GUIDE TO SINGING CHANT

(SFHPSJBO
DIBOU uses the traditional square notation, CSJFGMZ
EFTDSJCFE
IFSF

For a detailed explanation, consult the introduction to the Liber usualis
or any one of several chant textbooks. The following guide, which
follows the classic Solesmes interpretation, is necessarily brief.1

Notes and Groups of Notes

Traditional chant notation uses various types of individual notes and
groups of notes. Each note, either alone or in a group, receives a single,
equal pulse, regardless of its shape. The classic Solesmes method does
not recognize different proportionate note values (half-notes, sixteenth-
notes, etc.) among the different shapes.2 The basic individual pulse can
be considered the equivalent of an eighth note in modern music. It
may be stretched by the use of various rhythmic markings (see below).
Of the individual notes, the most basic are the punctum . and virga N .
These are combined to form groups of notes, called neums, which are
sung in consecutive order. For the clivis ,, the first and higher note is
sung first, followed by the second, lower note. For the podatus <, the
bottom note is sung first, followed by the top note. These two-note
groups may describe an interval of a second, third, fourth, or fifth.
The three-note groups include the torculus ], for which the three notes
are sung consecutively, the middle note always being the highest. For
the porrectus v, the extended diagonal element represents the progres-
sion of two descending pitches from one end to the other, while the
higher single note at the end is sung third in the series. The climacus c
uses a series of smaller notes. Each rhombus receives the same standard
pulse, and the shape merely indicates the downward progression of the
notes. For groups of three or more notes, any of the constitutive inter-
vals may describe a second, third, or fourth (rarely larger).
Repeated single notes in proximity are customarily rendered as a single
note of proportionate value: two punctums . . (bistropha) equal a note
two pulses in length; three punctums . . . (tristropha) equal three pulses.
The same applies to neums that contain repeated notes, like the pressus
. . Though repeated notes are treated as a single composite tone, the
,
passage of the individual notes may be marked by a slight swelling of
the voice (crescendo).
The final note in a two- or three-note group may sometimes appear
smaller than normal x. This small note is a liquescent, and is used for

1
For this guide, I have borrowed freely from the work of Dr. Lila Collamore.
2
This method follows the classic Solesmes interpretation, which is equalist. Some
theorists do recognize proportionate values in the manuscript notation.
Latin syllables whose final consonant is voiced (l, m, n, j, etc.) or whose
vowels are treated as a diphthong (au). It is rendered by singing the
regular notes of the group on the vowel (a, in the case of au), and clos-
ing to the voiced consonant or auxiliary vowel on the liquescent note.
Some conductors advocate closing to the auxiliary for the entire value
of the liquescent note, and an unusually warm acoustic might call for
this. Others may find it excessive, and will prefer to wait until roughly
halfway through the note before closing to the auxiliary. When the
liquescent takes the consonant t, it is best to treat it as a normal note.
The liquescent note receives the same rhythmic pulse as any other note.
Additional neums with special rhythmic properties appear below.
Beginning singers need not be overwhelmed by the terminology, but
they should learn to recognize the basic shapes and how to sing them.
(See the Table of Neums, p. 313)
Staff and Clef Signs
Chant is notated on a four-line staff. Notes fall on lines and spaces, as
they do on the modern five-line staff, and moving from a line to a
space represents the movement of one degree in the scale, at the inter-
val of either a whole step or a half step. The staff can be extended by
the use of ledger lines.
The chant staff accommodates melodies of varying range by using two
types of moveable clefs:

BvvVv Cv Do-clef
marks do on the staff XvvZv Fa-clef
marks fa on the staff

Clef signs are placed first on every line of chant. They mark the position
of either do m or fa, M on the staff (and thus the position of the semi-
tones, or half-steps, in a diatonic scale), and from these the singer
determines the relative positions of all the other degrees in the scale. In
longer pieces, the range might shift part way through the piece, and
may require a clef change. The new clef appears following a double bar,
and do (or fa) is repositioned accordingly. The various placements of
the different clefs can be confusing, but they are necessary to keep the
majority of notes for a given melody on the staff. In time, with con-
sistent use of solfeggio (do-re-mi), finding the relative position of notes on
the different clefs will become second nature.

Bbb zczbcvb cbxvbcbvbzcvbzcvbz czcv\c}vVbzczbcvbcbxvbcvbzcb }

DO
TI

|half-step|
LA
SOL
FA
MI

|half-step|
RE
DO
TE
DO
TI

|half-step|
LA
SOL
FA
MI

|half-step|

Xxbcvb cbxvbcvbzcvbzcvbz czb v vb }vZzczbcvbcbxvbcvbzcb}

DO
TI
LA DO
SOL TI
FA LA
MI SOL
RE FA
DO MI
Guide to Singing Chant

Other notational signs include the flat, natural, and custos:

flat sign natural sign

Bv»v creates te (ti-flat)

lasts for word or incise,
whichever is smaller Bv»vjv cancels
Bv\vjv
custos

vvõ cue to the first pitch of the

next line

The only accidental in chant is the one flat on ti, which lowers that tone
a half step (to te). The flat remains in effect until the end of the word
or until the next barline, whichever comes first. The natural sign is
used to cancel the flat, if necessary.
The custos (or guide) appears at the end of every line of chant. It is not a
note, but a visual cue for the first pitch on the next line.
Rhythmic Markings and Expressed Notes
Most characteristic of the classic Solesmes method is its use of special
rhythmic markings. These markings are not present in the chant manu-
scripts (though they are sometimes inferred), but are added as an aid
to singers in order to achieve an artful and coherent rendering of the
chant melodies.
As we have said, individual notes receive the same rhythmic value,
irrespective of their shape. However, notes can be expressed in several
different ways, which may affect their relative length:

1. by the addition of a dot:

cfb,c cfzfc
= i.e. two pulses

2. by the addition of a
horizontal episema: cëfc céYG c cìtfc cìfìgfc
3. in the context of a
special neum: cfÃYcv cèfYG cv
quilisma sung as

cfGßYcv cfYGé cv
salicus sung as

The most fundamental rhythmic marking is the dot, which doubles the
length of the note it follows (whether punctum, virga, or rhombus),
giving it two pulses instead of one. Dotted notes often precede a
barline, in which case they receive a slight relaxation (ritardando and
diminuendo). Those that appear in the middle of the phrase may mark
the end of a sub-phrase, and also may receive a slight relaxation of the
tone; however, this is followed by a re-energizing of the tone on the
dot, to propel the voice into the rest of the phrase.
The horizontal episema _ affects the sound of the note by adding expres-
sion. Such expression is best understood as a slight pressure and length-
ening (as in the description of the quilisma and salicus below). It is not
an accent, as understood in modern music. It is not a doubling of the
note value. It is much more nuanced and subtle, and should never
affect the overall rhythmic flow of the melody. Often, beginning sing-
ers adopt too rigorous an interpretation that does, in effect, double all
the notes marked with a horizontal episema. It might be more fruitful
for beginners to wait to include episemas until the melody itself, in its
rhythmic integrity, has been well absorbed.
The amount of expression given by the horizontal episema depends on
its context. It chiefly affects the note it is over (in the case of a podatus,
the first note of the group). However, as with the quilisma and salicus,
the horizontal episema should never be rendered rigorously or mechan-
ically. Expressed notes of all types may need to be prepared by a slight
anticipation, and their effect may need to linger by a slight reluctance
to return to tempo. The musical and textual context, the shape of the
phrase, and rules of good taste will, with practice, guide their ultimate
interpretation.
Longer episemas extending over two or more notes affect all the notes,
but with decreasing strength. The first note receives the most obvious
expression, and each subsequent note less expression. Long episemas at
the ends of phrases are most marked; those in the middle of the phrase

¿
less so, and in this case, the final note of the group generally should
return to the regular tempo.
The quilisma is a special note; as customarily rendered, it gives ex-
pression to the note preceding it (a slight pressure and lengthening).
Otherwise, the quilisma itself is sung like any other note. Despite
appearances, it is not treated as a vocal trill.3
The salicus b is a special neum; it can be recognized by the vertical
stroke that marks the middle note of the group (the same as an ictus
mark, described below, but in this context, it is called a vertical episema).
Like the quilisma, the salicus is rhythmically modified, with expression
given to the note marked with the vertical episema.
Except as part of a salicus, the vertical episema (ictus mark) + does not
affect the rendering of the notes to which it is attached, either through
length or stress. As discussed below, it is added purely as a guide to
ensure proper grouping of musical pulses into two- and three-note
rhythms.

3
The Liber usualis, in its guide to interpretation, hints at the possibility of a trill,
but recommends this more practical rendering “if one has not learnt how to execute
these tremolo or shaken notes, or, knowing how to render them, has nevertheless to
sing with others.” Most conductors consider this very good advice.
Guide to Singing Chant

Barlines
Chant is not measured; its notes fall into unequal groups of twos and
threes. Barlines in chant mark the ends of various types of phrases:

quarter bar half bar full bar double bar

Bzcccvvcc[v ccczcc{cczv ccc]cv cbccc}cv

Full and double bars mark the end of a significant phrase. They are
treated as full stops and preceded by a slight ritardando. Half bars mark
less significant sections; breath may be taken, but the rhythm should
not be significantly interrupted. Quarter bars mark shorter musical
phrases. The rhythm should not be interrupted, and breathing, if
needed, should steal time from the note preceding the barline.
Full barlines also serve as musical rests. Following the barline, time is
added using rests that are equal in value to either a single or double
pulse (where the punctum receives a single pulse, equivalent to an
eighth note). The value of the rest depends on the rhythm of the
phrase following the barline. If the first note of the next phrase receives
an ictus, it is treated as a downbeat, and is prepared by two pulses
(quarter rest). If the first note of the next phrase does not receive an
ictus, it is treated as an upbeat, and is prepared by one pulse (eighth
rest). This affects counting as follows:

if the note following

a full or double bar
is ictic:
v vbfzc, b ]v vzTF cbz fbvz bfv z cbfz,bcv z ]v zTF vzb fbvz bfv z
12 12 12 1 2
=
Œ

(downbeat)


if the note following
a full or double bar
is not ictic:
cbbfz,cbz]v z fcbtfv zbfbz zfc cbfz,cz]zv z fvbvtfvzvfvzvfv
12 1 2 12 1 2
=

(upbeat)

In a psalm recitation, the next known ictus following the full or double
bar may be many notes away. Counting back (described below) may
yield a result that seems especially counterintuitive, or that contradicts
the textual rhythm in a particularly unnatural way. In this case, the
conductor is free to add the value of rest that seems most natural.
It is important that the note before a full or double bar (invariably a
dotted note) be given its full value. Singers can enhance the sense of
cadence at these points by singing into the barline, placing any final
consonant on it, or even slightly after it.
The asterisk * is generally used to signal the end of an intonation (the
opening phrase of a piece, usually sung by a cantor) and the entrance
of all the singers. Some conductors treat the asterisk as a full or double
bar, always adding a rest after it. However, in cases where the note
before the asterisk is not dotted or lengthened in some other way (e.g.,
the first Kyrie from Mass I, page 60), it may be better not to add time,
but to keep the rhythm flowing, and have the singers enter as if they
had already been singing. Experience will judge whether this is practical.

Plainsong Rhythm
Nothing is more characteristic of the classic Solesmes method, nor has
been the source of more scholarly controversy, than the topic of plain-
song rhythm. Even a cursory discussion is beyond the scope of this
guide, but a thorough understanding of the Solesmes rhythmic method
is essential for the proper and artful singing of chant.4
Of basic consideration is the proper arrangement of notes into two-
and three-note groups, which form the basic pattern of “beats” in chant
rhythm. The beginning of each group receives the rhythmic ictus, or
touching point. Of itself, the ictus is purely organizational, and indicates
no qualitative change in the rendering of the note—not emphasis, not
lengthening. The basic rhythmic groups of twos and threes are further
combined to form larger groups that either tend to rise (arsis) or fall
(thesis). It is up to the conductor to expresses this pattern of rise and
fall. See a more detailed method for a complete discussion of chant
conducting (chironomy).
The musical ictus may or may not correspond to a textual ictus, the
strong syllable of the word. This subtle interweaving of the musical and
textual ictus is the defining characteristic of classic Solesmes rhythm,
and once grasped, it is the key that unlocks the magic of plainsong.
As a practical matter, the proper marking of the rhythmic ictus is
invaluable for keeping a schola together and moving forward at a
steady, deliberate pace. Within this firm, ictic framework, the rhythmic
markings (horizontal episemas and special neums) provide subtle,
supple points of relaxation and expression to the melody. Conductors
and singers both need to know the exact placement of the ictus and
how to find it, when it is not marked.

4
Consult one of the following:
Gajard, Joseph. The Rhythm of Plainsong According to the Solesmes School. New York:
J. Fischer and Bro., 1945; reprint ed., Richmond, VA: Church Music Association of
America, 2007.
Mocquereau, Dom Andre. A Study of Gregorian Musical Rhythm (Le Nombre Musical
Gregorien). Volume 1. Solesmes, trans. Aileen Tone, 1989; reprint ed., Church Music
Association of America, 2007.
Guide to Singing Chant

There are four ways to find the musical ictus, which are presented in
order of precedence:

1. as indicated by the
ictus mark: cÞfc
2. as it falls at the beginning
of a long or doubled note: z z ygbtfz vbzfbfbz vzszmvbv z z ygß btfz bvzÞfbfbz vzsÜ bmvbv
=

3. as it falls on the first note

of a neum: z z fYG vbz rdv fb,vbv z z fÞ YG vbz rá dv Þfb,vbv
=

4. by counting backward by twos from next known ictus:

written: Bzz fYG vbz fv zbgv fb,vbv z b\vjv vzhvbv vgbv z hb.v {v gv vz fbv bdv bfv bvGYv bgv tfb<M z }
A-ve verum Corpus na-tum de Ma-rí- a Vírgine:

known, using
rules 1–3: Bzz ÞfYG vbz fv zbgv Þfb,vbv bz\vjv vzhvbv vgbv z àhb.v b{v gv vz fbv bdv bfv bvßYG v bgv tâ Þfb<M z}
A-ve verum Corpus na-tum de Ma-rí- a Vírgine:

added,
counting back
by twos:
Bzz ÞfYG vbz Þfv zbgv Þfb,vbv bz\vjv vzàhvbv vgbv z àhb.v b{v ßgv vz fbv bÝdv bfv bvßGYv bgv tâ Þfb<M z}
A-ve verum Corpus na-tum de Ma-rí- a Vírgine:
Conductors may need to modify this rule for the sake of musical sense,
or to preserve a good ensemble. They must also determine how much
of this information they want to provide to their schola, or how much
they want to rely on conducting alone to communicate the rhythm.
Order of precedence means, for instance, that a note marked with an ictus
takes precedence over the first note of a neum; the first note of a
double note also takes precedence over the first note of a neum; etc.
Marking all the ictuses helps the conductor and singers see the groups
of twos and threes that form the basis of chant rhythm. This is further
reinforced by counting out the resulting patterns, beginning with one on
the ictus, followed by two and, as necessary, three on the non-ictic
notes. Just as solfeggio reinforces the relative pitches of a melody,
counting instills a clear sense of its underlying rhythm.
Repercussions
For groups of repeated notes that appear consecutively and slightly
separated, the second group receives a fresh impetus, called a reper-
cussion, which should mark, but not interrupt, the flow of the sound.
The same effect applies when a group of repeated notes precedes a
neum, or when it precedes the same note marked with a vertical epi-
sema. Some cases of repercussion follow:

1. on a new ictus:
cfzf„rá dc[cÞfzfzf„Þfzf„Þfzfc}
2. before a new neum:
cÞfzf„rÝdc
The repercussion is best rendered subtly. Should the effect prove too
strong when sung by the whole group, the conductor might want to
assign the task of rendering the repercussions to only a few singers.
Modal Melodies
Chant melodies are modal, and each is given a modal classification
(which appears as a Roman numeral on the first line of each chant),
based on one of the eight ecclesiastical modes.5 These eight modes cor-
respond to the first four modes in the ancient Greek system, with each
of the four appearing in two forms, either authentic or plagal, depending
on the range of the melody (higher or lower, respectively) and the pre-
vailing dominant (or tenor) of the scale. The final (or tonic) of the mode
is usually the last note of the chant, and gives the mode its tonality.

Ecclesiastical mode Greek mode Final Dominant

I and II Dorian RE I: LA II: FA
III and IV Phrygian MI III: TI (DO) IV: LA
V and VI Lydian FA V: DO VI: LA
VII and VIII Mixolydian SOL VII: RE VIII: DO

Modes are most easily understood as they correspond to scales played

on the white keys of a piano, starting on D (Dorian), E (Phrygian),
F (Lydian), and G (Mixolydian). This exercise shows how modal scales
are characterized by their arrangement of whole steps and half steps,
which fall in different places depending on the starting note, but it is
only for demonstration. In practice, modal scales can be sung starting
on any pitch. Chant does not have key signatures like those of modern
music.6 The modal classification, and the pattern of whole and half
steps it implies, is the only tonal information given. Chant notation
represents relative pitch only, not absolute pitch.
For this reason, it is essential that beginning students of chant use the
classic solfeggio system (do-re-mi) when learning a new melody, always

5
The corpus of Gregorian chant employs additional modes, but they do not appear
in this collection.
6
A few chants use a flatted ti throughout, and are written with a key signature of
one flat.
Guide to Singing Chant

remembering that do is moveable, and corresponds to whatever pitch is

chosen. It is up to the conductor to choose the absolute pitch for do (or
for the final of the mode), based on the melodic range of the piece, and
how it falls within the vocal range of the singers.
Although the eight ecclesiastical modes do not correspond to the major
and minor scales of modern music, each features either a major or minor
third, and can be described as major (V–VIII) or minor (I–IV). With prac-
tice, singers can begin to recognize the special character of each of the
modes, including certain characteristic melodic gestures.
Table of Neums
The following table lists the most basic notes and groups:

v gv punctum
v 7v virga

vTF v FYv podatus (pes)

bottom note sung first v yÎfv clivis
higher note sung first

bghgv ØfhÎfb torculus

all notes are of equal
value, sung consecutively
bÎhYvÎ6YØ b porrectus
three notes, the first two at
either end of the diagonal

v 6b%$v climacus
all notes, including the small rhombus, are
of equal value, and are sung consecutively

v gbgv bistropha (distropha)

repeated notes sung as a
single note of double length
zvgbgzgvz tristropha
repeated notes sung as a
single note of triple length

v gbtÍdv pressus
repeated notes sung as a
single note of double length
v fYÃ v quilisma
middle note of a three-note
group; the note before is
expressed

vYF 7v fGUv scandicus

all notes are of equal value vdYFÞ v AEFYÞ v salicus
the last two notes form a
podatus; the note marked

vyg v vF^v liquescent notes

pronounce a diphthong
(a-u) or voiced consonant
( l, m, n, j, etc.) on the v SYØã 7v
with the ictus is lengthened
when the first interval of
the salicus is a 5th, the first
two notes form the podatus;
small note the note marked with the
ictus is lengthened
 GUIDE TO PRONOUNCING LITURGICAL LATIN
Successful singing of plainsong requires attention to the proper pronunciation
of traditional liturgical Latin, which differs from that of classical Latin in
several ways. In the first place, vowels follow the Italianate model, and are
sung as purely as possible (within the bounds of vocal taste), with no hint of
diphthong—the elision of vowel sounds common in vernacular languages. The
following table gives equivalents, which hold true no matter the position of the
vowel in the word:
A as in father (never as in add)
E as in mellow, with no elision to the i, as in obey; singers who
chronically make diphthongs of e’s are encouraged to add an h
(Domin-eh D-eh-us)
I as in pizza; it is always sung with a long e sound, as in feet; this holds
true even when it is followed by a voiced consonant (in = een;
dimittimus = dee-meet-tee-mus)
[Obviously, this can be overdone, and should be modified for
grateful singing, especially on high notes.]
Note: In this edition, I is always used as a vowel; in some editions, I
is used as a consonant in place of J, following classical usage, and is
equivalent in sound to the English consonant y (iubilate = jubilate =
yoo-bee-lah-teh)
O as in motion or for, with never a hint of diphthong to u (as in mowing);
adding an h might help (n- oh-mine D- oh-mini)
U as in truth; it is always sung like a long oo sound (as in boot, not as in
foot), and is never shortened (but or put); it should never be preceded
by a diphthong (as in cute)
Y is always treated as a vowel, equivalent to I above
Æ and Œ, in sung Latin, are treated as E above; they are often written as
separate letters (ae, oe), but are always treated as a single vowel sound
AU is a true diphthong, with A receiving the majority of the pitch, and U
added just before the next syllable; especially in melismatic passages
where the syllable with au receives several notes, singers should main-
tain a pure ah sound for as long as possible (ca--usam, la--udate)
NGU– , when the U is followed by a vowel, creates another true diph-
thong, in which the second vowel receives the majority of the pitch
(san-guIs, san-guI-ne, lin-guA, un-guEn-tum, pin-guI-um, etc.)
Otherwise, consecutive vowels are almost always pronounced sepa-
rately and more or less equally (De- i, me- us, tu- um, tu- o, Evangeli- i)
Pay special attention to words that look like English words, but which never-
theless are sung using pure Latin vowels:
Immaculati = ee-mah-coo-lah-tee
In liturgical Latin, certain consonants receive special pronunciation:
C is hard, like k before hard vowels (a, o, u); but
is soft, like ch before soft vowels (e, i, æ, œ, y)
CC becomes t-ch before soft vowels (ecce = et-che)
CH is always hard, like k, before all vowels and consonants
Guide to Pronouncing Liturgical Latin

G is hard, as in got, before hard vowels (a, o, u); but

is soft, as in gentle, before soft vowels (e, i, æ, œ, y)
GN is pronounced as in Italian (signor), it sounds like n-(i), with a soft
diphthong before the vowel (Agnus = an-(i)us)
H is always silent, never aspirant; when sung at the beginning of a
word, it may modify the vowel slightly to prevent a glottal
J sounds like the English consonant y; in some editions, it is replaced
by I (see note above)
R is a troublesome consonant generally, as sung by Americans, and can
be excruciating when applied to Latin; great pains must be taken to
see that it is never significantly voiced (even when it falls on a
liquescent); it should receive a quick flip with the tip of the tongue,
nothing more; it should never modify or interrupt the vowel it
follows; beginning singers may need to draw a line through all
problematic r’s
PH is always pronounced like F
S is always hard, as in pass (never soft or z-like, as in was); however,
when it falls between two vowels, it may be softened somewhat
(miserere)
SC becomes sh before soft vowels (e, i, æ, œ, y) (ascendit = a-shen-dit);
before hard vowels (a, o, u), it is sk (scandalum, sculptus)
TH is always hard, as in Thomas
TI becomes tsi when followed by any vowel (lætitia = leh-tee-tsee-ah); the
rule does not apply when it is preceded by S, X, or T (hostiam)
X is always hard, like ks; however, when it falls between two vowels, it
may be softened somewhat (exercitus = ek-ze-rchi-tus)
XC becomes k-sh before a soft vowel (e, i, y) (excelsis = ek-shel-sis), but not
before a hard vowel (a, o, u) (excubo = eks-ku-bo)
Z is softened and dental, like dz (azymus = a-dzy-mus)
Finally, in order to achieve the sort of rhythmic “fluidity within solidity”
demanded by plainsong, care must be taken when executing the Latin accent.
For two-syllable words, the accent always falls on the first syllable; for longer
words, the accent is marked (Laudáte Dóminum). Unlike vernacular languages,
Latin employs a “quality” accent, rather than one of quantity, either of weight
or volume. Because this quality mostly involves lifting the pitch, it is difficult
to apply to a text that has a fixed melody. However—and especially in cases of
recitation, such as in Psalm verses and other passages with a single repeated
note on several syllables—the sense of a lifted accent can be achieved through a
slight heightening of the voice, both in strength and, to a lesser degree, duration.
But this can never sound mechanical, and is best achieved when thought, more
than sung.
Plainsong, following the classic Solesmes model, respects the rhythmic impulse
of the music in equal degree to that of the text; the seamless integration of these
two rhythmic elements is the ultimate goal in singing the chant, even in cases
where the two may seem at odds. A gently flowing, steady musical rhythm
need not be disrupted in an effort to convey the Latin accent; but neither
should the music obscure the meaning of the text, which is, after all, the
essence of the prayer it seeks to enliven.