks or even less able to activate all the relevant knowledge
structures and integrate them when translating. The process of knowledge
integration cannot be directly observed as it takes place in the translator’s
mind in a way which can perhaps be demonstrated by a connectionist
model of cascaded multilayered activation spreading into the translator’s
long term memory and triggering both conscious and unconscious thought
processes. However, following the principles of cognitive psychology and
psycholinguistics that mental processing can be inferred from behavioural
data an attempt was made to infer the adequacy of KIN from the actual
translation performance. Using a computer software program, Translog as
a research tool allowed accessing many pieces of information so far inac-
cessible in translation process research. Such data as the actual duration
of the translation process, which starts with the opening of a source lan-
guage text and ends with the final decision that the translation product is
finished, could be compared across the subjects who participated in the
two studies described and discussed in this chapter. Then through analyz-
ing such data as time taken to solve translation problems (including the
use of dictionaries, the amount of text eliminated during the TL text pro-
duction, the amount of navigation within the TL text exhibited by cursor
and mouse movement, and the typing speed) conclusions were drawn re-
ferring to the hypothesized ability to integrate knowledge.
Study 1 investigated three groups of translators who represented sig-
nificantly different stages in the process of translation skill development.
All the subjects were asked to translate a self-contained short text, a film
review, from Polish into English using the Translog program.
The data analysis carried out with statistical tools for inferential analysis
(ANOVA) showed no significant differences between the three groups of
subjects in all text processing features chosen for analysis, that is duration,
dictionary consultation, typing speed, on-line revision and post-draft revi-
sion. This result, however, was not taken as sufficient evidence to refute the
hypothesis that the ability to integrate knowledge for the purpose of a trans-
392 Chapter 7

lation task develops with experience. On the contrary, attention was drawn
to the individual differences between the subjects within each group which
were the most pronounced for the group of practicing translators. The raw
data obtained through Translog show that although statistically all the sub-
jects took a similar amount of time to translate the SL text, the time was
differently distributed throughout the translation process. The G1 group
spent the least amount of time to open the SL text (time prior to the onset of
translation) and to revise the TL text during post-draft revision and elimi-
nated the largest amount of text during the TL text production. This shows
that they most likely have not created an adequate knowledge integration
network (KIN) and their TL text production is not fluent, with frequent dic-
tionary consultations and a lot of navigation within the actual TL text dur-
ing the process of its production. The G2 group of translation trainees took
much longer to open the SL text and devoted much more time to revising
the TL text during the post-draft revision stage. They also introduced the
most changes when revising the TL text. They also frequently used external
sources to confirm their hypothesized solutions and there was a significant
amount of movement within the emerging text. Their controlled effort es-
pecially before the onset of translation and during the post-draft revision
can be interpreted as showing a conscious process of knowledge integra-
tion. However, their low L2 communicative confidence calls for using ex-
ternal sources to fill gaps in L2 communicative competence which slows
down the translation process. As a result, although they consciously built a
KIN they do not perform faster than the G1 group. The G3 group of prac-
ticing translators is the least homogenous in terms of both translation profi-
ciency and L2 communicative competence. It is also a group with most
pronounced differences in text processing features analyzed in Study 1. A
closer look at the correlations between years of experience and text produc-
tion features for individual translators reveals significant interdependencies
between years of practice and the time needed to translate the SL text used
in Study 1. The two most experienced translators (S1 with 30 and S2 with
20 years of practice) took the least time to translate the SL text, did not use
dictionaries, had the highest rate of text production per minute and spent
hardly any time to revise the TL text. This result allows the assumption that
all the knowledge which was needed to translate the SL text was made
available for their working memory via activation and integration processes
which resulted in an adequate KIN for the translation task ahead. Conse-
quently, the translation process was fast and fluent and resulted in a TL text
 Empirical investigation into the development of … 393

highly acceptable in their estimation as they both decided to leave it with-

out any post-draft revising. They however both admitted that they would
normally look at the TL text some time later, confirming in a way the opin-
ions of 40 practicing translators analyzed in chapter 6. As it was suggested
in chapter 6, leaving a draft translation aside before revision might indicate
the need to re-frame from a KIN created for the purpose of translating to a
KIN most effective for revision. The least experienced practicing transla-
tors provided the most intriguing results. In many respects they performed
very similar to the group of translation trainees and their reliance on exter-
nal sources shows their limited L2 communicative confidence as well as
low self-confidence as translators. The time devoted to post-draft revision
confirms their care for the quality of the emerging TL text but also displays
a lack of self-confidence in their decision-making. Consequently, the trans-
lation process is frequently disrupted and the insufficient KIN does not
yield productive results in terms of channeling lexical access to TL items.
A probably very obvious observation has to be made that translation
performance is very closely linked with L2 communicative confidence
which is qualitatively different from communicative competence (Whyatt
2009a). Unquestionably, the fact that well experienced translators are also
better L2 users is not to be underestimated. As shown in chapter 6 a vast
majority of practicing translators are aware of the fact that through their
translation practice the size of their bilingual mental lexicon is constantly
expanding. The results in Study 1 showed that it is not only the size of the
mental lexicon which keeps growing with practice but also the amount of
information about the words referring both to their detailed semantic
specifications, pragmatic behaviour and information about their gram-
matical behaviour in context (lemmas). This comprehensive knowledge
allows one to make quick informed decisions rather than vague guesses
which have to be confirmed by checking words in a dictionary.
The effect of the translation experience on the structure of bilingual
memory and its functional specialization requires a more thorough investi-
gation and the implications obtained in Study 1 confirm the need for further
research. The speed and efficiency of lexical access in professional transla-
tors shows that they have paved many interlingual routes in their bilingual
memory which remain unpaved for inexperienced and novice translators.
Contrary to Paradis (1994: 323, 2009: 180) however, who suggested that
well experienced simultaneous interpreters translate using only cross-
linguistic lexical links skipping the conceptual level, a strategy which may
394 Chapter 7

compromise the communicative quality of the emerging translation, I am

more inclined to expect after Séguinot (1997: 117) that both forms and
conceptual content can be stored connected across languages in the long-
term memory of experienced translators. Some connections enhanced by
experience (e.g., translation done previously on the same topic, and/or in-
cluding the same text type) can be very strong and therefore highly respon-
sive to the knowledge integration network (KIN) created in the translating
mind. Other connections both at the level of concepts and word forms can
be accompanied by a warning tag from previous translation experience in-
forming the translator to keep a watchful eye on possible cross-linguistic
interference (see Chmiel 2004) for which translating creates excellent con-
ditions by keeping two languages simultaneously activated.
It was my intention to demonstrate that the control function in the
translation process is performed by the metacognitve supervision which
relies on the adequately activated and integrated knowledge in the trans-
lating mind. Study 1 showed that the ability to create adequate KINs de-
velops with translation experience. Very experienced translators reap the
benefits of adequate knowledge integration both declarative (meanings of
words, knowledge of text types and various kinds of encyclopedic infor-
mation relevant to the subject matter of the SL text) and procedural (L1
and L2 competence, accessing relevant procedures to make chosen words
fit the contextual patterns, typing skills, self-monitoring and self-
correction). As a result their translation process is fast and fluent. Much
less experienced translators had problems with integrating even their bi-
lingual knowledge and more experienced translation trainees or beginners
in the profession put a lot of effort into consciously building an adequate
KIN but they still displayed a lot of uncertainty in their decision making.
The quality of the translations produced by the three groups of subjects
was not analyzed in the study. Nevertheless, a qualitative analysis could
yield more supportive evidence for the consequences of the hypothe-
sized ability to integrate knowledge provided that it is carried out fol-
lowing objective quality assessment procedures5. To provide some in-
sight into the TL text quality samples of TL texts from each experimen-
tal group together with their Translog files and the SL text are presented
in Appendix 2.

–––––––––
5
See charper 5 on problems with translation quality assessment.
 Empirical investigation into the development of … 395

Study 2 showed that an experimentally manipulated KIN for a task

ahead did not result in the improved translation performance of 2BA EFL
students. The KIN group (N24), which had to perform a task aimed at ac-
tivating relevant concepts and triggering a priming effect to TL lexical
forms after reading the SL text and before starting to translate it, did not
perform significantly better in terms of the speed and fluency of the TL
text production. The NO KIN group of subjects who were only translating
the SL text resembled the KIN group on all the text production features
apart from the overall keyboard activity and text production per minute
which was higher for the NO KIN group. This could suggest that the KIN
group was more channeled and more tuned in with the task whereas the
NO KIN group displayed more hit and miss tactics in their TL text pro-
duction. Possibly a qualitative analysis bringing together the data on the
process and the assessment of the product, the actual TL text could yield
more informative results on the effects of the experimentally induced
KIN. A qualitative analysis however was not intended either in Study 1 or
in Study 2. Translog as a new research tool allows for quantitative analy-
ses of the translation process by providing objective data whereas a quali-
tative analysis would introduce subjectivity in the choice of criteria used
to assess translation quality (see House 2001: 257). Using a common
sense approach it can be assumed that all the translations performed by
the Study 2 subjects have similar flaws as the ones described in chapter 3
discussing translations performed by L2 learners (see Appendix 3 for the
SL text used in the experiment and samples of translations with Translog
files). However, investigating lexical access in both groups could provide
a broader picture of the relevance of making an effort to activate and inte-
grate relevant knowledge (in the KIN group) before embarking on a trans-
lation task. This issue is open to investigation in my future studies.
To sum up, it seems that the human skill to translate relies on broad
cognitive abilities which can be initially present in some individuals (as
early as at the stage of translation predisposition and natural translation
ability) and which will continue to develop parallel to translation experi-
ence. Comparing the performance of translators at different stages on the
developmental continuum of translation skill it seems plausible to assume
that the differences in the TL text production features reflect their differ-
ent abilities of activating and integrating knowledge which is necessary
for efficient metacognitive supervision of the translation process. The link
between translation expertise and the control over the translator’s bilin-
396 Chapter 7

gual knowledge is clearly visible in the correlations between years of

practice and the time needed to produce a translation, the use of dictionar-
ies and text production per minute. Although the conclusion that the rela-
tionship between bilingual knowledge and translation experience is that of
mutual benefit may seem obvious, it points to the possible functional re-
organization and refinement of bilingual memory structures. Such func-
tional rewiring of the bilingual knowledge structures in expert translators
is consonant with suggestions made by Presas (2000) and Toury (1995)
pointing to “interlingual bridges” or developing “interlingualism” as a
consequence of repeated translation practice (see Tymczyńska 2011). A
further investigation into the nature of this mutual interdependence can
yield interesting information on bilingual language processing, bilingual
memory and language control in the bilingual mind.
The results of both studies point to the evolutionary nature of the proc-
ess in which the human skill to translate develops to reach the level of ex-
pertise. The hypothesis that the development of expertise in translation de-
pends on the ability to integrate knowledge for a task at hand seems to have
been validated by the Translog data on TL text production features. How-
ever, the results show that it is a slow cognitive process. There was hardly
any difference between the 2BA group (G1) and the Translation trainees
group (G2), and the novices in the profession in terms of duration and the
use of external resources. Clear differences appeared only in the group of
practicing translators. It seems that the development of translation expertise
can be seen as a learning process in which the translator with all the quali-
ties of an expert learner develops his/her self-confidence as an intercultural
mediator in otherwise impossible communication. It is a process of self-
involved conscious effort to acquire expertise which relies on declarative
and procedural knowledge integration (cf. Alves and Gonçalves 2004: 426).
Observing the performance of translators at different stages on the devel-
opmental continuum reveals important facts about translation as an activity.
Everybody who has access to two languages can translate, but in the proc-
ess of translation there will be different processing patterns which can be
objectively analyzed through Translog. Still, many questions remain includ-
ing those most fundamental: Can we, as translator trainers’ speed up the

–––––––––
6
Alves and Gonçalves (2004: 42) introduced a concept of translator’s rather than
translation competence. This change of perspective allows to account for the individual-
ity of expertise acquisition stressed in this analysis.
 Empirical investigation into the development of … 397

process in which future translators learn to integrate all the knowledge rele-
vant for a task at hand? Can we structure the training programmes in a way
which gives future professionals the best possible learning environment to
raise their self-awareness of their role in the learning process? Many voices
have been raised putting forward valuable suggestions. Kiraly (2000, 2005)
stresses the importance of realistic translation assignments, González Da-
vies (2004) opts for collaborative learning, Piotrowska (2007) suggests fo-
cusing on a strategic approach to translation tasks. With the research dis-
cussed in this chapter I add the implication of raising the awareness of the
developmental route to translation expertise and engaging the expert
learner’s self-involvement in the developmental journey from natural pre-
disposition to translate to expertise in translation. Looking at the translation
performance discussed in Study 1 and Study 2, it seems that this knowledge
is not subject to explicit instruction in an old-fashioned transmissionist
manner. For many translators-to-be it remains a process of self-discovery
and self-development also confirmed by the voices of practicing translators
analyzed in chapter 6. What can be done, however, is to make the implicit
learning process more explicit by raising the trainee’s awareness of it taking
place in their learning minds. The proposal of the ability to build Knowl-
edge Integration Networks implies that translation training practice apart
from real-life translation assignments7 could also apply knowledge integra-
tion for the task at hand in the form of simulated translation performance.
Decomposing a KIN into specific knowledge components including con-
ceptual knowledge (Common Underlying Conceptual Base (Kesckes and
Papp 2000) and domain specific knowledge) and bilingual knowledge (in-
cluding all levels of language as well as norms referring to social practice)
could allow to practice knowledge integration with focusing on intercon-
nections between various knowledge components, i.e., conceptual-lexical
within L1, L2 and including both languages. The fluency and speed of lexi-
cal access can be practiced for any chosen translation task prior to the onset
of the actual translation process. In other words, knowledge integration
tasks can be used in a similar manner brainstorming is used for practice in
creative writing. Putting beginning translators in the right frame of mind
through knowledge integration can warm up their cognitive resources and
help them in the translation process.

–––––––––
7
which as pointed to in chapter 4 are valuable but extremely time-consuming (see
Kiraly (2005) on his subtitling project).
398 Chapter 7

To provide a clear methodology of how to best structure the tasks in

which future translators can learn to integrate all relevant kinds of knowl-
edge needed for a translation task at hand more research is needed involv-
ing all the disciplines that Translation Studies are related to, SLA, psy-
chology, psycholinguistics, neurolinguistics and cognitive studies. Find-
ing synergies between language-related disciplines could further validate
the hypothesized ability to integrate knowledge into task-specific Knowl-
edge Integration Networks in expert performance. Translation after all is
an operation on sometimes very subtle information encoded in languages,
decoded in the translating mind and enciphered in another language for
others, for other generations for humanity. It is my conviction that the
service translators have provided throughout history makes them well in-
formed research subjects irrespective of the stage they occupy on the de-
velopmental continuum of translation as a human skill. Investigating how
translators process information encoded in languages can inform us about
the routes of development of human intellectual potential.
Conclusions and Implications

The aim of this book was to show the human ability to translate in its evo-
lutionary perspective. Suggestions were made that the human mind is in-
trinsically a translating mind which constantly is involved into decoding
and encoding meaning to meet the needs of social interaction and com-
munication. For the mind to be truly mind-sharing (Donald 2001) lan-
guages came to serve the purpose of communication. While language
shared by its community fulfils the communicative function it at the same
time creates a barrier to all people who do not share its knowledge as a
system or as a social practice. Most people have experienced this uncom-
fortable feeling of being barred from accessing information if it has been
encoded in an unknown language. Since, however, the human need to
communicate is possibly one of the strongest human desires, and follow-
ing the popular saying, ‘If there is a will, there is a way’ translation be-
came the way to overcome barriers imposed by languages.
Translation as a broad cluster concept involves many means (interlin-
gual, intralingual, intersemiotic) and different modalities (written, oral,
sign translation). Yet, with the exclusion of the still narrowly applied area
of machine translation, it is performed in the human mind before its so-
cially required product can enable an otherwise impossible exchange and
flow of information. Because everybody who has access to two languages
can translate, in the sense that he or she is able to perform this operation,
translation as a human skill is frequently perceived as a simple process of
transcoding, and it is therefore socially expected of bilinguals or multilin-
guals. A more detailed analysis of the different demonstrations of the hu-
man ability to translate, however requires to see translation skill in its de-
velopmental continuum from the human predisposition to translate to pro-
fessional expertise. The necessity to distinguish between such concepts as
translation ability, translation skill, translation competence and translation
expertise becomes essential to avoid confusion and misunderstanding.
400 Conclusions and Implications

Chapter 1 discussed these concepts and established common grounds for

a detailed analysis of the human ability to translate in its developmental
continuum.
Since interlingual translation as a communicative strategy involves the
knowledge and use of two different languages by one person, his/her bi-
lingual knowledge constitutes the foundations on which translation skill
and translation competence can be built. A thorough analysis of the under-
lying bilingual knowledge conducted in chapter 2 showed its internal
complexity and dynamic dependence on other factors such as language
acquisition history and the psycholinguistic processes of cross-linguistic
interaction. Discussing such essential concepts as bilingual memory, lan-
guage control and the cognitive consequences of bilingualism was con-
sidered essential to understanding the developmental nature of translation
as a human skill. This assumption was confirmed in chapter 3 which dis-
cussed translation as an untrained ability of bilinguals.
No matter whether a second language was acquired in a natural process
of acquisition or through learning, the users of two languages can act as
natural translators in the sense that when translating they rely on their bi-
lingual knowledge and have not received any training in translation. Trans-
lation is a widespread social phenomenon, a part and parcel of people‘s
lives as bilinguals in everyday interaction in multilingual communities and
natural translators provide a vital link to keep the communication going.
Acting as language brokers they can produce communicatively functional
translation although as the scant research has shown the form it takes does
not meet professional standards of quality. While natural translators do not
attach importance to the formal aspects of language and focus on meaning
transfer, L2 learners acting as natural translators attach more importance to
the form than to meaning. However, as research into L2 translation has
shown in the course of translation practice they fairly quickly learn to ad-
dress the form/meaning imbalance and as a result, and quite incidentally
they start to see their bilingual knowledge from a more holistic perspective.
Translation experience even if it is not aimed at training L2 learners as
translators but used as a pedagogical tool enhances their metalinguistic
abilities and helps them acquire intercultural competence. These obvious
assets of translation experience point to a transformation which has to occur
for the further development of translation as a human skill to become pos-
sible. Since, however, the natural ability to translate has not received ample
 Conclusions and Implications 401

attention from translation scholars its potential remains to be utilized and

can be expected to shed more light on the nature of transformation from a
bilingual into an interlingual individual. A point is made that in order to
evolve into a trained skill with its underlying professional competence a bi-
lingual individual has to make conscious intentional effort to reorganise and
restructure his/her bilingual foundations.
For those who choose to develop their untrained ability to translate into a
trained skill allowing for fluent performance the notion of translation com-
petence becomes of vital importance. A detailed discussion of one of the
most frequently discussed concepts in Translation Studies shows that Trans-
lation Competence (chapter 4) has defied an acceptable definition that
would satisfy, if not all then at least the majority of scholars. Whether seen
from an additive, componential or minimalist perspective Translation Com-
petence is seen as experience-derived knowledge allowing for fluent and
professionally adequate translation. Yet, the process of acquiring this in-
sider’s knowledge of how to produce professionally satisfying translations
cannot be dismantled into stages for a number of reasons. One of them is the
difficulty in defining the constituent concepts of translation quality and pro-
fessionalism in translation. In today’s world the need for translation services
is constantly growing, translator training institutions are set up and they
have to cope with the ‘pedagogical gap’. To counterbalance the lack of
guidance from translation theory effort is made to provide the trainees with
real-life translation experience. By incorporating the ideals of social con-
structivism and collaborative learning it is hoped that the trainees will for
themselves acquire the necessary knowledge to enter the translation market.
A glimpse into translation as a profession points to the extremely high re-
quirements imposed by translation training institutions which are considered
important for those who want to become professional translators. Many
points on the developmental route of translation competence acquisition re-
main unclear. Such points include different routes to translation competence,
including self-coaching and translator training, and personality factors. A
suggestion was made that the notion of the translator as an expert learner
might be able to account for a multitude of factors involved in translation
competence acquisition which apart from the underlying knowledge gives
one the personal self-confidence to provide translation services.
Chapter 5 discussed the possibility that translation as a human skill to
reach the level of professional competence relies on the ability to integrate
402 Conclusions and Implications

knowledge into what I termed a Knowledge Integration Network. The theo-

retical foundations of the Knowledge Integration Network were discussed
taking into account such assumptions as the limited capacity of the human
mind, automatized and controlled processing and meta-cognitive supervi-
sion. The hypothesized KIN is a dynamic network which facilitates the
translation process and enables the retrieval of information from the trans-
lator’s long term memory including the necessary linguistic means to pro-
duce a TL text. When faced with a specific translation task the translator ac-
tivates and integrates all the knowledge that he/she perceives as relevant for
(i.e., kin to) the task at hand. The ability to create adequate Knowledge In-
tegration Networks is developmental in nature and translators differ in their
knowledge integration ability depending on the stage they currently occupy
on the developmental continuum of translation as a human skill. The role of
adequately integrated knowledge was pointed out in a detailed comparison
of translation performance between inexperienced and experienced transla-
tors. The differences in how a KIN is utilized were visible at every stage of
the translation process, from reading a SL text for translation to the stage of
revision. Further suggestions were made that the developmental ability to
integrate knowledge is best viewed from the perspective of the translator as
an expert learner, a Systems Intelligent person who through practice and
self-reflection learns to engage and utilize his or her cognitive potential to
perform so much socially required translation services. This perspective al-
lows encompassing different routes to translation competence and expertise
and shows the development of the professional self as a process of develop-
ing personal professional self-confidence. Suggestions were made that to
make our knowledge of translation expertise development more compre-
hensive the role of affective factors and personality issues in the process of
TC acquisition need more attention in future research.
Chapters 6 and 7 attempted to empirically verify the hypothesized
ability to integrate knowledge (build a KIN) in translators occupying dif-
ferent stages on the developmental continuum. Chapter 6 included an
analysis of the questionnaire study conducted among potential and prac-
ticing translators. The data gathered from 200 respondents allowed for a
comprehensive study of ideas about translation as a profession and about
translation as an activity. The still high interest in pursuing a career in
translation among students of English as a foreign language shows that
the profession has a lot to offer to those who want to make a conscious in-
 Conclusions and Implications 403

tentional effort to acquire the necessary translation competence (TC).

Most gains were reported in the intellectual potential of translation as an
activity requiring the comprehensive use of knowledge both conceptual
and linguistic, confirming the role of knowledge integration in the form of
hypothesized Knowledge Integration Networks. A further analysis of the
interdependencies between bilingual knowledge, metalinguistic aware-
ness and translation experience provided insightful information confirm-
ing the hypothesized ability of integrating knowledge into a KIN as an
important part of developing the human ability to translate to the level of
professional competence. Mutual benefits between bilingual knowledge,
metalinguistic awareness and translation practice were pointed out as po-
tential assets to be used for translator training and L2 teaching method-
ologies. The suggestion discussed in chapter 5 that translators are best
perceived as expert learners devoted to a life-long learning process to
reach the level of translation expertise was confirmed by data elicited
from practicing translators. A detailed analysis of the translation process
carried out through responses to the questionnaire confirmed the integra-
tive nature of the translation process supporting the role of the hypothe-
sized ability to integrate knowledge into temporary Knowledge Integra-
tion Networks as an important part of TC acquisition. Although the num-
ber of subjects was not impressive (40 practicing translators) and the urge
to make generalizations has to be restrained the data gathered can serve as
a real-life illustration to counterbalance assumptions and misconceptions
about translation as an activity performed by professional translators.
Since the translation process always takes place in the translator’s mind
the interest in the translator as an individual provided an empirically valid
personal profile of those who have chosen and continue to enjoy transla-
tion as their professional career. The results confirmed that the translator
as an individual is devoted to continuous development through conscious
language maintenance (involving both working languages), intellectual
curiosity and self-reflection. This conclusion could be utilized in transla-
tion pedagogy. In terms of further research the personality determinants
and the role of affective factors in translation expertise development need
further investigation.
The tentative confirmation of the assumption that the ability to inte-
grate knowledge for the purpose of translation into an adequate KIN de-
velops over a period of time following practice and self-reflection was
404 Conclusions and Implications

further investigated in chapter 7. Translation process data were collected

through Translog, a computer key logging programme which allowed for
a thorough insight into TL text production including duration, erased sec-
tions of text, pauses, navigation within the text and dictionary use. Study
1 investigated the translation process of three groups of translators treated
as representatives of three different stages on the developmental contin-
uum of translation as a human skill, inexperienced translators (2BA EFL
students), translation trainees and experienced practicing translators (with
different length of practice). The results analyzed with the analysis of
variance ANOVA showed no statistically significant differences in TL text
production features available through Translog among the three groups of
subjects. This seemingly disappointing result summoned two interpreta-
tions. The samples of subjects were too small (N8 in each group) and
there was a wide range of individual differences within each group which
was obscured by the statistical tools. A closer look at the raw data pro-
vided more interesting results and confirmed that the time used for the TL
text production was differently distributed throughout the translation
process by the three groups of subjects. The subjects in the inexperienced
group spent the least amount of time to read the SL text which allowed to
assume that they did not create an adequate KIN for the task at hand. The
trainees and professional translators spent more time to read the SL text
and it can be tentatively concluded that they created a KIN which resulted
in a more fluent translation process and a more efficient meta-cognitive
supervision over the translation process. Similar differences reported in
terms of dictionary use, navigation within the text, on-line and post-draft
revision point to the developmental nature of the ability to build a Knowl-
edge Integration Network for a task at hand. The group of professional
translators was analyzed in more detail and significant correlations be-
tween length of practice and text processing features were established and
taken as supporting evidence for the proposed ability to build a KIN.
Study 2 aimed at the further testing of the hypotheses formulated on the
grounds of the empirical evidence gathered in Study 1 that a KIN created in
the translating mind will facilitate the translation process. 2BA subjects
were divided into two experimental groups who were translating the same
text. The KIN group apart from being asked to translate a SL text into Eng-
lish was asked to answer 10 questions aimed at creating in their mind an
induced Knowledge Integration Network. The statistical analysis conducted
 Conclusions and Implications 405

with ANOVA did not show any significant differences between the text
production features between both experimental groups apart from the over-
all keyboard activity and text production per minute. Again like in Study 1
there were considerable individual differences between the subjects within
both groups. The conclusion following Study 2 while seemingly disap-
pointing, points to the fact that developing the ability to integrate knowl-
edge for a translation task at hand is a slow process of development and
maturation. The limited capacity of the human mind did not allow for a
speeded up translation process following an artificially created Knowledge
Integration Network in the KIN group. Another possible explanation is that
the KIN created in the subjects’ minds was limited to integrating only the
bilingual knowledge (or only the lexical level) without the support of the
conceptual integration which could prime access to TL linguistic means. It
is also possible that the statistical tools used in the study are not efficient
and the very much individual translation process defies any generalizations.
Other interpretations invite future research.
The comprehensive study of translation as a human skill presented in
this volume aimed at showing its developmental continuum from predis-
position to expertise. This perspective on translation as an ability is able
to encompass all forms and functions translation serves and has served
throughout history. As a result it provides an all inclusive framework to
study translation performed by all possible human translators whether
trained or untrained and it does not deny the ability to anybody who helps
others divided by languages to communicate. In a sense we are all transla-
tors, but since we differ in the way we integrate knowledge for the pur-
pose of translation we produce different translations. This variability fac-
tor applies to all translators who, even if they share a seemingly similar
translation competence, will never produce completely identical transla-
tions. Using the framework in future research could allow for a more fine
grained distinctions not only between natural and trained translators but
also between competent, practicing, professional and expert translators.
If the ideas contained in this book can encourage others to further re-
search the developmental continuum of translation as a human skill, or if
some suggestions can be of any use to translator training and its closest
relative Second Language Acquisition research I will be more than happy.
References

Abutalebi, Jubin – Jean-Marie Annoni – Ivan Zimine – Alan J. Pegna – Mohamed L.

Seghiern – Hannelore Lee-Jahnke – François Lazeyras – Stefano F. Cappa –
Asaid Khateb. 2007. “Language control and lexical competition in bilinguals:
An event-related fMRI study”. Cerebral Cortex. (http://cercor.oxfordjournals.
org/cgi/content/full/bhm182v1) (date of access 20 May 2010).
Adab, Beverly. 2000. “Evaluating translation competence”. In: Schäffner, Christine –
Beverly Adab (eds.), 215-228.
Afendras, Evangelos A. (ed.). 1980. Patterns of bilingualism. Singapore: RELC.
Aitchison, Jean. 1996. Words in the mind. 2nd edition. Oxford: Blackwell.
Alatis, James E. (ed.). 1992. Language, communication and social meaning. Washington
DC: Georgetown University Press.
Albert, Martin L. – Loraine K. Obler. 1978. The bilingual brain: Neuropsychological
and neurolinguistic aspects of bilingualism. New York: Academic Press.
Altariba, Jeanette – Roberto Heredia (eds.). 2008. An introduction to bilingualism:
principles and processes. Mahwah NY: Lawrence Erlbaum Associates.
Altenberg, Evelyn P. – Helen S. Cairns. 1983. “The effects of phonotactic constraints on
lexical processing in bilingual and monolingual subjects”. Journal of Verbal
Learning and Verbal Behaviour 22: 174-188.
Álvarez, Roman – Carmen-África Vidal (eds.). 1996. Translation, power, subversion.
Clevedon: Multilingual Matters.
Alves, Fábio (ed.). 2003. Triangulating translation. Amsterdam: John Benjamins.
Alves, Fabio. 2005. “Ritmo cognitivo, meta reflexão e experiência: parãmetros de
análise processual no desempenho de tradutores novatos e experientes”. In:
Pagano Adriana – Celia Magalhães – Fabio Alves (eds.), 109-169.
Alves, Fabio – José L. Gonçalves – Rui Rothe-Neves. 2001. “In search of a definition of
translation competence: the structure and development of an ongoing research
project”. Quaderns. Revista de traduccio 6: 46-49. (http://ddd.uab.es/pub/
quaderns/11385790n6p46.pdf) (date of access 10 January 2010).
Alves, Fabio – José L. Gonçalves. 2004. “Modelling translator’s competence. Relevance
and expertise under scrutiny”. In: Gambier, Yves – Miriam Shlesinger –
Radegundis Stolze (eds.), 41-55.
Anderman, Gunilla. 1998. “Finding the right word: translation and language teaching.”
In: Malmkjaer, Kirsten (ed.), 39-48.
Anderman, Gunilla – Margaret Rogers (eds.). 1999. Word, text, translation. Liber
amicorum for Peter Newmark. Clevedon: Multilingual Matters.
408 References

Andersen, Roger W. 1983. “Transfer to somewhere”. In: Gass Susan – Larry Selinker
(eds.), 177-201.
Anderson, R. Bruce. 1978. “Interpreter roles and interpretation situations: Cross-cutting
typologies”. In: Gerver, David – Wallace H. Sinaiko, (eds.), 217-229.
Anderson, John R. 1980. Cognitive psychology and its implications. San Francisco:
Freeman.
Anderson, John. R. 1985/2005. Cognitive psychology and its implications. New York:
Freeman.
Anderson, John R. 1996. “ACT: A simple theory of complex cognition”. American
Psychologist 51,4: 355-365.
Angelleli, Claudia V. – Holly E. Jacobson (eds.). 2009. Testing and assessment in
translation and interpreting studies. Amsterdam: John Benjamins.
Appel, René – Pieter Muysken. 1996. Bilingüismo y contacto de lenguas. Barcelona:
Ariel Lingüística.
Arabski, Janusz. 1979. Errors as indicators of the development of interlanguage.
Katowice: Wydawnictwo Uniwersytetu Śląskiego.
Arabski, Janusz. 2006. “Language transfer in language learning and language contact”.
In: Arabski, Janusz (ed.), 12-21.
Arabski, Janusz (ed.). 2006. Cross-linguistic influences in the second language lexicon.
Clevedon: Multilingual Matters.
Arabski, Janusz – Adam Wojtaszek (eds.). 2009. Multidisciplinary perspectives on
second language acquisition and foreign language learning. Katowice:
Wydawnictwo Uniwersytetu Śląskiego.
Arabski, Janusz – Adam Wojtaszek (eds.). 2010. Neurolinguistic and psycholinguistic
perspectives on SLA. Bristol: Multilingual Matters.
Arntz, Reiner (ed.). 1988. Textlinguistik und Fachsprache – Akten des Internationalen
übersetzungswissenschaftlichen AILA-Symposions. Hildesheim: Olms.
Arntz, Reiner – Gisela Thome (eds.). 1990. Überssetzungswissenschaft. Ergebnisse und
Perspektiven. Festschrift für Wolfram Wilss zum 65. Geburstag. Tübingen: Narr.
Asadi, Paula – Candance Séguinot. 2005. “Shortcuts, strategies and general patterns in a
process study of nine professionals”, META 50,2: 522-547.
Atkins, Sue (ed.). 1998. Using dictionaries. Tübingen: Max Niemeyer Verlag.
Atkins, Sue – Krista Varantola. 1998. “Monitoring dictionary use”. In: Atkins, Sue (ed.),
83-122.
Auerbach, Elsa. 1993. “Reexamining English only in the ESL classroom. TESOL
Quarterly 27,1: 9-32.
Baddeley, Alan D. 1986. Working Memory. Oxford: Oxford University Press.
Baddeley, Alan D. 1996. “Exploring the central executive”. Quarterly Journal of
Experimental Psychology 49: 5-28.
Baer, Brian J. – Geoffrey S. Koby (eds.). 2003. Beyond the Ivory Tower. Amsterdam:
John Benjamins.
Baker, Mona. 1992. In other words. Manchester: St. Jerome.
Baker, Mona (ed.). 2001. Routledge encyclopedia of Translation Studies. London:
Routledge.
 References 409

Baker, Mona – Gabriela Saldanha (eds.). 2009. Routledge encyclopedia of Translation

Studies 2nd edition. New York: Routledge.
Bartlett, Frederic C. 1932. A study in experimental and social psychology. Cambridge:
Cambridge University Press.
Bassnett, Susan. 1996. “The meek or the mighty: rephrasing the role of the translator”.
In: Álvarez, Roman – Carmen-África Vidal (eds.), 10-24.
Bassnett, Susan – André Lefevere (eds.). 1990. Translation, history and culture.
London: Pinter.
Beatens Beardsmore, Hugo. 1982. Bilingualism: basic principles. Clevedon:
Multilingual Matters.
Beeby, Allison – Doris Ensinger – Marisa Presas (eds.). 2000. Investigating translation.
Amsterdam: John Benjamins.
Bell, Roger T. 1991. Translation and translating: theory and practice. London:
Longman.
Ben Zeev, Sandra. 1977. “The influence of bilingualism on cognitive strategy and
cognitive development”. Child Development 48,3: 1009-1018.
Bergström, Matti R. 1988. “Communication and translation from the point of view of
brain functions”. In: Holz-Mänttäri (ed.), 23-38.
Berman, Antoine. 1989. “La traduction et ses discourse”. META 34,4: 672-679.
Bereiter, Carl – Marlene Scardamalia. 1993. Surpassing ourselves: An inquiry into the
nature and implications of expertise. Chicago: Open Court Publishing
Company.
Bernardini, Silvia. 1999. Using think-aloud protocols to investigate the translation
process: methodological aspects. Bologna: University of Bologna.
Bernardini, Silvia. 2002. “Think Aloud Protocols in translation research : achievements,
limits, future prospects.” Target 13,2: 241-263.
Besner, Derek. – Marilyn C. Smith – Colin M. MacLeod. 1990. “Visual word
recognition: A dissociation of lexical and semantic processing”. Journal of
Experimental Psychology: Learning, Memory, and Cognition 16: 862-869.
Beylard-Ozeroff, Ann – Jana Králová – Barbara Moser-Mercer (eds.). 1998.
Translators’ strategies and creativity. Amsterdam: John Benjamins.
Bhatia, Tej K. – William C. Ritchie (eds.). 2004. A handbook of bilingualism. Malden,
MA: Blackwell.
Bialystok, Ellen (ed.). 1991. Language processing in bilingual children. Cambridge:
Cambridge University Press.
Bialystok, Ellen (ed.). 2001/2005. Bilingualism in development: language, literacy, and
cognition. Cambridge: Cambridge University Press.
Bialystok, Ellen. 2005. “Consequences of bilingualism for cognitive development”. In:
Kroll, Judith F. – Annette M.B. De Groot (eds.), 417-432.
Bialystok, Ellen – Fergus I.M. Craik – Raymond Klein – Mythili Viswanathan. 2004.
“Bilingualism, aging and cognitive control: evidence from the Simon task”.
Psychology and Aging 19: 290-303.
Biguenet, John – Rainer Schulte (eds.). 1984. The craft of translation. Chicago: The
University of Chicago Press.
410 References

Block, David. 2003. The social turn in second language acquisition. Edinburgh:
Edinburgh University Press.
Block, David. 2007. Second language identities. London: Continuum.
Boekaerts, Monique. 1981. “Is there a direct link between the comprehension and the
production process?”. In: Heid, Manfred (ed.), 26-60.
Börsch, Sabine. 1986. “Introspective methods in research on interlingual and
intercultural communication”. In: Juliane House – Shoshana Blum – Kulka
(eds.), 195-209.
Bowker, Lynne – Michael Cronin – Dorothy Kenny – Jennifer Pearson (eds.). 1998.
Unity in diversity? Manchester: St. Jerome Publishing.
Breedveld, Hella. 2002. “Writing and revising processes in professional translation”.
Across Languages and Cultures 3,1: 91-100.
Brislin, Richard W. (ed.). 1976. Translation: Application and research. New York:
Gardner.
Broeck, Raymond, van den (ed.). 1994. Translated!: Papers on literary translation and
Translation Studies/James S. Holmes. Amsterdam: Atlanta.
Bromberek-Dyzman, Katarzyna – Karolina Rataj – Jacek Dylak. 2010. “Mentalizing in
the second language: Is irony online inferencing any different in L1 and
L2?”. In: Witczak-Plisiecka, Iwona (ed.), 197-216.
Brower, Reuben A. (ed.). 1959. On translation. Cambridge, MA: Harvard University
Press.
Bruner, Jerome S. – Allison Jolly – Kathy Sylva (eds.). 1976. Play. Its role in
development and evolution. Harmondsworth: Penguin.
Buchweitz, Augusto – Fábio Alves. 2006. “Cognitive adaptation in translation: an
interface between language direction, time and recursiveness in target text
production”. Letras de Hoje 41,2: 241-272.
Bullock, Carolyn – Harris, Brian. 1997. “Schoolchildren as community interpreters”. In:
Carr Silvana E. – Roda P. Roberts – Aideen Dufour – Dini Steyn (eds.), 227-248.
Buttjes, Dieter – Michael Byram (eds.). 1990. Mediating language and cultures:
Towards an intercultural theory of foreign language education. Clevedon:
Multilingual Matters.
Byram, Michael – Anwei Feng. 2005. “Teaching and researching intercultural
competence”. In: Hinkel, Eli (ed.), 911-930.
Byram, Michael – Peter Grundy (eds.). 2002. Context and culture in language teaching
and learning. Clevedon: Multilingual Matters.
Caine, Renate N. – Geoffrey Caine. 1994. Making connections: teaching and the human
brain. Menlo Park, CA: Addison-Wesley.
Canale, Michael – Merrill Swain. 1980. “Theoretical bases of communicative approaches
to second language teaching and testing”. Applied Linguistics 1: 1-47.
Caramazza, Alfonso. 1997. “How many levels of processing are there in lexical
access?”. Cognitive Neuropsychology 14: 177-208.
Carr, Silvana E. – Roda P. Roberts – Aideen Dufour – Dini Steyn (eds.). 1997. The
critical link: Interpreters in the community. Amsterdam: John Benjamins.
 References 411

Carroll, John B. 1978. “How shall we study individual differences in cognitive abilities?
– Methodological and theoretical perspectives”. Intelligence 2: 87-115.
Carroll, John B. 1993. Human cognitive abilities. Cambridge: Cambridge University Press.
Carroll, John B. – Stanley M. Sapon. 2002. Modern Language Aptitude Test –
Elementary: Manual, 2002 Edition. Rockville, MD: Second Language
Testing, Inc.
Cameron, Deborah. 2001. Working with spoken discourse. London: Sage Publications.
Cao, Deborah. 1996. “Towards a model of translation proficiency”. Target 8,2: 325-340.
Catford, John. C. 1965. A linguistic theory of translation. Oxford: Oxford University
Press.
Cattell, Ray. 2006. An introduction to mind, consciousness and language. London:
Continuum International Publishing Group.
Chau, Simon. 1999. “From anonymous parasites to transformation agents – a ‘third
world’ vision of translation for the new millennium”. In: Anderman, Gunilla –
Margaret Rogers (eds.), 233-235.
Checkland, Peter. 1999. Systems thinking, systems practice. Chichester: John Wiley.
Chesterman, Andrew. 2002. “On the interdisciplinarity of Translation Studies”. Logos
and Language 3,1: 1-9.
Chesterman, Andrew – Natividad Gallardo San Salvador – Yves Gambier (eds.). 2000.
Translation in context. Amsterdam: John Benjamins.
Chesterman, Andrew – Eva Wagner. 2002. Can theory help translators? Manchester: St.
Jerome.
Chi, Michelene T. – Robert Glaser – Marshall Farr (eds.). 1988. The nature of expertise.
Hillsdale, NJ: Erlbaum.
Chen, Hsuan-Chih – Yuen-Sum Leung. 1989. “Patterns of lexical processing in a non-
native language”. Journal of Experimental Psychology: Learning, Memory,
and Cognition 15: 316-325.
Chen, Hsuan-Chih – Man-Lai Ng. 1989. “Semantic facilitation and translation priming
effects in Chinese-English bilinguals”. Memory and Cognition 17: 454-462.
Chmiel, Agnieszka. 2004. Neurocognitive plausibility of conference interpreting.
Application of cognitive neuroscience to interpreting research. Unpublished
doctoral dissertation. Poznań. Adam Mickiewicz University.
Chmiel, Agnieszka. 2006. “Skill-based approach to conference interpreting”. Translation
Ireland. New Vistas in Translator and Interpreter Training 17: 47-64.
Chmiel, Agnieszka. 2010. “Interpreting studies and psycholinguistics: A possible synergy
effect”. In: Gile, Daniel – Gyde Hansen – Nike K. Pokorn (eds.), 223-236.
Chomsky, Noam. 1965. Aspects of the theory of syntax. Masssachusetts: The MIT Press.
Chomsky, Noam. 1986. Knowledge of language. New York: Praeger Publisher.
Chomsky, Noam. 2006. Language and mind (3rd edition). Cambridge: Cambridge
University Press.
Chriss, Roger. 2006. Translation as a profession. Raleigh: Lulu.com.
Churchman, Charles W. 1968. The systems approach: New York: Delta.
Cieślicka, Anna. 2005. “Semantic priming effects in bilingual word fragment
completion”. Poznań Studies in Contemporary Linguistics 40: 111-133.
412 References

Cieślicka, Anna. 2004. On processing figurative language: towards a model of idiom

comprehension in foreign language learners. Poznań: Motivex.
Cieślicka, Anna – Anna Ekert. 2009. “Investigating asymmetry in lexical processing
with proficient Polish-English bilinguals”. In: Arabski, Janusz – Adam
Wojtaszek (eds.), 241-253.
Clark, Eve V. 1978. “Awareness of language: Some evidence from what children say and
do”. In: Sinclair, Anne – Robert J. Jarvella – Willem J. M. Levelt (eds.), 17-43.
Clyne, Michael G. 1980. “Triggering and language processing”. Canadian Journal of
Psychology 34,4: 400-406.
Cole, Peter – Jerry Morgan (eds.). 1975. Syntax and Semantics, 3: Speech Acts. New
York: Academic Press.
Colina, Sonia. 2002. “Second language acquisition, language teaching and Translation
Studies”. The Translator 8,1: 1-24.
Colina, Sonia. 2003a. “Towards an empirically-based translation pedagogy”. In: Baer,
Brian J. – Geoffrey S. Koby (eds.), 29-59.
Colina, Sonia. 2003b. Translation teaching from research to the classroom. New York:
Mc-Grow Hill.
Cook, Guy. 2001. “Language teaching, use of translation”. In: Baker, Mona (ed.), 117-120.
Cook, Guy. 2010. Translation in language teaching: An argument for reassessment.
Oxford: Oxford University Press.
Cook, Vivian J. 1991. “The poverty-of-the-stimulus argument and multicompetence”.
Second Language Research 7,2: 103-117.
Cook, Vivian J. 1992. “Evidence for multicompetence”. Language Learning 42,4: 557-591.
Cook, Vivian J. 2002. “Background to the L2 user.” In: Cook, Vivian J. (ed.), 1-28.
Cook, Vivian J. (ed.). 2002. Portraits of the L2 user. Amsterdam: John Benjamins.
Cook, Vivian J. 2003. “The changing L1 in the L2 user’s mind”. In: Cook, Vivian J.
(ed.), 1-18.
Cook, Vivian J. (ed.). 2003. Effects of the second language on the first. Clevedon:
Multilingual Matters.
Cook, Vivian J. 2007. “The nature of the L2 user”. EUROSLA Yearbook 7: 205-220.
Costa, Albert – Mikel Santesteban. 2004. “Lexical access in bilingual speech production:
evidence from language switching in highly proficient bilinguals and L2
learners”. Journal of Memory and Language 50: 491-511.
Cronin, Michael. 1995. “Keeping one’s distance: Translation and the play of
possibility”. TTR: Traduction, Terminologie, Rédaction, 8,2: 227-243.
Cronin, Michael. 2005. “Deschooling translation”. In Tennent, Martha (ed.), 249-265.
Crystal, David. 1986. Listen to your child. Harmondsworth: Penguin.
Cummins, Jim. 1991. “Language development and academic learning”. In: Malavé
Lillian M. – Georges Duquette (eds.), 161-174.
Dam, Helle V. – Jan Engberg – Heidrun Gerzymisch-Arbogast (eds.). 2005. Knowledge
systems and translation. Berlin: Muton de Gruyter.
Damasio, Antonio R. 1994. Descartes’ error. New York: Avon Books.
Damasio, Antonio R. 1999. The feeling of what happens. New York: Harcourt Brace.
 References 413

Dancette, Jeanne. 1992. “Measurements of comprehension of the source text: theory and
practice”. In: Lewandowska-Tomaszczyk, Barbara – Marcel Thelen (eds.),
379-386.
Danesi, Marcel. 1992. “Metaphorical competence in second language acquisition and
second language teaching: the neglected dimension”. In: Alatis, James E.
(ed.), 489-500.
Danesi, Marcel. 2003. Second language teaching: a view from the right side of the brain.
Dordrecht: Kluwer Academic Publishers.
Danks, Joseph H. – Gregory M. Shreve – Stephen B. Fountain – Michael K. McBeath
(eds.). 1997. Cognitive prosesses in translation and interpreting. Thousand
Oaks: Sage Publications.
Davidson, Donald. 1980. Essays on actions and events. Oxford: Oxford University
Press.
Davidson, Richard J. – Gary Schwartz – David Shapiro (eds.). 1986. Consciousness and
self-regulation: advances in research and theory (vol. 4). New York: Plenum
Press.
De Groot, Annette M.B. 1992. “Bilingual lexical representation: A closer look at
conceptual representations”. In: Frost, Ram – Leonard Katz (eds.), 389-412.
De Groot, Annette M.B. 1997. “The cognitive study of translation and interpretation:
Three approaches”. In: Danks, Joseph H. – Gregory M. Shreve – Stephen B.
Fountain – Michael K. McBeath (eds.), 25-56.
De Groot, Annette M.B. 2000. “A complex-skill approach to translation and
interpreting”. In: Tirkkonen- Condit, Sonja – Ritta Jääskeläinen (eds.), 53-68.
De Groot, Annette M.B. 2002. “Lexical representation and lexical processing in the L2
user”. In: Cook ,Vivian (ed.), 32-61.
De Groot, Annette M.B. – Ingrid K. Christoffels. 2006. “Language control in bilinguals:
Monolingual tasks and simultaneous interpreting”. Bilingualism: Language
and Cognition 9,2: 189-201.
De Groot, Annette M.B. – Lucia Dannenburg – Janet G. van Hell. 1994. “Forward and
backward word translation by bilinguals”. Journal of Memory and Language
33: 600-629.
De Groot, Annette M.B. – Rineke Keijzer. 2000. “What is hard to learn is easy to forget:
The role of word concreteness, cognate status, and word frequency in foreign
language vocabulary learning and forgetting. Language Learning 50: 1-56.
De Groot, Annette M.B. – Judith F. Kroll (eds.). 1997. Tutorials in bilingualism:
Psycholinguistic perspectives. Mahwah, NJ: Lawrence Erlbaum Associates.
De Groot, Annette M.B. – Gerard L.J. Nas. 1991. “Lexical representation of cognates
and noncognates in compound bilinguals”. Journal of Memory and Language
30: 90-123.
Delabastita, Dirk. 1993. There’s a double tongue. An investigation into the translation of
Shakespeare’s wordplay with special reference to Hamlet. Amsterdam: Rodopi.
Delisle, Jean. 1988. Translation – an interpretative approach. Ottawa: University of
Ottawa Press.
414 References

Delisle, Jean – Judith Woodsworth (eds.). 1995. Translators through history.

Amsterdam: John Benjamins.
De Ment, Terri L. – Raymond Buriel – Christina M. Villanueva. 2005. “Children as
language brokers”. In: Rumjahn, Hossain – Salili Farideh (eds.), 255-272.
De Oliveira Harden, Alessandra. 2009. “The rules of the game: Translation as a
privileged learning resource”. In: Witte, Arnd – Theo Harden – Alessandra R.
de Oliveira Harden (eds.), 361-374.
Derrida, Jacques. 1992. “From des tours de Babel” [translated by Joseph F. Graham]. In:
Schulte, Rainer – John Biguenat (eds.), 218-227.
Dijkstra, Ton A. – Walter J.B. Van Heuven. 2002. “The arcitecture of the bilingual word
recognition system: from identification to decision”. Bilingualism, Language
and Cognition 5: 175-197.
Dimitrova, Birgitta Englund. 2005. Expertise and explicitation in the translation
process. Amsterdam: John Benjamins.
Dimitrova, Brigitta Englund – Kerstin Jonasson. 1999. “Translation ability and
translatorial competence: expert and novice use of dictionaries”. Paper
presented at the Symposiun on translation and Interpreting. Stockholm
(http://www2.tolk.su.se/Birgitta_Englund_Dimitrova/documents/englund_dim
itrova__jonasson.1999.pdf) (date of access 20 August 2010).
Dodds, John. 1999. “Friends, false friends and foes or back to basics in L1 to L2
translation”. In: Anderman, Gunilla – Margater Rogers (eds.), 56-65.
Dollerup, Cay – Anne Loddegaard (eds.). 1992. Teaching translation and interpreting.
Amsterdam: John Benjamins.
Dollerup, Cay – Annette Lindegaard (eds.). 1994. Teaching translation and interpreting
2. Amsterdam: John Benjamins.
Donald, Merlin. 1991. Origins of the modern mind: Three stages in the evolution of
culture and cognition. Harvard: Harvard University Press.
Donald, Merlin. 2001. A mind so rare: The evolution of human consciousness. New
York: Norton.
Donald, Merlin. 2006. “Cognitive Cultural Networks”. Paper presented at the 2nd Culture
Language and Mind Conference. Paris.
Donald, Merlin. 2007a. “The slow process: A hypothetical cognitive adaptation for
distributed cognitive networks”. Journal of Physiology 101: 214-222.
Donald, Merlin. 2007b. “Consciousness and governance: from embodiment to
enculturation – an interview”. In: Lars Andreassen – Line Brandt – Jes Vang
(eds.), 68-83.
Donald, Merlin. 2008. “How culture and brain mechanisms interact in decision making”.
In: Engel, Christoph – Wolf Singer (eds.), 191-225.
Dornic, Stan. 1980. “Language dominance, spare capacity and perceived effort in
bilinguals”. Ergonimics 23,4: 369-377.
Dudai, Yadin. 1989. The neurobiology of memory: Concepts, findings, trends. New
York: Oxford University Press.
Duff, Alan. 1981. The third language. Oxford: Pergamon Press.
 References 415

Durban, Chris. 2003. Translation: Getting it right – A guide to buying translations.

Milton Keynes: Institute of Translation and Interpreting.
Dynel, Marta (ed.). 2009. Advances in discourse approaches. Cambridge: Cambridge
Scholars Publishing.
Dziubalska-Kołaczyk, Katarzyna (ed.). 2006. IFAtuation: A life in IFA. A Festschrift for
Professor Jacek Fisiak on the occasion of his 70th birthday. Poznań:
Wydawnictwo Naukowe UAM.
Eichenbaum, Howard. 2002. The cognitive neuroscience of memory: An introduction.
New York: Oxford University Press.
Engel, Christoph – Wolf Singer (eds.). 2008. Better than conscious? Decision-making,
the human mind, and implications for institutions. Cambridge, MA: The MIT
Press.
English, Horace B. – Ava C. English. 1958. A comprehensive dictionary of psychological
and psychoanalytical terms: A guide to usage. New York: Mc Kay.
Ericsson, Karl A. – Jacqui Smith (eds.). 1991. Towards a general theory of expertise.
Cambridge: Cambridge University Press.
Ericsson, Karl A. – Herbert. A. Simon. 1980. “Verbal reports as data”. Psychological
Review 87: 215-251.
Ertmer, Peggy – Timothy Newby. 1996. “The expert learner, strategic, self-regulated,
and reflective”. Instructional Science 24,1: 1-24.
Ervin, Susan M. – Charles E. Osgood. 1954. “Second language learning and bilingualism”.
Journal of Abnormal Social Psychology, Supplement 52: 139-146.
Esch, Kees van – John Oliver St. (eds.). 2004. New insights in foreign language learning
and teaching. Frankfurt: Peter Lang.
Ewert, Anna. 2009. L2 User’s L1. Poznań: Pracownia Wydawnicza Instytutu Filologii
Angielskiej.
Fabbro, Franco. 1999. The neurolinguistics of bilingualism: An introduction. East
Sussex: Psychology Press Ltd.
Fabiszak, Małgorzata (ed.). 2007. Language and meaning. Frankfurt: Peter Lang.
Fauconnier, Giles. 1998. “Mental spaces, language modalities, and conceptual
integration”. In: Tomasello, Michael (ed.), 251-279.
Fawcett, Peter. 1997. Translation and language: Linguistic theories explained.
Manchester: St Jerome.
Feltovich, Paul J. – Kenneth M. Ford – Robert R. Hoffman (eds.). 1997. Expertise in
context. Cambridge, MA: MAT Press.
Férailleur-Dumoulin, Carline. 2009. A Career in Language Translation. Bloomington:
Author House.
Fillmore, Charles J. 1977. “Scenes -and –frames semantics”. In: Zampolli, Antonio (ed.),
55-81.
Firth, Alan – Johannes Wagner. 1997. “On discourse, communication, and (some)
fundamental concepts in SLA research”. Modern Language Journal 81: 285-300.
Firth, Alan – Johannes Wagner. 2007. “Second/foreign language learning as a social
accomplishment: Elaborations on a reconceptualized SLA”. Modern
Language Journal 91: 800-819.
416 References

Fisher, Kurt – Sandra Pipp. 1984. “Processes of cognitive development: optimal level
and skill acquisition”. In: Steinberg, Robert J. (ed.), 45-80.
Fisiak, Jacek (ed.). 1991. Further insights into contrastive analysis. Amsterdam: John
Benjamins.
Fisiak, Jacek (ed.). 2006. English language, literature and culture. Poznań: Uni-Druk.
Flood, Robert L. 1999. Rethinking the fifth discipline. Learning within the unknowable.
London: Routledge.
Fodor, Jerry. 1975. The language of thought. Harvard: University Press.
Fodor, Jerry. 2008. The language of thought revisited. Oxford: OUP.
Forbach, Garry B. – Robert F. Stanners – Larry Hochhaus. 1974. “Repetition and
practice effects in a lexical decision task”. Memory and Cognition 2: 337-339.
Fortunato, Israel (ed.). 1998. Quelle formation pour le traducteur de l’an 2000? Paris:
Didier.
Frank, Armin P. – Norbert Greiner – Theo Hermans – Haral Kittel – Werner Koller –
José Lambert – Paul Fritz (eds.). 2004. An International Encyclopaedia of
Translation Studies, Art. 48. Vol. 1. Berlin: Mouton de Gryuter.
Fraser, Janet. 1999. “The translator and the word: The pros and cons of dictionaries in
translation”. In: Anderman, Gunilla – Margater Rogers (eds.), 25-34.
Fraser, Janet. 2000. “The broader view: How freelance translators define translation
competence”. In: Schäffner, Christina – Beverly Adab (eds.), 51-62.
Frith, Uta (ed.). 1980. Cognitive processes in spelling. London: Academic Press.
Frost, Ram – Leonard Katz (eds.). 1992. Orthography, phonology, morphology, and
meaning. Amsterdam: North Holland .
Gambier, Yves – Jorma Tommola (eds). 1993. Translation and knowledge. Turku:
University of Turku Press.
Gambier, Yves – Miriam Shlesinger – Radegundis Stolze (eds.). 2004. Doubts and
directions in Translation Studies. Amsterdam: John Benjamins.
Gardner, Robert C. 1979. “Social psychological aspects of second language acquisition.
In: Giles, Howard – Robert N. St Clair (eds.), 193-220.
Gardner, Howard – Michael Connell. 2000. “Response to Nicholas Allix”. Australian
Journal of Education 44: 288–293.
Gass, Susan – Larry Selinker (eds). 1983. Language transfer in language learning.
Rowley, MA: Newbury House.
Gatenby, Edward V. 1967 “Reasons for failure to learn a foreign language”. In: Lee,
William R. (ed.), 1-10.
Genesee, Fred – Elena Nicoladis. 2006. “Bilingual acquisition”. In: Hoff, Erika –
Marilyn Shatz (eds.), 324-342.
Gentzler, Edwin. 1993. Contemporary translation theories. New York: Routledge.
Gerard, Linda D. – Don L. Scarborough. 1989. “Language specific lexical access of
homoghraphs by bilinguals”. Journal of Experimental Psychology: Learning,
Memory, and Cognition 15: 305-315.
Gerloff, Pamela. 1986. “Second language learners’ reports on the interpretive process:
talk-aloud protocols of translation”. In: House, Juliane – Shoshana Blum–
Kulka (eds.), 243-262.
 References 417

Gerver, David – Wallace H. Sinaiko (eds). 1978. Language interpretation and

communication. New York: Plenum Press.
Giddens, Anthony. 1991. Modernity and self-identity. Self and society in the late modern
age. Cambridge: Polity Press.
Gile, Daniel. 1993. “Translation/interpretation and knowledge”. In: Gambier, Yves –
Jorma Tommola (eds.), 67-86.
Gile, Daniel. 1995/2009. Basic concepts and models for interpreter and translator
training. Amsterdam: John Benjamins.
Gile, Daniel. 1997. “Conference interpreting as a cognitive management problem”. In:
Danks, Joseph E. – Gregory M. Shreve – Stephen B. Fountain – Michael K.
McBeath (eds.), 196-214.
Gile, Daniel. 2004. “Integrated problem and decision reporting as a translation training
tool”. Journal of Specialized Translation 2, (http://www.jostrans.org/issue02/
art_gile.php) (date of access 14 March 2010).
Gile, Daniel – Gyde Hansen – Nike K. Pokorn (eds.). 2010. Why Translation Studies
Matters. Amsterdam: John Benjamins.
Giles, Howard – Robert N. St Clair (eds). 1979. Language and social psychology.
Oxford: Basil Blackwell.
Girard, Denis (ed.). 1992. Enseignement des langues dans le monde d'aujourd'hui. Paris:
Hachette.
Goleman, Daniel. 1995. Emotional intelligence. New York: Bantam Books.
Goleman, Daniel. 2011. The brain and emotional intelligence: New insights.
Northampton: More than Sound.
Gollan, Tamar H. – Judith F. Kroll. 2001. “Bilingual lexical access”. In: Rapp Brenda
(ed.), 321-345.
Gollan, Tamar H. – Lori-Ann Acenas. 2004. “What is a TOT? Cognate and translation
effects on tip-of-the-tongue states in Spanish-English and Tagalog-English
bilinguals”. Journal of Experimental Psychology: Learning, Memory, and
Cognition 30: 246-269.
Gollan, Tamar H. – Rosa I. Montoya – Christine Fennema-Notestine – Shaunna K.
Morris. 2005. “Bilingualism affects picture naming but not picture
classification”. Memory & Cognition 33: 1220-1234.
Gómez Hurtado, Isabel. 2006. Traducir, ¿capacidad innata o destreza adquirida?
[Translating. Innate ability or acquired skill?]. Doctoral dissertation. Granada:
University of Granada.
González Davies, Maria. 2004. Multiple voices in the translation classroom.
Amsterdam: John Benjamins.
Gopher, Daniel. 1992. “The skill of attention control: acquisition and execution of
attention strategies”. In: Meyer, David E. – Sylvan Kornblum (eds.), 299-322.
Gorlee, Dinda L. 1994. Semiotics and the problem of translation. Amsterdam: Atlanta.
Green, Alison J.K. – Kenneth J. Gilhooley. 1992. “Empirical advances in expertise
research”. In: Keane, Mark T. – Kenneth J. Gilhooley (eds.), 45-70.
Green, David W. 1986. “Control, activation and resource: a framework and a model for
the control of speech in bilinguals”. Brain and Language 27,2: 210-223.
418 References

Green, David W. 1993. “Towards a model of L2 comprehension and production”. In:

Schreuder, Robert – Bert Weltens (eds.), 249-277.
Green, David W. 1998. “Mental control of the bilingual lexico-semantic system”.
Bilingualism: Language and Cognition 1,2: 67-81.
Green, David. 2000. “Control, activation and resource: a framework and a model for the
control of speech in bilinguals”. In: Wei, Li (ed.), 374-385.
Grice, Paul. 1975. “Logic and conversation”. In: Cole, Peter – Jerry Morgan (eds.), 41-58.
Grosjean, Francis. 1982. Life with two languages: An introduction to bilingualism.
Cambridge, MA: Harvard University Press.
Grosjean, Francois. 1997. “The bilingual individual”. Interpreting 2, 1-2: 163-187.
Grosjean, Francois. 2001. “The bilingual’s language modes.” In: Nicol, Janet (ed.), 1-21.
Grosjean, Francois. 2000. “Processing mixed languages: Issues, findings and models”.
In: Wei, Li (ed.), 408-434.
Grosjean, Francois. 2002. “Interview”. The Bilingual Family Newsletter 19,4: 4-7.
Gouadec, Daniel. 2007. Translation as a profession. Amsterdam: John Benjamins.
Gudykunst, William W.B. – Young Yun Kim. 1992. Communicating with strangers: An
approach to intercultural communication. 2nd edition. New York: McGraw
Hill Inc.
Gutt, Ernst-August. 1991. Translation and relevance: cognition and context. Oxford:
Blackwell.
Hall, Joan K. – An Cheng – Matthew T. Carlson. 2006. “Reconceptualizing
multicompetence as a theory of language knowledge”. Applied Linguistics
27,2: 220-240.
Halliday, Michael A.K. – John Gibbons – Howard Nicholas (eds.). 1990. Learning,
keeping and using language. Amsterdam: John Benjamins.
Hämäläinen, Raimo P. – Esa Saarinen (eds.). 2007. Systems intelligence in leadership
and everyday life. Helsinki: University of Technology.
Hamers, Josiane F. – Michel H. Blanc. 1990. Bilinguality and bilingualism. Cambridge,
MA: Harvard University Press.
Hansen, Gyde (ed.). 2002. Empirical Translation Studies: Process and product.
Copenhagen: Samfundslitteratur.
Hansen, Gyde. 2003. “Controlling the process”. In : Alves, Fábio (ed.), 25-43.
Hansen, Gyde. 2006. “Retrospective methods in translator training and translation
research”. Journal of Specialized Translation 5: 1-40. (http:www.jostrans.org)
(date of access 20 May 2010).
Hansen, Gyde. 2008. “The speck in your brother’s eye – the beam in your own: Quality
management in translation and revision”. In: Hansen, Gyde – Andrew
Chesterman – Heidrun Gerzymisch-Arbogast (eds.), 255-277.
Hansen, Gyde – Andrew Chesterman – Heidrun Gerzymisch-Arbogast (eds.). 2008.
Efforts and models in interpreting and translation research. A tribute to
Daniel Gile. Amsterdam: John Benjamins.
Hansen, Gyde – Kirsten Malmkjaer – Daniel Gile (eds.). 2004. Claims, changes and
challenges in Translation Studies. Amsterdam: John Benjamins.
 References 419

Hansen, Kit. 2009. Literacy and Second Language Oracy. Oxford: Oxford University
Press.
Harden, Theo. 2009. “Accessing conceptual metaphors through translation”. In: Witte,
Arnd – Theo Harden – Alessandra Ramos de Oliveira Harden (eds.), 119-133.
Harley, Birgit. 1986. Age in second language acquisition. Clevedon: Multilingual
Matters.
Haritos, Calliope. 2003. “Listening, remembering, and speaking in two languages: How
did you do that?” Bilingual Research Journal 27,1: 73-99.
Harré, Rom – Roger Lamb (eds). 1983. The encyclopedic dictionary of psychology.
Massachusetts: The MIT Press.
Harris, Brian. 1977. “The importance of natural translation”. Working papers in
bilingualism 12: 96-114.
Harris, Brian. 1980. “How a three-year-old translates”. In: Afendras, Evangelos A. (ed.),
370-393.
Harris, Brian. 1992. “Natural translation: A reply to Hans P. Krings”. Target 4,1: 97-
103.
Harris, Brian. 1997. “Foreword: A landmark in the evolution of interpreting”. In: Carr
Silvana E. – Roda P. Roberts – Aideen Dufour – Dini Steyn (eds.), 1-6.
Harris, Brian – Bianca Sherwood. 1978. “Translating as an innate skill”. In: Gerver,
David – Wallace H. Sinaiko (eds.), 155-170.
Harris, Richard J. (ed.). 1992. Cognitive processing in bilinguals. Amsterdam: North-
Holland.
Hatano, Giyoo – Kayoko Inagaki. 1992. “Desituating cognition through the construction of
conceptual knowledge”. In: Light, Paul – George Butterworth (eds.), 115-133.
Hatim, Basil – Ian Mason. 1997. The Translator as communicator. London: Routledge.
Hayes, John R. 1996. “A new framework for the understanding of cognition and affect in
writing”. In: Levy, Michael C. – Sarah Ransdell (eds.), 1-28.
Healey, Frank. 1978. “Translators made not born?”. The Incorporated Linguist 17,3: 54-
58.
Hebb, Donald O. 1949. The organization of behavior. New York: Wiley & Sons.
Heid, Manfred (ed.). 1981. New Yorker Werkstattgespräch 1980: Beziehungen zwischen
Sprachrezeption un Sprachproduktion im Fremdsprachenuterricht. Munich:
Goethe House.
Hejwowski, Krzysztof. 1992. Psycholingwistyczny model procesu tłumaczenia.
Unpublished doctoral dissertation. Warszawa: Uniwersytet Warszawski.
Hejwowski, Krzysztof. 2004. Kognitywno-komunikacyjna teoria przekładu. Warszawa:
Wydawnictwo Naukowe PWN.
Henderson, John S. 1987. Personality and the linguist. Bradford: University of Bradford
Press.
Hentschel, Elke. 2009. “Translation as an inevitable part of foreign language
acquisition”. In: Witte, Arnd – Theo Harden – Alessandra Ramos de Oliveira
Harden (eds.), 15-30.
Herdina, Philip – Urlike Jessner. 2002. A Dynamic model of multilingualism.
Perspectives of change in psycholinguistics. Clevedon: Multilingual Matters.
420 References

Heredia, Roberto. 1997. “Bilingual memory and hierarchical models: A case for
language dominance”. Current Directions in Psychological Science 6: 34-39.
Heredia, Roberto. 2008. “Mental models of bilingual memory”. In: Altariba, Jeanette –
Roberto Heredia (eds.), 39-67.
Heredia, Roberto – Jeffrey Brown. 2004. “Bilingual memory”. In: Bhatia, Tej K.-
William C. Ritchie (eds.), 225-249.
Hermans, Theo. 1991. “Translational norms and correct translations”. In: Leuven-Zwart,
Kitty M. – Ton Naaijkens (eds.), 155-169.
Hickey, Leo. 1998. The pragmatics of translation. Clevedon: Multilingual Matters.
Hickey, Raymond – Stanislaw Puppel (eds.). 1997. Language history and linguistic
modeling. A Festschrift for Jacek Fisiak on his 60th birthday Vol. II. Berlin:
Mouton de Gruyter.
Hinkel, Eli (ed.). 2005. Handbook of research in second language teaching and
learning. Mahwah, NJ: Lawrence Erlbaum Associates.
Hoff, Erika – Marilyn Shatz (eds.). 2006. Handbook of language development. Oxford:
Blackwell.
Hoffman, Robert (ed.). 1992. The psychology of expertise. Cognitive research and
empirical AI. Mahwah, NJ: Lawrence Erlbaum Associates.
Hoffman, Robert R. 1997. “The cognitive psychology of expertise and the domain of
interpreting”. Interpreting 2,1-2: 189-230.
Holmes, James. 1988/1994. Translated! Papers on literary translation and Translation
Studies. Amsterdam: Rodopi.
Holquist, Michael. 2003. “What’s the ontological status of bilingualism”. In: Sommer,
Doris (ed.), 21-34.
Holz Mänttäri, Justa. 1984. Translatorisches Handeln. Theorie und Methode. Helsinki:
Suomalainen Tiedeakatemia.
Holz-Mänttäri (ed.). 1988. Translationstheorie – Grundlagen und Function. Tampere:
University of Tampere.
House, Juliane. 1977. A model for translation quality assessment. Tübingen: Narr.
House, Juliane. 1986. “Acquiring translational competence in interaction”. In: House,
Juliane – Shoshana Blum-Kulka (eds.), 179-191.
House, Juliane. 1997. Translation quality assessment: a model revisited. Tübingen:
Gunter Narr Verlag.
House, Juliane. 2001. “Translation quality assessment: linguistic description versus
social evaluation”. META 46,2: 243-257.
House, Juliane. 2009. “Quality of translation”. In: Baker, Mona – Gabriela Saldanha
(eds.), 252-256.
House, Juliane – Shoshana Blum–Kulka (eds.). 1986. Interlingual and intercultural
communication. Tübingen: Narr.
Howatt, Anthony P. 1984. A history of English language teaching. Oxford: Oxford
University Press.
Hönig, Hans G. 1990. “Sagen, was macht nicht weiss – wissen was macht nicht sagt.
Überlegungen zur übersetzerischen Intuition”. In: Arntz, Reiner – Gisela
Thome (eds.), 152-161.
 References 421

Hönig, Hans G. 1991. “Holmes’ ‘Mapping Theory’ and the landscape of mental translation
processes”. In: Leuven-Zwart, Kitty M. – Ton Naaijkens (eds.), 77-89.
Humboldt, Wilhelm von. 1903-1936. Wilhelm von Humboldts Werke (edited by Albert
Leitzmann). Berlin: Behr.
Hymes, Dell. 1971. On communicative competence. Philadelphia: University of
Pennsylvania Press.
Ivanova, Adelina. 1998. “Educating the ‘language elite’. Teaching translation for
translator training”. In: Malmkjaer, Kirsten (ed.), 92-109.
Jakobsen, Arnt L. 1994. “Starting from the (other) end: Integrating translation and text
production”. In: Dollerup, Cay – Annette Lindegaard (eds.), 143-150.
Jakobsen, Arnt L. 2002. “Orientation, segmentation and revision in translation”. In:
Gyde Hansen (ed.), 191-204.
Jakobsen, Arnt. L. 2005. “Investigating expert translators’ processing knowledge”. In:
Dam, Helle V. – Jan Engberg – Heidrun Gerzymisch-Arbogast (eds.), 173-
189.
Jakobsen, Arnt L. 2006. “Research methods in translation – Translog”. In: Sullivan, Kirk
P.H. – Eva L. Lindgren (eds.), 95-105.
Jakobsen, Arnt L. 2011. “What's in the eye of translators?: Translog with eyetracking”.
In: Schmitt, Peter – Susann Herold – Anette Weilandt (eds.), 343-353.
Jakobsen, Arnt L. – Lasse Schou. 1999. “Translog documantation”. Copenhagen Studies
in Language 24: 149-184.
Jakobson, Roman. 1959. “On linguistic aspects of translation”. In: Brower, Reuben A.
(ed.), 232-239.
Jääskeläinen, Riitta. 1989. “Translation assignment in professional versus non-
professional translation: a think aloud protocol study”. In: Séguinot, Candance
(ed.), 87-98.
Jääskeläinen, Ritta. 1996. “Hard work will bear beautiful fruit. A comparison of two
think-aloud protocol studies”. META 41,1: 61-74.
Jääskeläinen, Riitta. 1999. Tapping the process: an explorative study of the cognitive
and affective factors involved in translating. Joensuu: University of Joensuu.
Jääskeläinen, Riitta. 2010. “Are all professionals experts? Definitions of expertise and
reinterpretation of research evidence in process studies”. In: Shreve, Gregory,
M. – Erik Angelone (eds.), 213-227.
Jääskeläinen, Riitta – Tirkkonen-Condit, Sonja. 1991. “Automatised processes in
professional vs. non-professional translation: A think-aloud protocol study”.
In: Tirkkonen-Condit, Sonja (ed.), 89-109.
Jarvis, Scott – Aneta Pavlenko. 2008. Crosslinguistic influence in language and
cognition. New York: Routledge.
Jensen, Astrid. 1999. “Time pressure in translation”. Copenhagen Studies in Language
24: 103-119.
Jensen, Astrid – Arnt L. Jakobsen. 2000. “Translating under time pressure”. In:
Chesterman, Andrew – Natividad Gallardo San Salvador – Yves Gambier
(eds.), 105-116.
422 References

Jensen, Eric. 1998. Teaching with the brain in mind. Alexandria, VA: Association for
Supervision and Curriculum Development.
Joerges, Bernward. 1977. “Wissenschaftliche Kreativtät: empirische und
wissenchaftliche Hinweise”. In: Schneider, Johann – Marianne Schneider-
Düker (eds.), 143-182.
Jonasson, Kerstin. 1998. “Degree of text awareness in professional vs. non-professional
translators”. In: Beylard-Ozeroff, Ann – Jana Králová – Barbara Moser-
Mercer (eds.), 189-200.
Jones, Roderic. 2002. Conference interpreting explained. Manchester: St. Jerome
Publishing.
Jung, Carl G. 1923/1971. The collected works of C.G. Jung: Vol 6. Psychological types.
Princeton: Princeton University Press.
Kalat, James. 2007. Biological psychology. Belmont: Wadsworth.
Kalat, James. 2010. Introduction to psychology. 9th edition. Belmont: Wadsworth.
Károly, Krisztina – Ágota Fóris (eds.). 2005. New trends in Translation Studies.
Budapest: Akadémiai Kiadó.
Keane, Mark T. – Kenneth J. Gilhooley (eds.). 1992. Advances in the psychology of
thinking. Hempstead, UK: Hemel.
Kearns, John. 2006. “In the broader context: Reflections on curriculum design principles
for training translators”. Translation Ireland. New Vistas in Translator and
Interpreter Training (Special issue) 17, 1: 205-218.
Kearns, John. 2008. “The academic and the vocational in translator education”. In:
Kearns John (ed.), 184-211.
Kearns, John (ed.). 2008. Translator and interpreter training: Issues, methods and
debates. London: Continuum.
Kecskes, Istvan – Liliana Albertazzi (eds.). 2007. Cognitive aspects of bilingualism.
New York: Springer.
Kecskes, Istvan – Tunde Papp. 2000. Foreign language and mother tongue. Mahwah:
Lawrence Erlbaum.
Kelly, Dorothy. 2005. A handbook for translator trainers. Manchester: St. Jerome
Publishers.
Kiraly, Donald. 1995. Pathways to translation: pedagogy and process. Kent, Ohio: Kent
State University Press.
Kiraly, Donald. 1997. “Think-Aloud Protocols and the construction of a professional
translator self-concept”. In: Danks, Joseph H. – Gregory, M. Shreve –
Stephen, B. Fountain – Michael, K. McBeath (eds.), 137-160.
Kiraly, Donald. 2000. A social constructivist approach to translator education.
Manchester: St. Jerome Publishing.
Kiraly, Donald. 2005a. “Situating praxis in translator education”. In: Károly, Krisztina –
Ágota Fóris (eds.), 117-138.
Kiraly, Donald. 2005b. “Project-based learning: A case for situated translation”. META 50,4:
1098-1111. ( http://www.erudit.org/revue/meta/2005/v50/n4/012063ar.html) (date
of access 10 March 2010).
 References 423

Kirsner, Kim – Marilyn C. Smith – R.S. Lockhart – M.L. King – M. Jain. 1984. “The
bilingual lexicon: Language-specific units in an integrated network”. Journal
of Verbal Learning and Verbal Behavior 23: 519-539.
Kirsner, Kim – Craig Speelman. 1998. “Introduction and overview”. In: Kirsner, Kim –
Craig Speelman – Murray Maybery – Angela O’Brien-Malone – Mike
Anderson – Colin MacLeod (eds.), 3-12.
Kirsner, Kim – Craig Speelman – Murray Maybery – Angela O’Brien-Malone – Mike
Anderson – Colin MacLeod (eds.). 1998. Implicit and explicit mental
processes. Mahwah, NJ: Lawrence Erlbaum Associates.
Klaudy, Kinga. 2003. Languages in translation. Lectures on the theory, teaching and
practice of translation. Budapest: Scholactica.
Klir, George J. 2001. Facets of Systems Science. New York: Plenum Publishers.
Koh, Hian Ch. 1996. “Testing hypothesis of entreprenourian characteristics”. Journal of
Managerial Psychology 11,3: 12-25.
Kolers, Paul A. 1963. “Interlingual word association”. Journal of Verbal Learning and
Verbal Behavior 2: 291-300.
Kolers, Paul A. 1973. “Translation and bilingualism”. In: Miller, George A. (ed.), 280-290.
Koller, Werner. 1972. Grundprobleme der Übersetzungstheorie. Unter besonderer
Berücksichtigung schwedisch-deutscher Übersetzungsfälle. Bern: Francke.
Koller, Werner. 1979. Einführung in die Übersetzungswissenschaft. Heidelberg: Quelle
& Meyer.
Koller, Werner. 1992. Einführung in die Übersetzungswissenschaft, Völlig neu
bearbeitete Auflage. Heidelberg, Wiesbaden: Quelle & Meyer.
Komissarov, Vilen. N. 1991. “Language and culture in translation: Competitors or
collaborators?” TTR: Traduction, Terminologie, Rédaction 4,1: 33-47.
Kopczyński, Andrzej. 1994. “Quality in conference interpreting: Some pragmatic problems”.
In: Snell-Hornby, Mary – Franz Pöchhacker – Klaus Kaindl (eds.), 189-198.
Kordes, Hagen. 1990. “Intercultural learning at school: Limits and possibilities”. In:
Buttjes, Dieter – Michael Byram (eds.), 287-305.
Korunka, Christian – Hermann Frank – Manfred Lueger – Josef Mugler. 2003. “The
entrepreneurial personality in the context of resources, environment, and the
startup process – a configurational approach”. Entrepreneurship Theory and
Practice 28,1: 23-42.
Korzeniowska, Aniela – Piotr Kuhiwczak. 1994. Successful Polish-English translation.
Warszawa: Wydawnictwo Naukowe PWN.
Korzeniowska, Aniela. 1998. Explorations in Polish-English mistranslation problems.
Warszawa: Wydawnictwo Uniwersytetu Warszawskiego.
Kramsch, Claire. 1993. Context and culture in language teaching. Oxford: Oxford
Univeristy Press.
Kramsch, Claire. 2009. The multilingual subject. What language learners say about their
experience and why it matters. Oxford: Oxford Univeristy Press.
Krings, Hans P. 1986a. “Translation problems and translation strategies of advanced
German learners of French (L2)”. In: House, Juliane – Shoshana Blum-Kulka
(eds.), 263-276.
424 References

Krings, Hans. P. 1986b. Was in den Köpfen von Übersetzern vorgeht. Tübingen: Gunter
Narr.
Krings, Hans P. 1988. “Blick in die black box: Eine Fallstudie zum Übersetzungsprozess
bei Berufsübersetzern”. In: Arntz, Reiner (ed.), 393-411.
Krashen, Stephen. 1982. Principles and practice in second language acquisition.
Oxford: Pergamon Press.
Kroll, Judith F. 1993. “Accessing conceptual representations for words in a second
language”. In: Schreuder, Robert – Bert Weltens (eds.), 53-82.
Kroll, Judith – Annette M.B. De Groot. 1997. “Lexical and conceptual memory in the
bilingual: Mapping form to meaning in two languages”. In: Kroll, Judith –
Annette M.B. De Groot (eds.), 169-197.
Kroll, Judith – Annette M.B. De Groot (eds.). 1997. Tutorials in bilingualism:
Psycholinguistic perspectives. Mahwah: Lawrence Erlbaum Publishers.
Kroll, Judith F. – Annette M.B. De Groot (eds.). 2005. Handbook of bilingualism:
Psycholinguistic approaches. New York: Oxford University Press.
Kroll, Judith F. – Jared A. Linc. 2007. “Representation and skill in second language
learners and proficient bilinguals”. In: Kecskes, Istvan – Liliana Albertazzi
(eds.), 237-269.
Kroll, Judith F. – Alexandra Sholl. 1992. “Lexical and conceptual memory in fluent and
nonfluent bilinguals.” In: Harris, Richard J. (ed.), 191-203.
Kroll, Judith F. – Erika Stewart. 1994. “Category interference in translation and picture
naming: evidence for asymmetric connections between bilingual memory
representations. Journal of Memory and Language 33: 149-174.
Kuliniak, Joanna. 2009. On the use and overuse of dictionaries by inexperienced
translators. Unpublished BA thesis. Poznań. Adam Mickiewicz University.
Kurz, Ingrid. 2000. “Translators and interpreters: different learning styles?”. Across
Languages and Cultures 1,1: 71-83.
Kussmaul, Paul. 1991. “Creativity in the translation process. Empirical approaches”. In:
Leuven-Zwart, Kitty M. – Ton Naaijkens (eds.), 91-101.
Kussmaul, Paul. 1995. Training the translator. Amsterdam: John Benjamins.
Kussmaul, Paul. 2005. “Translation through visualization”. META 50,2: 378-391.
Kussmaul, Paul – Sonja Tirkkonen-Condit. 1995. “Think-aloud protocol analysis in
Translation Studies. TTR: Traduction, Terminologie, Rédaction 8,1: 177-199.
Lado, Robert. 1957. Linguistics across cultures: Applied linguistics for language
teachers. Michigan: University of Michigan Press.
Lado, Robert. 1964. Language teaching: A scientific approach. New York: McGraw
Hill.
La Heij, Wido. 2005. “Selection processes in monolingual and bilingual lexical access”.
In: Kroll, Judith F. – Annette M.B. De Groot (eds.), 289-307.
Lakoff, George. 1982. Categories and cognitive models. Trier: LAUT.
Lambert, Wallace M. 1978. “Psychological approaches to bilingualism, translation and
interpretation”. In: Gerver, David – Wallace H. Sinaiko (eds.), 131- 143.
Landers, Clifford, E. 2001. Literary translation: A practical guide. Clevedon:
Multilingual Matters.
 References 425

Lantolf, James P. (ed). 2000. Sociocultural theory and second language learning.
Oxford: Oxford University Press.
Lars, Andreassen – Line Brandt – Jes Vang (eds.). 2007. Cognitive semiotics. Oxford:
Peter Lang.
Larson, Mildred. 1984. Meaning-based translation: a guide to cross-language
equivalence. Langham: University of America Press.
Larson, Mildred (ed.). 1991. Translation: theory and practice, tension and
interdependence. New York: Binghampton.
Lauffer, Sabine. 2002. “The translation process: an analysis of observational methodology”.
CAD. De Traducao 2,10: 59-75. (http://www.periodicos.ufsc.br/index.php
/traducao/article/viewFile/6144/5702) (date of access 18 May 2010).
Lee, William R. (ed.). 1967. E.L.T. Selections: Articles from the journal English
Language Teaching. London: Oxford University Press.
Leonardi, Vanessa. 2010. The role of pedagogical translation in second language
acquisition. Bern: Peter Lang.
Leopold, Werner F. 1939-1949. Speech development of a bilingual child. A linguist’s
record (4 volumes). Evanston, IL: Northwestern University Press.
Leuven-Zwart, Kitty M. – Ton Naaijkens (eds.). 1991. Translation Studies: The state of
the art. Proceedings of the first James Holmes symposium on Translation
Studies. Amsterdam: Rodopi.
Levelt, Willem J.M. 1989. Speaking: From intention to articulation. Cambridge, MA:
The MIT Press.
Levelt, Willem J.M. 2001. “Spoken word production: A theory of lexical access”. PNAS.
Proceedings of the National Academy of Sciences 98,23: 13464-13471.
Levy, Michael C. – Sarah Ransdell (eds.). 1996. The science of writing. Theories,
methods, individual differences, and applications. Mahwah, NJ: Lawrence
Erlbaum.
Levý, Jiri. 1967. “Translation as a Decision Process”. In: Venuti (ed.), 148-159.
Lewandowska-Tomaszczyk, Barbara. 1985. “ Współczesne prądy w językoznawstwie a
teoria tłumaczenia”. Folia Linguistica 21, Acta Universitatis Lodziensis: 59-71.
Lewandowska-Tomaszczyk, Barbara. 1990. “Overt and hidden negations in translation”.
In: Thelen, Marcel – Barbara Lewandowska-Tomaszczyk (eds.), 234-244.
Lewandowska-Tomaszczyk, Barbara. 2004. “Lexical problems of translation”. In: Frank,
Armin P. – Norbert Greiner – Theo Hermans – Haral Kittel – Werner Koller –
José Lambert – Paul Fritz (eds.), 455-465.
Lewandowska Tomaszczyk, Barbara – Marcel Thelen (eds.). 1992. Translation and
meaning. Part 2. Rijkshogeschool Maastricht: Faculty of Translation and
Interpreting.
Light, Paul – George Butterworth (eds.). 1992. Context and cognition. Hillsdale:
Lawrence Erlbaum.
Ljudskanov, Alexandre. 1969. Traduction humaine et traduction mécanique. Paris:
Dunod.
Lowe, Pardee. 1987. “Revising the ACTFL/ETS scale for a new purpose: rating skill in
translating”. In : Rose, Marilyn G. (ed.), 53-61.
426 References

Lörscher, Wolfgang. 1986. “Linguistic aspects of translation processes: towards an

analysis of translation performance.” In: House, Juliane – Shoshana Blum-
Kulka (eds.), 277-292.
Lörscher, Wolfgang. 1991. Translation performance, translation process and translation
strategies. Tübingen: Narr.
Lörscher, Wolfgang. 1992. “Form- and sense-oriented approaches to translation”. In:
Lewandowska Tomaszczyk, Barbara – Marcel Thelen (eds.), 403-414.
Lörscher, Wolfgang. 1996. “A psycholinguistic analysis of translation processes”. META
41,1: 26-32.
Lörscher, Wolfgang. 1997. “Translation process analysis and its implications for translation
teaching”. In: Hickey, Raymond – Stanislaw Puppel (eds.), 2005-2018.
Mackenzie, Rosemary. 1998. “Creative problem-solving and translator training”. In:
Beylard-Ozeroff, Ann – Jana Králová – Barbara Moser-Mercer (eds.), 201-
206.
Mackintosh, Kristen. 1998. “An empirical study of dictionary use in L2-L1 translation”.
In: Atkins, Sue (ed.), 123-144.
Mackenzie, Rosemary. 2004. “The competencies required by the translator’s roles as
professionals”. In: Malmkjaer, Kirsten (ed.), 31-81.
Malakoff, Marguerite E. 1992. “Translation ability: A natural and metalinguistic skill”.
In: Harris Richard J. (ed.), 515-529.
Malakoff, Marguerite – Kenji Hakuta. 1991. “Translation skill and metalinguistic
awareness in bilinguals”. In: Bialystok, Ellen (ed.), 141-166.
Malavé, Lillian M. – Georges Duquette (eds.). 1991. Language, culture and cognition.
Clevedon: Multilingual Matters.
Malmkjaer, Kirsten (ed.). 1998. Translation and language teaching: language teaching
and translation. Manchester: St. Jerome Publishing.
Malmkjaer, Kirsten (ed.). 2004. Translation in undergraduate degree programmes.
Amsterdam: John Benjamins.
Malmkjaer, Kirsten. 2005. Linguistics and the language of translation. Edinburgh:
Edinburgh University Press.
Marcus, Gary F. 2001. The algebraic mind: Integrating connectionism and cognitive
science (Learning, development, and conceptual change). Cambridge, MA:
The MIT Press.
Marian, Viorica – Michael Spivey. 2003. “Competing activation in bilingual language
processing: within- and between-language competition”. Bilingualism:
Language and Cognition 6,2: 97-115.
Martin, Ricardo M. 2006. “Expertise and environment in translation”. Unpublished
paper presented at the Second IATIS Conference. (http://www.petraweb.
org/Munoz%20Expertise%20and%20Environment%20in%20Translation.pdf)
(date of access 28 February 2010).
Massey, Gary. 2005. “Process-oriented translator training and the challenge for e-
learning”. META 50,2: 626-633.
Matsumoto, David. 2000. Culture and psychology: People around the world. Belmont,
CA: Wadsworth.
 References 427

McClelland, David. 1973. “Testing for competence rather than “intelligence”. American
Psychologist 28,1: 1-14.
McKay, Corrine. 2006. How to succeed as a freelance translator. Lulu.com: Two Rat
Press.
McLaughlin, Barry – Roberto Heredia. 1996. “Information processing approaches to
research on second language acquisition and use”. In: Ritchie, William – Tej
Bhatia (eds.), 213-228.
Meyer, David E. – Sylvan Kornblum (eds.). 1992. Synergies in experimental psychology,
artificial intelligence and cognitive neuroscience. Cambridge, MA: The MIT
Press.
Miller, George A. (ed.). 1973. Communication, language, and meaning, psychological
perspectives. New York: Basic Books, Inc. Publishers.
Mitchell, Rosamond – Florence Myles. 1998/2006. Second Language Learning
Theories. London: Arnold.
Mora, Jill. K. 2007. “Metalinguistic awareness as defined through research”.
(http://coe.sdsu.edu/people/jmora/moramodules/MetaLingResearch.htm) (date
of access 28 October 2009).
Morales, Alejandro – William, E. Hanson. 2005. “Language brokering: an integrative
review of the literature”. Hispanic Journal of Behavioral Sciences 27,4: 471-
503.
Morton, John. 1980. “The logogen model and orthographic structure”. In: Frith, Uta
(ed.), 117-133.
Mossop, Brian. 2001. Revising and editing for translators. Manchester: St. Jerome
Publishing.
Mossop, Brian. 2007. “Empirical studies of revision: what we know and need to know”.
(http://www.jostrans.org/issue08/art_mossop.pdf) (date of access 15 March
2010).
Neubert, Albrecht. 1985. Text and translation. Leipzig: Enzyklopädie.
Neubert, Albrecht. 1992. “Competence in translation: a complex skill, how to study and
how to teach it”. In: Snell-Hornby, Mary – Franz Pöchhacker – Klaus Kaindl
(eds.), 411-420.
Neubert, Albrecht. 1997. “Postulates for a theory of translation”. In: Danks, Joseph H. –
Gregory M. Shreve – Stephen B. Fountain – Michael K. McBeath (eds.), 1-24.
Neubert, Albrecht. 2000. “Competence in language, languages, and in translation”. In:
Schäffner, Christine – Beverly Adab (eds.), 3-18.
Neumann, John Von – Oscar Morgenstern. 1944. The theory of games and economic
behavior. Princeton: Princeton University Press.
Newson, Dennis. 1998. “Translation and foreign language learning”. In: Malmkjaer
(ed.), 63-68.
Newmark, Peter. 1969. “Some notes on translation and translators”. Incorporated
Linguist 12,1: 9-15.
Newmark, Peter. 1988. A Textbook of translation. New York: Prentice Hall.
Nicol, Janet (ed.). 2001. One mind, two languages: Bilingual language processing.
Oxford: Blackwell.
428 References

Nida, Eugene A. 1964. Towards a science of translating. Leiden: E. J. Brill.

Nida, Eugene A. 1981. “Translators are born not made”. Babel 15,4: 214-215.
Nida, Eugene A. 2001. Contexts in translating. Amsterdam: John Benjamins.
Nida, Eugene A. 2002. “Translators confrontations with false ideas about language”.
(http://translatoruy.files.wordpress.com/2009/12/nida-european-commission-
translation-service.pdf) (date of access 10 January 2010).
Norberg, Ulf. 2003. Übersetzen mit doppeltem skopos. Eine empirische Prozess und
Produktstudie. Uppsala: Uppsala University Library.
Nord, Christine. 1988. Textanalyse und Übersetzen. Heilderberg: Groos.
Nord, Christine. 1991. Text analysis in translation. Amsterdam: Rodopi.
Nord, Christiane. 2005. Text analysis in translation: Theory, methodology and dydactic
application. Amsterdam: Rodopi.
Norman, Don A. – Tim Shallice. 1986. “Attention to action: willed and automatic
control of behaviour”. In: Davidson, Richard J. – Gary Schwartz, Gary –
David Shapiro (eds.), 1-18.
Odlin, Terence. 1989. Language transfer: cross-linguistic influence in language
learning. Cambridge: Cambridge University Press.
O’Malley, Michael J. – Anna U. Chamot. 1990. Learning strategies in second language
acquisition. Cambridge: Cambridge University Press.
Orozco, Mariana – Amparo Hurtado Albir. 2000. “Measuring translation competence
acquisition”. META 47,3: 375-402.
PACTE. 2000. “Acquiring translation competence: Hypotheses and methodological
problems in a research project”. In: Beeby, Allison – Doris Ensinger – Marisa
Presas (eds.), 99-106.
PACTE. 2003. “Building a translation competence model”. In: Alves, Fabio (ed.), 43-66.
Pagano, Adriana – Celia Magalhães – Fabio Alves (eds.). 2005. Competência em
tradução: cognição e discurso. Belo Horizonte: Editora UFMG.
Paradis, Michel. 1992. “Neurosciences et apprentissage des langues.” In: Girard, Denis
(ed.), 21-31.
Paradis, Michel. 1994. “Toward a neurolinguistic theory of simultaneous translation:
The framework. International Journal of Psycholinguistics 10: 319-335.
Paradis, Michel. 1997. “The cognitive neuropsychology of bilingualism”. In: De Groot,
Annette M.B. – Judith F. Kroll (eds.), 331-354.
Paradis, Michel. 1998. Neurolinguistics of bilingualism and the teaching of languages”.
(www.semioticon.com/virtuals/multimodality/paradis.pdf) (date of access 10
March 2008).
Paradis, Michel. 2004. A neurolinguistic theory of bilingualism. Amsterdam: John
Benjamins.
Paradis, Michel. 2005. “Introduction to part IV. Aspects of bilingualism”. In: Kroll,
Judith F. – Annette M.B. De Groot (eds.), 411-415.
Paradis, Michel. 2007. “The neurofunctional components of the bilingual cognitive
system”. In: Kecskes, Istvan – Liliana Albertazzi (eds.), 3-28.
Paradis, Michel. 2009. Declarative and procedural determinants of second languages.
Amsterdam: John Benjamins.
 References 429

Paradis, Michel – Marie-Claire Goldblum – Raouf Abidi. 1982. “Alternate antagonism

with paradoxical translation behaviour in two bilingual aphasic patients”.
Brain and Language, 15,1: 55-69.
Pavlenko, Aneta. 1999. “New approaches to concepts in bilingual memory”.
Bilingualism: Language and Cognition 2: 209-230.
Pavlenko, Aneta. 2005. “Bilingualism and thought”. In: Kroll, Judith – Annette M.B. De
Groot (eds.), 433-453.
Pavlenko, Aneta (ed.). 2009. The bilingual mental lexicon: Interdisciplinary approaches.
Bristol: Multilingual Matters.
Pavlenko, Aneta – Adrian Blackledge – Ingrid Piller – Marya Teutsch-Dwyer (eds.).
2001. Multilingualism, second language learning and gender. Berlin: Mouton
de Gruyter.
Pavlenko, Aneta – James P. Lantolf. 2000. “Second language learning as participation
and the (re)construction of selves”. In: Lantolf James P. (ed), 155-177.
Pawlak, Mirosław (ed.). 2008. Investigating English language learning and teaching.
Poznań & Kalisz: Faculty of Fine Arts in Kalisz /Adam Mickiewicz
University in Poznań.
Paz, Ottavio. 1971. Traduccion: Literatura y Literalidad. Barcelona: Tusquets.
Peal, Elizabeth – Wallace E. Lambert. 1962. “The relationship of bilingualism to
intelligence”. Psychological Monographs: General and Applied 76, 27: 1-23.
Pellatt, Valerie. 2002. “Anaphoric resolution of zero pronouns in Chinese in translation
and reading comprehension”. Unpublished doctoral dissertation. Edinburgh:
Edinburgh University.
Pellatt, Valerie. 2009. “Translation as a reading comprehension test: Schemata and the
role of the ‘write it down protocol”. In: Witte, Arnd – Theo Harden –
Alessandra Ramos de Oliveira Harden (eds.), 343-357.
Penfield, Wilder G. 1953. “A consideration of the neurophysiological mechanisms of
speech and some educational consequences”. Proceedings of the American
Academy of Arts and Science 82,5: 201-214.
Penfield, Wilder G. – Lamar Roberts. 1959. Speech and brain mechanisms. New York:
Atheneum Press.
Perdek, Magdalena. 2011. English phrasal verbs in translation: A lexicographic and
corpus study of equivalence. Unpublished doctoral dissertation. Poznań.
Adam Mickiewicz University.
Pimsleur, Paul – Daniel J. Reed – Charles W. Stansfield. 2004. Pimsleur language
aptitude battery: Manual, 2004 Edition. Bethesda, MD: Second Language
Testing, Inc.
Ping, Ke. 2009. “Machine Translation”. In: Baker, Mona – Gabriela Saldanha (eds.),
162-169.
Pinker, Steven. 2007. The stuff of thought. New York: Viking.
Piotrowska, Maria. 2002. A compensational model for strategy and techniques in
teaching translation. Kraków: Wydawnictwo Naukowe Akademii
Pedagogicznej.
430 References

Piotrowska, Maria. 2007. Proces decyzyjny tłumacza: podstawy metodologii nauczania

przekładu pisemnego. Kraków: Wydawnictwo Naukowe Akademii
Pedagogicznej.
Pisarska, Alicja. 1989. Creativity of translators. Poznań: Adam Mickiewicz University.
Pisarska, Alicja. 2004 “Metaphor and other tropes as translation problems: A linguistic
perspective.” In: Frank, Armin P. – Norbert Greiner – Theo Hermans – Haral
Kittel – Werner Koller – José Lambert – Paul Fritz (eds.), 520-527.
Płusa, Paweł. 2000. Kształcenie tłumaczy. Częstochowa: Wydawnictwo Wyższej Szkoły
Pedagogicznej w Częstochowie.
Płusa, Paweł. 2007. Rozwijanie kompetencji przekładu i kształcenie tłumaczy. Katowice:
Wydawnictwo Uniwersytetu Śląskiego.
Poincaré, Henri. 1913. The foundations of science. New York: Science Press.
Pokorn, Nike. 2005. Challenging the traditional axioms: translation into the non-mother
tongue. Amsterdam: John Benjamins.
Pöchhacker, Franz – Miriam Shlesinger (eds.). 2002. The Interpreting Studies reader.
London: Routledge.
Potter, Mary C. – Kwok-Fai So – Barbara Von Eckardt – Laurie B. Feldman. 1984.
“Lexical and conceptual representation in beginning and more proficient
bilinguals. Journal of Verbal Learning and Verbal Behavior 23: 23-38.
Poulisse, Nanda – Theo Bongaerts. 1994. “First language use in second language
production”. Applied Linguistics 15: 36-57.
Prat, Chantel S. – Timothy Keller – Marcel A. Just. 2007. “Individual differences in
sentence comprehension: A functional magnetic resonance imaging
investigation of syntactic and lexical processing demands”. Journal of
Cognitive Neuroscience 19,12: 1950-1963.
Presas, Marisa. 2000. “Bilingual competence and translation competence.” In: Schäffner,
Christine – Beverly Adab (eds.), 19-31.
Preston, Malcolm S. – Wallace E. Lambert. 1969. “Interlingual interference in a
bilingual version of the Stroop color-word task”. Journal of Verbal Learning
and Verbal Behavior 8: 295-301.
Price, Cathy J. – David Green – Roswitha von Studnitz. 1999. “A functional imaging
study of translation and language switching”. Brain 122,12: 2221-2235.
Prior, Anat – Brian MacWhinney. 2010. “A bilingual advantage in task switching”.
Bilingualism: Language and Cognition 13: 253-262.
Pym, Anthony. 1993. Epistemological problems in translation and its teaching.
Calaceite: Caminade.
Pym, Anthony. 1998. “Okay, so how are translation norms negotiated? A question for
Gideon Toury and Theo Hermans”. Current Issues in Language and Society
5,1-2: 107-113.
Pym, Anthony. 2003. “Redefining translation competence in an electronic age. In
defence of a minimalist approach”. META 48: 481-497.
Rapoport, Anatol. 1961. Fights, games and debates. Michigan: Michigan University
Press.
 References 431

Rapp, Brenda (ed.). 2001. The handbook of cognitive neuropsychology. New York:
Psychology Press.
Rayner, Keith. 1998. “Eye movements in reading and information processing: 20 years
of research”. Psychological Bulletin 124: 372–422.
Reiss, Katharina. 1981. “Type, kind and individuality of text. Decision making in
translation”. Poetics Today 2,4: 121-131.
Reiss, Katharina – Hans, J. Vermeer. 1984. Grundlegung einer allgemeinen
Translationstheorie. Tübingen: Niemeyer.
Rihimaki, Mikko. 2005. “Game theory in translation: memetics as a recourse”.
(www.uta.fi/mikko.rihimaki) (date of access 12 June 2009).
Risku, Hanna. 1998. Translatorische Kompetenz. Kogmitive Grundlegen des
Übersetzens als Expertentätigkeit. Tübingen: Stauffenburg.
Risku, Hanna. 2002. “Situatedness in Translation Studies”. Cognitive Systems Research
3: 523-533.
Ritchie, William – Tej Bhatia (eds.). 1996. A handbook of Second Language Acquisition.
New York: Academic Press.
Robinson, Douglas. 1997. Becoming a translator. London: Routledge.
Roberts, Roda P. 1992. “Translation pedagogy: strategies for improving dictionary use”.
TTR: Traduction, Terminologie, Rédaction 5,1: 49-76.
Roberts, Roda. P. 1997. “Using dictionaries efficiently”. Paper presented at 38th Annual
Conference of the American Translators Association, San Francisco,
California. (http://www.dico.uottawa.ca/articles-en.htm) (date of access: 20
September 2010).
Rodriguez-Fornells, Antoni – Ruth De Diego Balaguer – Thomas F. Münte. 2006.
“Executive control in bilingual language processing”. Language Learning 56:
133-190.
Roehr, Karen. 2008. “Metalinguistic knowledge and language-analytic ability in
university-level L2 learners”. Applied Linguistics 29: 173-199.
Rose, Marilyn G. (ed.). 1987. Translation excellence: assessment, achievement,
maintenance. New York: SUNY Binghamton Press.
Rose, Marilyn G. 1991. “Seeking synapses: Translators describe translating.” In: Larson,
Midred (ed.), 5-12.
Rosi-Landi, Ferruccio. 1973. Ideologies of linguistic relativity. The Hague: Mouton.
Roussey, Jean-Yves – Annie Piolat – F. Guercin. 1990. “Revising strategies for different
text types”. Language and Education 4,1: 51-65.
Ruiz, Carmen – Natalia Paredes – Pedro Macizo – Maria T. Bajo. 2008. “Activation of
lexical and syntactic target language properties in translation. Acta
Psychologica 128: 490-500.
Rumjahn, Hossain, – Salili Farideh (eds.). 2005. Language in multicultural education.
Connecticut: Information Age Publishing Inc.
Rysiewicz, Jacek. 2008. “Measuring foreign language learning aptitude. Polish
adaptation of The Modern Language Aptitude Test by Carroll and Sapon”.
Poznań Studies in Contemporary Linguistics 44,4: 569-595.
432 References

Saarinen, Esa – Raimo P. Hämäläinen. 2007. “Systems intelligence: Connecting

engineering thinking with human sensitivity”. In: Hämäläinen, Raimo P. –
Esa Saarinen (eds.), 51-77.
Salas, Eduardo – Michael A. Rosen – Deborah Diaz Granados. 2010. “Expertise-based
intuition and decision making in organizations”. Journal of Management 36,4:
941-973.
Samuelsson-Brown, Geoffrey. 2010. A practical guide for translators [fifth revised
edition]. Clevedon: Multilingual Matters.
Schäffner, Christina. 1998. “Parallel texts in translation”. In: Bowker, Lynne – Michael
Cronin – Dorothy Kenny – Jennifer Pearson (eds.), 83-90.
Schäffner, Christine (ed.). 2002. The role of discourse analysis for translation and in
translator training. Clevedon: Multilingual Matters.
Schäffner, Christine – Beverly Adab (eds.). 2000. Developing translation competence.
Amsterdam: John Benjamins.
Schank, Roger. 1986. Explanation patterns: Understanding mechanically and creatively.
Hillsdale, NJ: Lawrence Erlbaum Associates.
Schilperoord, Joost. 1996. It’s about time: Temporal aspects of cognitive processes in
text production. Amsterdam: Rodopi.
Schmidt, Morena A. 2005. How do you do it anyway? A longitudinal study of three
translator students translating from Russian into Swedish. Doctoral
dissertation. Stockholm: Acta Universitatis Stockholmiensis.
Schmitt, Peter – Susann Herold – Anette Weilandt (eds.). 2011. Translationsforschung:
Tagungsberichte der LICTRA IX. Frankfurt am Main: Peter Lang.
Schneider, Johann – Marianne Schneider-Düker (eds.). 1977. Interpretationen der
Wirkichkeit. Saarbrücken: Schriften.
Schön, Donald. 1983. The reflective practitioner: How professionals think in action.
London: Temple Smith.
Schottländer, Rudolf. 1972. “Paradoxien der Kreativität”. Zeitschrift für philosophische
Forschung XXVI,2: 153-170.
Schreuder, Robert – Bert Weltens (eds.). 1993. The bilingual lexicon. Amsterdam: John
Benjamins.
Schulte, Rainer – John Biguenat (eds.). 1992. Theories of translation. Chicago:
University of Chicago Press.
Schumann, John H. 1998. The neurobiology of affect in language. Oxford: Blackwell.
Schweda-Nicholson, Nancy. 2005. “Personality characteristics of interpreter trainees: the
Myers-Briggs Type Indicator (MBTI)”. The Interpreters’ Newsletter 13: 109-142.
Segalowitz, Norman S. – Sidney J. Segalowitz. 1993. “Skilled performance, practice,
and the differentiation of speed-up from automatization effects; evidence from
second language word recognition”. Applied Psycholinguistics 14: 369-385.
Segalowitz, Sidney J. – Norman S. Segalowitz – Anthony G. Wood. 1998. “Assessing
the development of automacity in second language word recognition”.
Applied Psycholinguistics 19: 53-67.
Segalowitz, Norman S. – Jan Hulstijn. 2005. “Automaticity in bilingualism and second
language learning”. In: Kroll, Judith – Annette M.B. De Groot (eds.), 371-388.
 References 433

Séguinot, Candance (ed.). 1989a. The translation process. Toronto: H.G. Publications.
Séguinot, Candance. 1989b. “Understanding why translators make mistakes”. TTR:
Traduction, Terminologie, Rédaction 2,2: 73-81.
Séguinot, Candance. 1997. “Accounting for variability in translation”. In : Danks, Joseph
H. – Gregory M. Shreve – Stephen B. Fountain – Michael K. McBeath (eds.),
104-119.
Séguinot, Candance. 2000. “Management issues in the translation process”. In:
Tirkkonen-Condit, Sonia – Ritta Jääskeläinen (eds.), 143-148.
Seleskovitch, Danica. 1976. “Interpretation: a psychological approach to translating”. In:
Brislin Richard W. (ed.), 92-116.
Seleskovitch, Danica. 1978. “Language and cognition”. In: Gerver, David – Wallace H.
Sinaiko (eds.), 333-341.
Selinker, Larry. 1972. “Interlanguage”. International Review of Applied Linguistics 10,3:
209-231.
Senge, Peter. 1990. The fifth discipline: The art and practice of the learning
organization. New York: Doubleday Currency
Senge, Peter. 1999. The dance of change: The challenges of sustaining momentum in
learning organizations. London: Brealey.
Sercu, Lies. 2004. “Researching the acquisition of intercultural communicative
competence in a foreign language: Setting the agenda for a research area”. In:
Esch, Kees van – John, Oliver St. (eds.), 131-156.
Sewell, Penelope. 1996. “Translation in the curriculum”. In: Sewell, Penelope – Ian
Higgins (eds), 135-159.
Sewell, Penelope. 2004. “Students buzz round the translation class like bees round the
honey pot – why?”. In: Malmkjaer, Kirsten (ed.), 151-162.
Sewell, Penelope – Ian Higgins (eds). 1996. Teaching translation in universities. Present
and future perspective. London: CILIT.
Sfard, Anna. 1998. “On two metaphors for learning and the danger of choosing just
one”. Educational Researcher 27,2: 4-13.
Shannon, Sheila M. 1987. English in the Barrio: a sociolinguistic study of second
language contact. Unpublished doctoral dissertation. Stanford University.
Shannon, Sheila M. 1990. “English in the Barrio: The quality of contact among
immigrant children”. Hispanic Journal of Behavioral Sciences 12: 256-276.
Shlesinger, Miriam. 2000. Strategic allocation of working memory and other attentional
resources in simultaneous interpreting. Unpublished doctoral dissertation. Bar
Ilan.
Shreve, Gregory M. 1997. “Cognition and the evolution of translation competence”. In:
Danks, Joseph H. – Gregory M. Shreve – Stephen B. Fountain – Michael K.
McBeath (eds.), 120-136.
Shreve, Gregory M. – Geoffrey S. Koby. 1997. “Introduction”. In: Danks, Joseph E. –
Gregory M. Shreve – Stephen B. Fountain – Michael K. McBeath (eds.). xi-
xviii.
Shreve, Gregory M. – Erik Angelone (eds.). 2010. Translation and cognition.
Amsterdam: John Benjamins.
434 References

Sinclair, Anne – Robert J. Jarvella – Willem J. M. Levelt (eds.). 1978. The child’s
conception of language. Berlin: Springer Verlag.
Sirén, Seija – Kai Hakkarainen. 2002. “Expertise in translation”. Across Languages and
Cultures 3.1: 71-82.
Snell-Hornby, Mary. 1988. Translation Studies: An integrated approach. Amsterdam:
John Benjamins.
Snell-Hornby, Mary. 1992. “The professional translator of tomorrow: language specialist
or all-around expert?” In: Dollerup Cay – Anne Loddegaard (eds.), 9-22.
Snell-Hornby, Mary. 2002. “Back to square one? – on the troubled relationship between
Translation Studies and linguistics”. Logos and Language 3,1: 21-30.
Snell-Hornby, Mary. 2006. The turns of Translation Studies. Amsterdam: John
Benjamins.
Snell-Hornby, Mary – Doina Motas – Jennifer Williams. 1992. “New translation
departments – challenges of the future”. In: Snell-Hornby, Mary – Franz
Pöchhacker – Klaus Kaindl (eds.), 431-434.
Snell-Hornby, Mary – Franz Pöchhacker – Klaus Kaindl (eds.). 1992. Translation
Studies: An interdiscipline. Amsterdam: John Benjamins.
Sofer, Morry. 1997/2006. The translator’s handbook. Rockville: Schreiber Publishing.
Sommer, Doris (ed.). 2003. Bilingual games. Some literary investigations. New York:
Palgrave Macmillan.
Sperber, Dan – Deadry Wilson. 1986. Relevance: communication and cognition. Oxford:
Blackwell.
Squire, Larry – Eric R. Kandel. 2009. Memory: From mind to molecules. Greenwood:
Roberts & Company.
Stansfield, Charles W. – Mary L. Scott – Dorry M. Kenyon. 1992. “The measurement of
translation ability”. The Modern Language Journal 76,4: 455-467.
Steinberg, Robert J. (ed.). 1984. Mechanisms of cognitive development. San Francisco:
Freeman.
Stibbard, Richard. 1998. “The principled use of oral translation in foreign language
teaching”. In: Malmkjaer, Kirsten (ed.), 69-76.
Sternberg, Robert. 1998. “Abilities are forms of developing expertise”. Educational
Researcher 27,3: 11-20.
Sullivan, Kirk P.H. – Eva L. Lindgren (eds.). 2006. Computer keystroke logging and
writing: methods and applications. Oxford: Elsevier.
Stolze, Radegundis. 2004. “Creating “presence” in translation”. In: Hansen Gyde –
Kirsten Malmkjaer – Daniel Gile (eds.), 39-71.
Stroop, John R. 1935. “Studies of interference in serial verbal reactions”. Journal of
Experimental Psychology 18: 643-662.
Superczyński, Tomasz. 2009. Translators as critical readers of their translations. A
study of revising strategies. Unpublished BA thesis. Poznań. Adam
Mickiewicz University.
Swain, Merril – Guy Dumas – Neil Naiman. 1974. “Alternatives to spontaneous speech:
elicited translation and imitation as indicators of second language
competence”. Working Papers on Bilingualism 3: 68-79.
 References 435

Swain, Merril – Sharon Lapkin. 2000. “Task-based second language learning: the uses of
the first language”. Language Teaching Research 4,3: 251-274.
Tabakowska, Elżbieta. 2001. Językoznawsto kognitywne a poetyka przekładu. Kraków:
Universitas.
Tabakowska, Elżbieta. 2003. O przekładzie na przykładzie. Kraków: Znak.
Taylor, Chris. 1990. The error analysis of translation and its application to language
teaching. Trieste: SSLMIT, University of Trieste.
Tennent, Martha (ed.). 2005. Training for the new millennium. Amsterdam: John
Benjamins.
Thelen, Marcel – Barbara Lewandowska-Tomaszczyk (eds.). 1990. Translation and
Meaning, Part 1. Maastricht: Euroterm.
Thiery, Christopher. 1978. “True bilingualism and second language learning”. In:
Gerver, David – Wallace H. Sinaiko (eds.), 145-153.
Thierry, Guillaume – Yan J. Wu. 2007. “Brain potentials reveal unconscious translation
during foreign-language comprehension”. PNAS 30,104: 12530-12535.
Tirkkonen-Condit, Sonja. 1990. “Professional vs. non-professional translation: a think
aloud protocol study”. In: Halliday, Michael A.K. – John Gibbons – Howard
Nicholas (eds.), 381-394.
Tirkkonen-Condit, Sonja (ed.). 1991. Empirical research in translation and intercultural
studies: Selected papers of the TRANSIF seminar, Savonlinna 1988.
Tübingen: Narr.
Tirkkonen-Condit, Sonja – John. Laffling (eds.). 1993. Recent trends in empirical
translation research. Joensuu: University of Joensuu.
Tirkkonen-Condit, Sonja – Ritta Jääskeläinen (eds.). 2000. Tapping and mapping the
translation process. Amsterdam: John Benjamins.
Tokowicz, Natasha – Erika B. Michael – Judith F. Kroll. 2004. “The roles of study
abroad experience and working memory capacity in the types of errors made
during translation”. Bilingualism: Language and Cognition 7: 255-272.
Tomasello, Michael (ed.). 1998. The new psychology of language. London: Lawrence
Erlbaum Associates.
Toury, Gideon. 1984. “The notion of ‘native translator’ and translation teaching”. In:
Wilss, Wolfram – Gisela Thome (eds.), 186-195.
Toury, Gideon. 1986. “Monitoring discourse transfer: A test case for developmental model
of translation”. In: House, Juliane – Shoshana Blum-Kulka (eds.), 79-94.
Toury, Gideon. 1995. Descriptive Translation Studies and beyond. Amsterdam: John
Benjamins.
Tunmer, William E. – Marion E. Myhill. 1984. “Metalinguistic awareness and
bilingualism”. In: Tunmer, William E. – Chantel Prat – Michael L. Herriman
(eds.), 169-187.
Tunmer, William E. – Chantel Prat – Michael L. Herriman (eds.). 1984. Metalinguistic
awareness in children. Berlin: Springer.
Tymczyńska, Maria. 2011. Lexical processing in online translation tasks. The case of
Polish-English-German professional and trainee conference interpreters.
Unpublished doctoral dissertation. Poznań. Adam Mickiewicz University.
436 References

Tymoczko, Maria. 2005. “Trajectories of research in Translation Studies”. META 50,4:

1082-1097.
Tymoczko, Maria. 2007. Enlarging translation, empowering translators. Manchester: St
Jerome.
Tymoczko, Maria – Edwin Gentzler (ed.). 2007. Translation and power. Amherst:
University of Massachusetts Press.
Uwajeh, M.K.C. 1994. “Teaching linguists translation”. In: Dollerup, Cay – Annette
Lindegaard (eds.), 287-294.
Varantola, Krista. 1998. “Translators and their use of dictionaries: User needs and user
habits”. In: Atkins, Sue (ed.), 179-192.
Venuti, Lawrence. 1995. The translator’s invisibility: a history of translation. New
York: Routledge.
Venuti, Lawrence (ed.). 2000. The Translation Studies reader. London: Routledge.
Verhoeven, Ludo – John de Jong (eds.). 1992. The construct of language proficiency.
Amsterdam: John Benjamins.
Vermeer, Anne. 1992. “Exploring the second learner language lexicon”. In: Verhoeven,
Ludo - John de Jong (eds.), 147-162.
Vermes, Albert. 2010. “Translation in foreign language teaching: a brief overview of
pros and cons”. Eger Journal of English Studies X: 83-93.
Vienne, Jean. 1998. “Teaching what they didn’t learn as language students”. In:
Malmkjaer, Kirsten (ed.), 111-116.
Volkovitch, Michel. 1998. “Une idée neuve en Europe”. In Fortunato, Israel (ed.), 235-
244.
Waddington, Christopher. 2000. Estudio comparativo de diferentes métodos de
evaluación de traducción general (inglés-español). Madrid: Publicaciones de
la Universidad Pontificia Comillas.
Walters, Joel. 2005. Bilingualism: The sociopragmatic – psycholinguistic interface.
London: Lawrence Erlbaum Associates.
Waterhouse, Lynn. 2006. “Multiple intelligences, the Mozart effect and emotional
intelligence: a critical review”. Educational Psychologist 41,4: 207-225.
Wei, Li (ed.). 2000. The bilingualism reader. London: Routledge.
Weinreich, Uriel. 1953. Languages in contact: Findings and problems. The Hague:
Mouton.
Whorf, Benjamin. 1956. Language, thought, reality. New York: Wiley.
Whyatt, Bogusława. 2000. A psycholinguistic investigation into the processes of
comprehension and production: a decision making approach towards the
preservation of meaning in translation. Unpublished doctoral dissertation.
Poznań. Adam Mickiewicz University.
Whyatt, Bogusława. 2003. “Reading for translation: investigating the process of foreign
language text comprehension from an information processing perspective”.
Studia Anglica Posnaniensia, 39: 133-147.
Whyatt, Bogusława. 2006a. “Psycholingwistyczne spojrzenie na użycie słowników w
procesie tłumaczenia tekstu z języka obcego (angielskiego) na język rodzimy
(polski)”. Neofilolog 28: 70-77.
 References 437

Whyatt, Bogusława. 2006b. “Discovering the bilingual mind – some evidence for the
reorganization of the mental lexicon in translation students”. In: Dziubalska-
Kołaczyk, Katarzyna (ed.), 729-747.
Whyatt, Bogusława. 2006c. “The essence of translation, Fillmore’s frames and scenes
and the communicative quality of Polish-English menu translations”. In:
Fisiak, Jacek (ed.), 101-111.
Whyatt, Bogusława. 2007a. “Meaning as a dynamic concept. A Think Aloud Protocol
study into strategies performed by inexperienced translators” In: Fabiszak,
Małgorzata (ed.), 329-341.
Whyatt, Bogusława. 2007b. “Two Languages, two cultures, one mind: a study into
developmental changes in the students’ view of language as a tool in cross-
cultural communication”. Poznań Studies in Contemporary Linguistics 43,2:
133-145.
Whyatt, Bogusława. 2008. “Bilingual knowledge and interlingual skills: A discussion of
the benefits of translation tasks for second language learners”. In: Pawlak,
Mirosław (ed.), 197-210.
Whyatt, Boguslawa. 2009a. “Building L2 communicative confidence through
interlingual tasks; towards function-focused L2 learning”. In: Dynel, Marta
(ed.), 365-388.
Whyatt, Bogusława. 2009b. “Translating as a way of improving language control in the
mind of an L2 learner: assets, requirements and challenges of translation
tasks”. In: Witte, Arnd – Theo Harden – Alessandra Ramos de Oliveira
Harden (eds.), 181-202.
Whyatt, Bogusława. 2009c. “The interlingual individual”. Paper presented at the 7th
Symposium on Bilingualism. Utrecht.
Whyatt, Bogusława. 2010. “Bilingual language control in translation tasks. A TAP study
into mental effort management by inexperienced translators”. In: Arabski,
Janusz – Adam Wojtaszek (eds.), 79-92.
Whyatt, Bogusława. 2011. “Review of: A Career in Language Translation by
Férailleur-Dumoulin’s. The Interpreter and Translator Trainer 5,2: 366-371.
Wierzbicka, Anna. 1991. Cross-cultural pragmatics: the semantics of human
interaction. Oxford: Oxford University Press.
Wilss, Wolfram. 1976. “Perspectives and limitations of a didactic framework for the
teaching of translation”. In: Brislin, Richard W. (ed.), 117-137.
Wilss, Wolfram. 1982. The science of translation. Problems and methods. Tübingen:
Gunter Narr.
Wilss, Wolfram. 1988. Kognition und Übersetzen. Tübingen: Niemeyer.
Wilss, Wolfram. 1996. Knowledge and skills in translator behaviour. Amsterdam: John
Benjamins.
Wilss, Wolfram. 1997. “Translator awareness”. Revista Ilha do Desterro 33: 89-100.
Wilss, Wolfram. 1999. Translation and interpreting in the 20th century. Focus on
German. Amsterdam: John Benjamins.
Wilss, Wolfram – Gisela Thome (eds.). 1984. Die Theorie des Übersetzens und ihr
Aufschlusswert für dir Übersetzungs- und Dolmetschdidaktik. Tübingen: Narr.
438 References

Witczak-Plisiecka, Iwona (ed.). 2010. Pragmatic perspectives on language and

linguistics; Vol.1: Speech actions in theory and applied studies. Newcastle:
Cambridge Scholars Publishing.
Witte, Arnd. 2009. “From translating to translation in foreign language learning”. In:
Witte, Arnd – Theo Harden – Alessandra Ramos de Oliveira Harden (eds.),
79-97.
Witte, Arnd – Theo Harden – Alessandra Ramos de Oliveira Harden (eds.). 2009.
Translation in second language learning and teaching. Oxford: Peter Lang.
Wright, Richard D. – Lawrence M. Ward. 2008. Orienting of attention. New York:
Oxford University Press.
Zampolli, Antonio (ed.). 1977. Linguistic structures processing. Amsterdam: North-
Holland.
Zawada, Anna. 2009. Modelling personality profile of a good translator. A
questionnaire study of foreign language learners, translation trainees and
professional translators. Unpublished BA thesis. Poznań. Adam Mickiewicz
University.
Zojer, Heidi. 2009. “The methodological potential of translation in second language
acquisition: Re-evaluating translation as a teaching tool”. In: Witte, Arnd –
Theo Harden – Alessandra Ramos de Oliveira Harden (eds.), 31-51.
Appendix 1

ANOVA

Sum of Mean
Squares df Square F Sig.
Between Groups 287 892 2 143 946 0,368 0,696
Duration of the translation
Within Groups 8 213 950 21 391 140
task
Total 8 501 843 23
Between Groups 6 491 2 3 246 1,102 0,351
Time prior to the onset of
Within Groups 61 856 21 2 946
translation
Total 68 347 23
Between Groups 82 054 2 41 027 0,589 0,564
Time for post-draft
Within Groups 1 462 747 21 69 655
revision
Total 1 544 801 23
Between Groups 3 089 2 1 545 0,334 0,720
Production during
Within Groups 97 119 21 4 625
post-revision
Total 100 208 23
Between Groups 2 2 1 0,045 0,956
Consulting external
Within Groups 524 21 25
sources
Total 527 23
Between Groups 6 500 2 3 250 0,465 0,634
Text elimination Within Groups 146 632 21 6 982
Total 153 132 23
Between Groups 34 010 2 17 005 0,180 0,837
Cursor and mouse
Within Groups 1 988 712 21 94 701
movement
Total 2 022 722 23

Graph 17. Significant test – no differences between the groups.

440 Appendix 1

Descriptive Statistics

Std. S.E.
N Mean
deviation mean
Total User Events KIN/ NO no KIN 24 1 701,21 449,910 91,837
(TUE) KIN KIN 24 1 656,71 399,848 81,619
KIN/ NO no KIN 24 1 281,33 95,982 19,592
Text production (TP)
KIN KIN 24 1 290,63 145,211 29,641
KIN/ NO no KIN 24 145,54 92,483 18,878
Text elimination (TE)
KIN KIN 24 134,17 72,598 14,819
Cursor navigation KIN/ NO no KIN 24 211,83 394,614 80,550
(CN) KIN KIN 24 187,17 241,972 49,392
KIN/ NO no KIN 24 62,29 48,881 9,978
Mouse Event (ME)
KIN KIN 24 44,46 32,011 6,534
Navigation in text KIN/ NO no KIN 24 274,13 384,743 78,535
(CN + ME) KIN KIN 24 231,63 243,320 49,667
KIN/ NO no KIN 24 12,38 8,032 1,639
Dictionary Use (DU)
KIN KIN 24 13,13 5,841 1,192
KIN/ NO no KIN 24 32:17:23 12:02:32 02:27:29
Duration (Dur)
KIN KIN 24 35:14:55 10:40:54 02:10:49
Events per minute KIN/ NO no KIN 24 58,8888 22,52180 4,59724
(EPM) KIN KIN 24 49,9367 14,42069 2,94361
Text Production per KIN/ NO no KIN 24 44,6613 14,43733 2,94701
minute (TPM) KIN KIN 24 39,2254 10,04288 2,04999

Graph 18. Descriptive statistics for both groups.

 Appendix 1 441

Significant correlations

Text Navigation
Duration Production Dictionary in text
(Dur) per minute Use (DU) (CN +
(TPM) ME)
Pearson
-,929(**)
Text Production Correlation
per minute (TPM) Sig. (2-tailed) 0,000
N 48
Pearson
,581(**) -,535(**)
Dictionary Use Correlation
(DU) Sig. (2-tailed) 0,000 0,000
N 48 48
Pearson
0,021 -0,029 -0,075
Navigation in text Correlation
(CN + ME) Sig. (2-tailed) 0,888 0,847 0,614
N 48 48 48
Pearson
0,189 -0,029 0,006 ,371(**)
Text elimination Correlation
(TE) Sig. (2-tailed) 0,199 0,845 0,970 0,009
N 48 48 48 48
. **. Correlation is significant at the 0.01 level (2-tailed).

Graph 19. Significant correlations between text production features.

Appendix 2

Study 1
SL Text
„Tatarak”

Film oparty na opowiadaniu Jarosława Iwaszkiewicza, wydaje się być

subtelną i wzruszającą historią o niemożliwej miłości. Jednak Andrzej
Wajda idzie dalej, tworząc wielowymiarową opowieść o miłości, która
przychodzi za późno i o śmierci, która zawsze przychodzi za wcześnie.
Marta (Krystyna Janda) jest żoną doktora (Jan Englert) z małego
miasteczka. Pewnego dnia Marta poznaje znacznie młodszego od siebie
mężczyznę, prostego robotnika Bogusia (Paweł Szajda), który oczarowuje
ją swoją młodością i niewinnością. Ich spotkania na brzegu rzeki
porośniętej tatarakiem są naznaczone wzajemną fascynacją. Jedno z nich
zmierza do przedwczesnego końca, drugie zaś wkracza dopiero w
dojrzałość jednak los obchodzi się z nimi okrutnie…

Tatarak to również film o tworzeniu filmu, a główną bohaterką jest nie

tylko fikcyjna Marta, ale i aktorka, która ją gra. Andrzej Wajda wplótł w
opowiadanie Iwaszkiewicza autentyczne monologi Krystyny Jandy na
temat przedwczesnej śmierci jej męża, cenionego operatora Edwarda
Kłosińskiego. W ten sposób obie kobiety – Krystyna i Marta – zlewają się
w jedną głęboko zranioną istotę, która musi znaleźć w sobie siłę i
poradzić sobie z perspektywą nieuniknionej śmierci.

Source: Advertising leaflet available in Polish cinemas

Randomly chosen samples of translations with their Translog files and

statistics
444 Appendix I1

S7 G1

“Sweet flag”

The film based on the story of Jarosław Iwaszkiewicz seems to be a subtle

and touching history of an impossible love. However, Andrzej Wajda goes
further, creating a mutlidimensional story about love which comes late
and death which always comes too early. Marta (Krystyna Janda) is the
wife of a doctor (Jan Englert) from a small town. One day, Marta meets a
man named Boguś, much younger than she, who is a simple worker
charming her with his youth ind innocence. Their meetings on the sweet
flag- overgrown river bank are full of fascination towards each other. One
of them is heading for a precocious end of life and the other one has just
entered the world of maturity, but the fate is cruel with them...

“Sweet flag” is also a film about making a film, and the main character is
not only fictional Marta but also the actress playing her. Andrzej Wajda
included in the film the authentic monologues by Krystyna Janda talking
about her precociously descended husband, a respected cameraman
Edward Kłosiński. By this, both women- Krystyna and Marta- are
merging together into one deeply harmed individual, who must find a
strenght in herself and cope with the pespective of the inescapable death.

[Start][ 49.875]" Tar⌫tarak" The film based on t

he story of Jaros łam Iwaszkiewicz seems t
o be [ 12.805]subtle and [ 18.113]history ⌫ ⌫⌫
ry of an impossible love . However, An
drzej Wajda goes fut ⌫rther , creating a
mutlidimensional story about love which com
es late and death which always comes too early. Ma
rta ( Krystyna Ja nda) is the doc
tor ’s wife ⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫ wife of the d
octor ( Jak⌫n E nglert) from a small city ⌫⌫⌫⌫
village ⌫⌫⌫⌫⌫⌫⌫town. O ne day, Marta meets
a ⌫ , much younger than she ⌫ , [ 26.293][ ] man
is [ 13.949]
⌫⌫⌫ a simple worker
 Appendix II 445

whose ⌫⌫ is [ 19.569][
] named Bogus [ ] [ ]⌫ś [ ] charmi
ng her ⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫ charmi ng her wit
h his youth ind innocence . Their meetre⌫⌫i
ngs on the tatarak- river bank [ ]
⌫⌫⌫⌫⌫⌫⌫[ 46.434]sweet flag[ 17.133] overgrown
[ ]⌫⌫⌫⌫⌫⌫⌫ Sweet flag [ ] [ ] B
[ 02:18.496][ ] h ⌫ [ ] [ ] I [ ]
are full of [ 18.213]fascination towards e
ach other. O ne of them [ 46.600]is headin
g towards the [ 20.578]⌫⌫⌫⌫a preco[ ] [ ]cio
us end ⌫ of life[ 01:10.052][ ] ⌫⌫
⌫⌫⌫⌫⌫for [ ] and the other one is [ 26.713]
just enteri ⌫ ⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫ has just en
tered the mature life ⌫⌫⌫⌫⌫⌫it u ⌫y[ 37.493]
[ ] world of [ ] . ⌫ , but t he fa
te is [ 29.613]cruel for them....⌫ [ 10.318]"Th e
⌫⌫⌫ Sweet flag" is also a film about making a film
, a nd the main chracter ⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫
⌫⌫⌫⌫⌫ ⌫⌫⌫ fictional Martha ⌫⌫⌫⌫⌫⌫
⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫ the role of the main chra⌫⌫⌫ hara
cter ⌫⌫⌫⌫
⌫⌫⌫⌫⌫⌫⌫⌫ [ ] is not only t⌫fictional Marta but a
lso the actress playing her. Andrzej Wajda
[ 17.411]interlaced the film with[ 03:26.137]⌫⌫⌫⌫⌫⌫
⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫ included in the film
the authentic monologues [ ] B M [ ] by Krystyn
a Janda talking about her pre cio ⌫⌫ o ciousl
y dead ⌫⌫⌫escended husband. ⌫, t⌫
a [ 11.378]respected operator[ 14.061]⌫⌫⌫⌫⌫⌫⌫
⌫cameraman Edward Kło siński . By this-
⌫⌫ both women - Krystyna and Marta - are
⌫⌫⌫[ 26.355]are merging together into one deepl
y harmed indiwidual[ 16.644], whi⌫o must find
a strenght in herself and [ 02:23.102]cope M [ ]
with the pespective of the [ 15.250]inescapa ble de
ath. [ ] B [ ]⌫w [ ] a [ ][ 12.773]B
446 Appendix I1

touh⌫ching [ ]and [ ]⌫⌫⌫⌫[ 33.234][ ]⌫⌫⌫a

[ 17.193][ ] ⌫⌫⌫ her [ ] [ ] [ ][ 22.003]
[ ] ⌫⌫⌫ she[ 29.152][ ]⌫⌫⌫ with[ 20.035][ ]
, [ ] B M [ ] ⌫v[ 33.699][ ] M [ ] M
[ ] [Stop]

Subject: S7 G1
Total user events 1853
Text production 1457
Text elimination 220
Cursor navigation 93
Mouse events 83
Dictionary consultations 5 bilingual, 5 monolingual
Internet look ups
Duration 36:57.039
User events per minute 50.15
Text production per minute 39.43

S8 G2

“Sweet flag”

The film, based upon Jarosław Iwaszkiewicz’s short story, seems to be

nothing more than an impossible moving love story. However, Andrzej
Wajda, the director, moves further in order to create a multidimensional
story about a late-in-life love and a premature death. Marta (Krystyna
Janda) is the wife of a small town doctor (Jan Englert). One day Marta
meets Boguś, a simple worker who is far younger than her. She is
charmed by his innocence and youth. They meet on a river bank covered
with sweet flag where their mutual fascination grows. One of them is
approaching a premature end, the other is stepping into maturity but a
cruel fate awaits them...

The Sweet flag also tells a story of film-making in which the main hero is
not only a ficticious character called Marta but also the actress who acts
her role. Andrzej Wajda interweaves Krystyna Janda’s true monologues
on the premature death of her husband, a renowned cameraman Edward
 Appendix II 447

Kłosiński. Thus, both women, Krystyna and Marta, mix into one deeply
hurt person who needs to find strength in order to face the gloom of
inevitable death.

[Start][ 26.692]b[ 22.089]i[ 48.624] "Sweet flag"

The film based upon Jarosław Iwaszkiewicz’s short
story [ 10.121]seems to be a moving impossible love
stroty. ⌫⌫⌫or Hi⌫owever, [ 10.130]An drze
j Wajda , the director, goes even further
as to create a multide⌫imen sional lov
es story which ⌫⌫⌫⌫⌫⌫ [Ctrl ] ⌫[Ctrl
] [Ctrl ] late t[ 42.024]i[ 40.
507] ⌫
⌫⌫⌫⌫[Ctrl ][Ctrl ][Ctrl ][Ctrl ]late-in ⌫ -life
and a [ 11.347]premature death. [ 10.981]Marta
(Krystyna Janda) is doctor’s wife[Ctrl ] ( Jan
Engl ert) .[Ctrl ][Ctrl ][Ctrl ][Ctrl ][
Ctrl ][Ctrl ][Ctrl ] a small town One day Mar
ta meets a man a lot yp⌫ounger than herself
[Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ]who is ⌫
⌫⌫⌫⌫ [Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctr
l ][Ctrl ][Ctrl ][Ctrl ] Bogu ś ⌫ , ⌦⌦
⌦worker [Ctrl ][Ctrl ][ShftCtrl ][ShftCtrl ]⌦far
i[ 56.281][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ]
[Ctrl ][Ctrl ]simple [ 10.530]

, that ⌦ charmes her with

his innocence and youth. [ 10.748]They meet o
n a river bank covered with swee t flag[ 11.963] and th
ey are fascinated about each other, ⌫⌫.
One is heading ⌫⌫⌫⌫⌫⌫⌫⌫approaching as ⌫
⌫ premature end, the other is stepping into matur
i ty but the fate is cruel to the,m⌫⌫m...
The Sweet flag i⌫ tells also a story of movi
e⌫⌫⌫⌫⌫film–making in which the ma in hero is
not o nly a ficticious character calledc⌫ Mart
a but also the actress who [ 30.180]plays her.
448 Appendix I1

b[ 26.215]m[ 52.113][Ctrl ][Ctrl ][ShftCtrl ][ShftCtrl

]⌦acts her role Andrzej Wajda[ 14.545] b[ 48.971]
interweaves [ 10.125]true stories ⌫⌫⌫⌫⌫⌫⌫⌫ K
rystyna Janda’s monolgues on g⌫her husband’s
premature death [ 10.061][Ctrl ][Ctrl ][Ctrl ][Ctrl ]
[ShftCtrl ][ShftCtrl ] the of [Paste]⌫⌫⌫,
b[ 38.874] a renowned cameraman Dward Kłosiński. In th
e⌫is way both women,.⌫ Krystyna and Marta, mi
xed to create one deeply hut⌫rt person who needs to
find strent ⌫gth and face the
gloom of inevitable death .
is ⌫⌫⌫⌫,

, nothing more than

n [ShftCtrl ] [Ctrl ] [Paste]

[ShftCtrl ]⌦moved⌫s ⌦⌦
[ShftCt
rl ]⌦⌦ ⌦⌦in order ⌦ la

story about [Ctrl ][

Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ][ShftCtrl ]⌦[ 12.493
][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ]a [ 10.500]

⌫the wio⌫i⌫fe od⌫f a[Ctrl ][Ctrl ][Ctrl ][C

trl ]⌫⌫⌫ [Ctrl ][Ctrl ][Ctrl ][Ctrl ][ShftCtrl ]⌦⌫
i[ 13.439]m[ 31.858] [ 19.199][ ]

. He ⌦⌦⌦⌦⌦⌦[ 12.036]⌫⌫ She

is [Ctrl ][Ctrl ][Ctrl ] ⌫d by his [ShftCtrl ][Shf
tCtrl ][ShftCtrl ][ShftCtrl ]⌦ [ 22.125]

where their [ShftCtrl ][Sh

ftCtrl ][ShftCtrl ]⌦ [Ctrl ] ⌫⌫ion ⌦
[Ctrl ][Ctrl ][Ctrl ] g [Ctrl ][ShftCtrl ][Shft
 Appendix II 449

Ctrl ][ShftCtrl ]⌦⌦[Ctrl ][Ctrl ][Ctrl ]mutual [Ctrl

][Ctrl ] [Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ]
[Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl
][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ] g
rowa⌫s [Ctrl ][Ctrl ]of them [ 13.473]i [ 01:03.411]
[ ] y musr⌫t [ 10.278]i⌫m [ 51.461][Ctrl ][ShftCtrl ]
⌦awai ⌫⌫⌫⌫⌫⌫⌫ ⌫⌫a cruel fate awaits the m
[Ctrl ][ShftCtrl ][ShftCtrl ][ShftCtrl ][ShftCtrl ][Shft
Ctrl ]⌦ i[ 20.560][ ]⌫⌫⌫⌫⌫[Ctrl ] also [ 30.
516][ ]⌫⌫⌫⌫⌫ [Ctrl ][Ctrl ][Ctrl ][Ctrl ]true
[ ]o [ ]⌫Ed [ ]⌦Thus,
[ ]=⌫⌫⌫ int⌫ \⌦⌫⌦⌦⌦⌦⌦⌦[ 13.209][ ]⌫
⌦⌦in order tp⌫o[ 23.856][Stop]

Subject: S8 G2
Total user events 2271
Text production 1391
Text elimination 149
Cursor navigation 709
Mouse events 18
Dictionary consultations 4 bilingual, 3 monolingual
Internet look ups 5
Duration 31:24.057
User events per minute 72.32
Text production per minute 44.30

S1 G3

“Sweet Flag”

This screening of a short novel by Jarosław Iwaszkiewicz seems to be a

subtle and yet moving story of an impossible love. But Andrzej Wajda,
the director, had decided to make it a multi-faceted story about love,
which comes too late, and about death, which comes prematurely. Marta
(starring Krystyna Janda) is the wife of a doctor (starring Jan Englert)
from a small town. One day Marta meets a much younger man, a simple
worker Buguś (Paweł Szajda), who enchants her with his youth and
450 Appendix I1

innocence. Their meeting at the bank of a river grown with sweek flag is
mutually facinating. One of them is on the road to a premature end, the
other is just entering maturity. But their future together is very painful...

“Sweet Flag” is also a film about film making. The main heroine is
featured in two roles - that of Marta and that of an actress, whom she
plays. Andrzej Wajda used Krystyna Janda’ genuine words about the
premautre death of her husband, Edward Kłosiński, a much appreciated
film operator, and made them part of his film. In this way both women,
Krystyna and Marta, bond into one deeply hurt being, which must find
strength in itself and cope with the prospect of unavoidable death.

[Start] [ ][ 15.627]I[ 47.668]" Sweet Flat⌫g"

The film ⌫⌫⌫⌫⌫⌫⌫is is a screening of a novel
by⌫⌫⌫⌫⌫⌫⌫⌫short novel by Jarosław Iwaszkiewicz .
[Ctrl⌦][Ctrl⌦] s⌫⌫⌫ seems to be a subtl
e and yet moving history[Ctrl⌫]story of an impossible lov
e. But Andrzej Wajda, the director, goes ⌫[Ctrl⌫]takes
a step forward [Ctrl⌫][Ctrl⌫][Ctrl⌫][Ctrl⌫] had decide
d to make it a multi–faceted story about love, which comes
too late a⌫⌫, and about death , which comes t⌫premature
ly. Marta (Kr⌫⌫starring Krystyna Janda) is the wife of
a doctor (starring Jan Englert) from a small town. One day
Marta meets a much younger man , a simple worker Buguś (
sta⌫⌫⌫Paeł⌫⌫we S⌫⌫ł Szajda_⌫), who enchants her
with his youth and innocence. Their meei⌫ting at the
bank of a river grown with se⌫week flag is mutually fa
cinating. One of them [ 13.236]is on the road to pre⌫⌫
⌫a premature end, the other ju⌫⌫is just entering t⌫ma
turity . But their future together is ver
y painful... T⌫"Sweet Flat" ⌫⌫⌫g" is also a fil
m about film making. The main heroine p⌫ is featured
in two roles - Marta[Ctrl⌫]that of Marta and that of an
actree⌫ss, whom she st ⌫⌫po⌫lays. Andrzej Wajda in
corporated a⌫genuine dia logu[Ctrl⌫] words [Ct
rl ][Ctrl ]Krystyna Janda’ w⌫about [Ctrl ][Ctr
l ][Ctrl ][Ctrl ][Ctrl ][Ctrl ][Ctrl ] [Ctrl⌦] i⌫us
 Appendix II 451

ed the premautre death of her husband, Edward Kłosińs

ki, a much appreciated film operator , and made them
part of his film. In this way both women, Krysyt⌫⌫tyna an
d Marta, bond into one deeply hurt being , which must find
strength in itself and cope with the prospect of i
⌫ uh⌫navoidable death. [Stop]

Subject: S1 G3
Total user events 1413
Text production 1325
Text elimination 68
Cursor navigation 18
Mouse events 2
Dictionary consultations 0
Internet look ups 1
Duration 07:49.390
User events per minute 180.62
Text production per minute 169.37
Appendix 3

Study 2
SL text
„Legenda Zorro”

W kontunuacji „Maski Zorro”, barwnego przeboju sprzed dziewięciu lat,

Alejandro de la Vega (Banderas) wiedzie poukładane życie statecznego
obywatela, męża pięknej Eleny (Zeta-Jones) i ojca 10-letniego Joaquina
(Alonso). Gdy jednak zachodzi potrzeba, przystojny szlachcic zakłada
czarną maskę, płaszcz, wsiada na konia i pędzi ratować świat. Ten zresztą
potrzebuje go nieustannie, co mocno irytuje donnę de la Vega. Przez
dziesięć lat znosiła niecodzienne hobby małżonka, teraz mówi "basta!" i
każe bohaterowi wybierać między zabawą w obrońcę uciśnionych a
rodziną. Na tym kłopotów Zorro nie koniec, bo oto tajne stowarzyszenie
Rycerzy Aragonii chce zapobiec przyłączeniu Californii do USA, a z
zagranicznych wojaży wraca jego dawny wróg hrabia Armand (Sewell),
który ostrzy sobie zęby na Elenę. Efektowne galopady, machanie
szabelkami i fruwanie na linach zachwycą małych chłopców, a uroda
Zeta-Jones tych większych.

Source: TV magazine
Randomly chosen samples of translations with their Translog files and
statistics

S16 KIN group

The legend of Zorro

In a sequel of “The Mask of Zorro”, a colorfull hit from the previous nine
years, Alejandro de la Vega (Banseras) leads a well-composed life of a
sedate citizen, husband of a beautiful Elena (Zeta-Jones) and a father of a
10-year-old Joaquin (Alonso). When the need occurs, the handsome
nobleman puts on a black mask, cloak, mounts the horse and hurries to
save the world. After all, this needs him incessantly, what strongly
454 Appendix III

irritates donna de la Vega. For ten years, she has tolerated the unusual
hobby of her hubby, and now she says: “enough!” and tells the hero to
chose between a game in fighting for the opressed and family. This is no
end to Zorro’s troubles, because now a secret assosiation of the Aragon
Knights wants to prevent the incorporation of California to the US, and
count Armand (Sewell), his old-time enemy, who whets his appetite for
Elena, comes back from the overseas voyages. Spectacular gallopades,
brandishing sabers and flying on robes will fascinate little boys, and the
charm of Zeta-Jones will do the same for the big ones.

[Start][ 08:57.090][ ] The legend of Zorro In a [ 1

5.865]se quel of "The m⌫Ma sk of Zorro" ⌫, [
21.739]a colorfull sto ⌫⌫⌫[ 12.219]hit o ⌫ fr
om the previous nine years , Ale jandro de la Ve
ga ( Banseras) leads a [ 43.302]life of a
static citizen[ 19.750][ ] well–composed ⌫ [ ]
[ 18.003][ ]⌫⌫edate[ ] , husband o f a beu ⌫auti
ful He⌫⌫Elen a (z⌫Zeta ⌫-Jo nes) and a father of
a 10-year-old [ ] Jo aquin (alonso)
⌫A [ ] . When the [ 24.106]need [ 17.924]occurs , th
e handsome [ 20.742]nobleman puts on a black ma
sk ⌫, his ⌫⌫⌫⌫[ 20.396]cloak nad ⌫⌫⌫⌫
⌫ , [ 12.670]mounts the ho rse and [ 22.832]hurrie
s to save thge ⌫⌫e worls. ⌫⌫⌫d. [ 23.373]After all,
this needs him un[ 18.291][ ]incessantly , what s
⌫strongly irritates donn a th⌫⌫de la Vega .
For ten years , she has [ 34.502]tolerated [ 16.5
67]u⌫t he un usual hobby o h⌫f her hubby. ⌫. now
⌫⌫⌫⌫⌫⌫, and noe⌫w she sa ⌫⌫[ 14.373]says: "E
no⌫⌫⌫enough! " and [ 20.111]tell the [ ]
s[ 12.306][ ]hero to chose between fun ⌫⌫⌫
and family. [ ][ ][ 55.787]a game in fi
ghting for the opressed [ ] T⌫[
30.498]This in⌫s no end to t he [ ] [ ] [ ]of
the Zorro’s troubles , bev⌫cause noe⌫w the
sec [ ]a ⌫ ret [ 19.588][ ]s⌫assosiation of
t⌫the [ 22.920]Aragon Kni ghts wants
to prevent ⌫⌫⌫⌫⌫⌫⌫⌫[ 23.730]prevent the e⌫i
 Appendix III 455

ncorporation [ 14.218][ ]⌫⌫ ⌫⌫to[ 10.984][ ]⌫ [

] of California rto⌫⌫⌫to the US , a ⌫nd [
15.373]count Armand ( Sewell) comes ba ck from the [
54.910]overseas voyages [ ], his old ⌫-time
enemy, [ ] who [ 30.362]whet s his appetit
e for el⌫⌫Elena, [ ] . [ 17.738]Spectat⌫c
ular [ 20.781]gallopades ⌫, [ 55.649]brandishing sab ers
and [ 36.205]flying on robes [ 23.197]will fa
scinate little boys, and the beu ⌫⌫⌫ charm of Zeta-
Jones will⌫⌫⌫⌫will do thge d⌫same⌫⌫⌫⌫⌫⌫e⌫⌫e
same for the big ones. [Stop]

Subject: S16 KIN group

Total user events 1349
Text production 1189
Text elimination 97
Cursor navigation 15
Mouse events 48
Dictionary consultations 5 bilingual
Internet look ups 0
Duration 33:02.941
User events per minute 40.82
Text production per minute 35.98

S12 NO KIN group

The Legend of Zorro,

In the continuation of “The Mask of Zorro”, a colourful blockbuster from

before nine years, Alejandro de la Vega (Banderas) leads a peaceful life of
a sedate citizen, a husband of beautiful Elena (Zeta-Jones) and a father of
a 10-years old Joaquin (Alonso). But whenever there is a need, the
handsome nobleman puts the black mask and his cloak on, mounts his
horse and and dashes to save the world. The latter needs him all the time
in any case, which immensly iritates donna de la Vega. She has been
standing her husband’s unusual hobby for ten years, and now she says
‘enough!’ and tells the heroe to choose between playing a guardian of the
oppressed and his family. It is only the beginning of Zorro’s troubles, as a
456 Appendix III

secret guild of The Knights of Aragon wants to prevent the incorporation

of California to the USA, whereas his arch-enemy count Armand (Sewall)
returns from his abroad voyages, and who has his eye on Elena. Effective
galopades, sword fights and swinging on ropes will ravish the little boys,
whereas beauty of Zeta-Jones the bigger ones.

[Start] The Le gend of Zorro, In the continuatio n of

"The Mask of Zorro ", a clo ⌫⌫olour fu l blockbuster f
ro m before nine years, A lejandro de la Vega ( Band
feras⌫⌫⌫⌫⌫eras ) leads qa ⌫⌫⌫a peacefu l life
a ⌫⌫ of a [ 20.730](BL ) sedate citizen , a
husband of the beas⌫⌫autifu l Elena (
⌫⌫⌫⌫
ZEta⌫⌫⌫eta_⌫_⌫-Jones ) and a father of a 10-
years old Jo a qui n ( al⌫⌫Alonso ) . [ 12.629]But
when [ 38.664]⌫ever thert⌫e ia⌫s a need , the han
dsome noble man [ 21.273]puts the black am⌫⌫mm⌫ask o
n ⌫⌫ and g⌫his [ 20.750](BL) cloak on ,
,ounts ⌫⌫⌫⌫⌫⌫⌫mounts his horse a nd [ 23.02
8]<B:⌫⌫⌫<⌫( BL) and dashes to save the day.
⌫⌫⌫⌫⌫worls⌫d. [ 22.237]T he latter nee ds
him all the time in any case
(BL) , which iritates
immensly donna de la VEG
⌫⌫ega . Fo r t⌫⌫⌫⌫⌫She has been st
anding her hus bands⌫’s unu sual hobby
for ten years , and now she says " en⌫⌫⌫’ e
nough!’ and teel ⌫⌫lls the heros⌫e to c
hoose betwee n[ 14.430] [ 59.453]t he ga ⌫⌫⌫⌫⌫⌫ pl
aying a guardia n of [ 18.152](BL) ippress
e⌫⌫⌫⌫⌫⌫⌫⌫oppe⌫ressed
thge⌫⌫e and his family .
B ut⌫⌫⌫It is only the beginni ng of Zorro’s
t roubles , as a [ 26.511](BL) a⌫⌫ guild of
The Knight s of Ara go n[ 13.373]

secret
wants to prevent the incorporation of
 Appendix III 457

California to the USA , whereas his arc

henemy count Armand (Sewa ll) returns from
his abroad voyaga ⌫es⌫⌫es , and who [
01:16.363]whert⌫⌫tsw ⌫⌫ his appetite for ⌫⌫⌫⌫⌫⌫
⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫⌫[ 01:37.651](BL) has his
wye ⌫⌫⌫⌫eye on Elena . Effective [ 19.482]
ga l opades , sword fih⌫ghts and s winging
on ropes will [ 24.070]ravis h t he lii⌫tt
le boys , ahereas⌫⌫⌫⌫⌫⌫⌫whereas
<⌫(BL)
beauty of
Zeta -Jones the bigger o ne s.[ 11.300][Stop]

Subject: S12 NO KIN group

Total user events 1631
Text production 1251
Text elimination 130
Cursor navigation 250
Mouse events 0
Dictionary consultations 8 bilingual
Internet look ups 0
Duration 21:33.675
User events per minute 75.64
Text production per minute 58.02
Przekład jako umiejętność człowieka

From predisposition to expertise

(Streszczenie)

Książka podejmuje złożoną tematykę przekładu jako umiejętności

człowieka. Z jednej strony panuje powszechne przekonanie, że tłu-
maczenie z jednego języka na drugi nie wymaga szczególnych
umiejętności poza znajomością dwóch języków. Z drugiej strony często
słyszy się opinie, że tłumaczenia powszechnie dostępne są słabej jakości.
Najwyraźniej mamy do czynienia z całym spektrum czynników, które
determinują ludzką umiejętność dokonywania tłumaczeń z jednego języka
na drugi. Aby uwzględnić czynniki wpływające na to jak dana osoba
wykonuje tłumaczenie autorka przedstawia tłumaczenie jako umiejętność,
która obejmuje ewolucję od danej człowiekowi predyspozycji do
interpretacji znaczenia i przenoszenia odczytanego sensu w inne formy
językowe (lub pozajęzykowe) do kompetencji przypisywanych ekspertom
w sztuce przekładu. Dokonując wnikliwej analizy obecnego stanu wiedzy
autorka wysuwa hipotezę, iż rozwój umiejętności dokonywania przekładu
do poziomu kompetencji i ekspertyzy opiera się na odpowiedniej
integracji wiedzy językowej i konceptualnej. Tłumacz, aby przenieść sens
z tekstu w języku źródłowym na tekst w języku docelowym uczy się
budować Sieć Integracji Wiedzy (Knowledge Integration Network)
koniecznej do wykonania danego tłumaczenia.
Książka składa się z dwóch zasadniczych części. Rozdziały 1-5 to
teoretyczne rozważania nad ewolucją i rozwojem umiejętności
dokonywania przekładu. Rozdział 6 i 7 to empiryczna część relacjonująca
przeprowadzone przez autorkę badania nad rozwojem umiejętności
dokonywania przekładu z uwzględnieniem hipotezy Sieci Integracji
Wiedzy.
Rozdział pierwszy wprowadza czytelnika w tematykę przekładu jako
umiejętności. Autorka wychodzi z założenia, że tłumaczenie jest
czynnością powszechną dla użytkowników nawet tylko jednego języka.
Komunikacja, to przecież ciągłe wyrażanie sensu w różnych kodach.
Użycie języka, to odpowiednie zakodowanie treści w formę, która jest
460 Streszczenie

odczytywana i interpretowana i ukazuje ową treść odbiorcy przekazu.

W pewnym sensie zatem wszyscy jesteśmy tłumaczami. Powszechne jest
przekonanie, że każdy, kto zna dwa języki potrafi również tłumaczyć. Nie
każdy natomiast będzie w stanie wykonać nawet dość proste tłumaczenie
pisemne nie popełniając często zasadniczych błędów językowych bądź
też znaczeniowych. Tak jak w przypadku wielu złożonych umiejętności,
umiejętność tłumaczenia nie rozwinie się samoistnie nawet jeżeli nasza
kompetencja językowa w dwóch językach jest dobra. Słusznym zatem
wydaje się przeanalizowanie ewolucyjnego spektrum, które prowadzi od
predyspozycji do ekspertyzy w sztuce przekładu.
Rozdział drugi analizuje znajomość dwóch języków jako podstawę, na
której można zbudować kompetencje tłumacza. Autorka stawia
podstawowe pytania: Co to znaczy znać drugi język? Jak ludzki umysł
rozróżnia pomiędzy językami? Jak wygląda struktura pamięci, w której
muszą pomieścić się różne słowa (w języku pierwszym i drugim) na
określenie tego samego pojęcia? Co dzieje się w sytuacji kiedy nie tylko
słowa, ale i pojęcia są rożne, albo tylko trochę podobne? Bazując na
obecnym stanie wiedzy z zakresu psycholingwistyki, bilingwalizmu
i psychologii kognitywnej coraz mniej jest wątpliwości, że dwa języki w
jednym umyśle są w stanie ciągłej, dynamicznej interakcji. Użytkownik
dwóch języków, a zwłaszcza tłumacz musi wypracować odpowiednie
mechanizmy kontroli, aby nie mylić form lub treści przypisanych swoim
dwóm językom. Taki obraz skomplikowanej i dynamicznej wiedzy
językowej niezwykle rzadko jest uwzględniany przez naukę o przekładzie
jako źródło problemów czysto tłumaczeniowych.
Rozdział trzeci i czwarty zajmuje się dwoma istotnymi etapami w
rozwoju umiejętności tłumaczenia. W rozdziale trzecim analizie poddano
tłumaczenie jako umiejętność naturalną, a zatem to jak tłumaczą osoby
znające dwa języki, ale nie posiadające żadnego przygotowania lub
wykształcenia jako tłumacze. Autorka podkreśla, że analiza takich
tłumaczeń jest ważna dla zrozumienia ewolucyjnej ciągłości wpisanej w
rozwój kompetencji przyszłych tłumaczy zawodowych. Bazując na
skąpych badaniach w tym zakresie, naturalni tłumacze, zarówno osoby
wychowane (tzw. ‘natural bilinguals’) i funkcjonujące w dwóch językach
(np. immigranci) lub uczące się języków obcych są w stanie tłumaczyć,
ale ich tłumaczenia są problematyczne pod względem formy lub/i treści.
Aby zatem umiejętność tłumaczenia mogła się rozwinąć do poziomu
 Streszczenie 461

oczekiwanego od zawodowych tłumaczy niezbędne jest nabycie szeregu

kompetencji, które pozwolą na uniknięcie błędów językowych i tłumacze-
niowych. Rozdział czwarty omawia obecny stan wiedzy dotyczący
zakresu kompetencji tłumacza, zagadnienia z zakresu pedagogiki
przekładu i analizę zawodu tłumacza.
Rozdział piąty przedstawia propozycję autorki dotyczącą rozwoju
kompetencji tłumaczy zawodowych postrzeganej jako proces integracji
wiedzy. Hipoteza, iż doskonalenie umiejętności dokonywania przekładu
jest świadomym i celowym rozwojem umiejętności budowania tzw. Sieci
Integracji Wiedzy (Knowledge Integration Network) oparta jest na
założeniach przyjętych w naukach kognitywnych i psycholingwistyce
przekładu, i szczegółowo wyjaśniona w odniesieniu do podobieństw
i różnic pomiędzy niedoświadczonymi i doświadczonymi tłumaczami. W
takim ujęciu tłumacz, to uczący się ekspert, który wykorzystuje swoje
doświadczenie i autorefleksję, aby rozwijać pewność siebie konieczną do
rozwiązywania nowych problemów w trakcie dokonywania przekładu.
Hipoteza Sieci Integracji Wiedzy zostaje poddana weryfikacji poprzez
badania empiryczne. W rozdziale szóstym, autorka analizuje ankiety
wypełnione przez potencjalnych i praktykujących tłumaczy, a w rozdziale
siódmym analizie poddaje proces tłumaczenia tekstu przez tłumaczy
znajdujących się na różnych etapach rozwoju kompetencji, od począ-
tkujących (studenci Filologii Angielskiej) do tłumaczy z 30-o letnim
doświadczeniem. Metoda badawcza, to program komputerowy Translog,
(Jakobsen and Schou 1999) który pozwala na wgląd w proces
tłumaczenia zapisując wszystkie czynności wykonane na klawiaturze
podczas tłumaczenia tekstu.
Wnioski z przeprowadzonej analizy potwierdzają ewolucyjny
charakter rozwoju kompetencji tłumaczeniowych jako procesu integracji
wiedzy językowej i konceptualnej. Implikacje, które wyłaniają się z prze-
prowadzonej rozprawy dotyczą pedagogiki przekładu, wykorzystania
czynności tłumaczenia w procesie nauczania języka obcego i przyszłych
badań nad tłumaczeniem jako umiejętnością człowieka, która angażuje
całkowity potencjał intelektualny.