Mapping Place Names

Søren Wichmann
Kiel University, GERMANY
Lennart Chevallier
Kiel University, GERMANY

ans-names.pitt.edu
ISSN: 0027-7738 (print) 1756-2279 (web) Vol. 73 No. 2, Summer 2025 DOI 10.5195/names.2025.2616

Articles in this journal are licensed under a Creative Commons Attribution 4.0 International License.

This journal is published by Pitt Open Library Publishing.

2
NAMES: A JOURNAL OF ONOMASTICS
Søren Wichmann and Lennart Chevallier

Abstract
This paper demonstrates how to leverage the GeoNames data for seeking patterns in toponymic data using
the software package ‘toponym’, which we wrote for the R computational environment. After discussing a
distinction between particularistic and pattern-seeking approaches to toponymics, we go on to characterize
the data of GeoNames, which are particularly appropriate for the latter type of approach. Then, we present
two cases studies. The first case study is on Xincan place names in Guatemala, and the second is on Slavic
place names in eastern Germany. These explorations support our hypothesis that toponymics may benefit
from new computational tools.
Keywords: toponym, hydronym, R, software, GeoNames, Xincan, Slavic

General Introduction
Place names (toponyms) are perceived as potentially imbued with meaning (Radding & Western 2010; Algeo
& Algeo 2000), yet we also accept the apparently arbitrary relationship between a designation and its referent
as we encounter it. Thus, newcomers in a region often adopt existing place names, even if they are opaque to
them, and toponyms can become important sources of information that “permit historical inferences about
languages and the people who spoke them” (Campbell 2013, 436).
For instance, the US states of Mississippi (Baca 2007) and Alabama (Read 1994) are teeming with
Choctaw names, bearing witness to parts of the extension of this ethnic group before most Choctaws
underwent forced relocation to Oklahoma in 1831–33. As the editor of the Vicksburg Daily Sentinel,
eyewitness to the removal, wrote, “They leave names to many of our rivers, towns, and counties, and so long
as our State remains the Choctaws who once owned most of her soil will be remembered” (DeRosier 1970, 4).
Today only native speakers and linguistic specialists will be able to interpret the meaning of old Choctaw
place names, for instance (in Mississippi, following Baca 2007) Tchoutacabouffa ‘Broken Pottery’ River, Itta
Bena ‘Forest Camp’, Tallahoma ‘Red Rock’ Creek, and (in Alabama, following Read 1994) Buttahatchee
‘Sumac River’, Coatopa ‘Wounded Panther Creek’, Nanafalia ‘Long Hill’, and so on. Even for specialists, the
majority of names of likely Choctaw origin remain more or less obscure.
The opposite case, where new names are given to locations already carrying names in a language other
than the one of the newcomers, is of course also common. Sometimes this is done systematically for political
motives. For instance, in 757 CE King Kyeongdeok of Silla changed toponyms in the realm of his kingdom
(within modern-day North and South Korea) into standardized names written with two Chinese characters
(Endo 2021, 35). Closer to the present, in the years after 1945, hundreds of place names in formerly German-
occupied areas of Czechoslovakia were renamed from German into Czech (Lehmann 1999). Finally, during
1949–1960, the vast majority of Arabic place names were replaced with Hebrew ones in the new state of Israel
(Azaryahu & Golan 2001), a practice which continued as new territories were included after the Six Days War
of 1967 (Cohen & Kliot 1992). Interestingly, in all three cases translating the place names was a common,
consciously applied strategy. This testifies to the conservative power inherent in toponyms, even in the face of
radical reformative stances.
Detailed philological work on place names commonly involves seeking out documentation for the
earliest attestations (cf. already Petersen 1833) and carrying out lexical, morphological, and phonological
analysis (e.g., Bright 1998). In addition to this line of work, onomastic research may profit from the
inspection of geographical distributions of crude phoneme/letter sequences. For instance, linguistic work tells
us that the suffix -ham contained in many British place names is West Germanic and means something like
‘dwelling, homestead, hamlet, for example.’, and the study of the documentary record indicates that it begins
to appear in the 5th century (Copley 1988, 31). Even without linguistic analysis and archival work, however,
we can hypothesize just from the distribution of this string that -ham is a suffix with a relatively generic
meaning belonging to a stratum of English place names which is neither of Celtic nor of Scandinavian origin.
Figure 1 shows 817 contemporary occurrences of this string in Great Britain and Ireland. It is infrequent in
Scotland, Wales, and Ireland, and not in any way confined to the Danelaw area. Clearly the gazetteer of
Copley (1988, 31–35), which discusses nine early place names in -ham, is well complemented by a raw
distributional map of place names ending in the letters -ham, even if the latter does not directly tell us
anything about the meaning of the apparent suffix and contains no historical stratification.

ans-names.pitt.edu
DOI 10.5195/names.2025.2616 Vol. 73 No. 2, Summer 2025 ISSN: 0027-7738 (print) 1756-2279 (web)
 NAMES: A JOURNAL OF ONOMASTICS 3

Mapping Place Names

Figure 1: The Distribution of -ham in the United Kingdom and Ireland as Displayed Using the Toponym
Package
With the advent of the personal computer it has become easier to generate distributional maps of place names
containing particular strings. An early example is the overview of common terms for “stream” (creek, branch,
run, brook, and so on) in Campbell (1991), which draws upon the Geographic Names Information System
(GNIS) of the United States.1 The author acknowledges technical assistance for producing the maps. The
introduction of geographic information systems (GIS) has made it easier to produce custom digital maps, and
publications in toponymy have profited from these technological advances. Chen et al. (2014, 6,315–6,316)
briefly reviews many early GIS-based toponymic studies. Among later studies we can mention Fuchs (2015),
which employs ArcGIS in the plotting and distributional analysis of German place names in the US Midwest
and Endo (2021) which, using ArcGIS online, plots Korean place names from the first millennium CE with
different linguistic backgrounds (Japonic, Koreanic, Sinitic, Tungusic, Mongolic).
In the preceding we have exemplified two general approaches to toponymy—a mainly philological
approach, focusing on individual names, including their etymologies, and a pattern-seeking approach,
simultaneously considering multiple names. Obviously the two types of approach are not sharply distinct, nor
do they reflect some kind of opposition—instead, they are better characterized as complementary. Tent (2015)
suggests distinguishing them terminologically. This author discusses contrastive terms that emphasize
different facets of the two different approaches, including “micro” vs. “macro”, “case” vs. “pattern” analysis,
and “qualitative” vs. “quantitative” research, but his preferred terms are “intensive” vs. “extensive” toponymic
research. We would concur with this terminology, although other candidates might be “particularistic” (a
term also used by Hunn 1996, 4) vs. “pattern-seeking”. As for the terms ‘qualitative’ vs. ‘quantitative’ for the
two approaches, also employed by Tent (2015), we would argue that a quantitative approach is one in which
numbers and statistics matter, not simply one that draws upon many data-points. Thus, when Tent (2015, 67)
presents a table according to which the quantitative approach is used in 15.6% of papers in the journal Names
in 1952–2014, with the rest representing qualitative, mixed or other approaches, we suspect that the
percentage of papers that are truly quantitative in nature is actually smaller. It is in any case clear that there is
plenty of room for applications and developments of the pattern-seeking approach to toponomy, and even
more room for the introduction of statistical methods. We see our own work as a contribution to the pattern-
seeking approach to toponymy, and our hypothesis in this paper is that new tools aiding research using this
type of approach have a great potential.
The present paper illustrates the use of a package written in the R computational environment (R Core
Team 2022) which is designed to filter and display any subset of millions of place names from across the

ans-names.pitt.edu
ISSN: 0027-7738 (print) 1756-2279 (web) Vol. 73 No. 2, Summer 2025 DOI 10.5195/names.2025.2616
4
NAMES: A JOURNAL OF ONOMASTICS
Søren Wichmann and Lennart Chevallier

world, producing maps like the one in figure 1 and allowing for outputting tables that include relevant names
and corresponding geographical coordinates. We offer this tool for the science of onomastics in particular and
also for any other purpose for which it might prove useful. Onomastics, as is well known, is not restricted to
place names, but its branch toponymy is. Subbranches hereof include hydronomy (the study of names of
rivers, lakes or other bodies of water), oronymy (the study of the names of hills, mountains or mountain
ranges), speleonymy (the names of caves, chasms, grottoes, mines, and so on.), and hodonymy (the names of
streets, avenues, lanes, and so on.) (Perono Cacciaforo & Cavallaro 2023, 3; Coates 2013). All these
subbranches of toponymy are served at least to some extent by our package. In the following sections, we first
introduce the data that the package draws upon and finally we provide two case studies illustrating the utility
of the tool. The nuts and bolts of the package itself are presented in a separate Note (Chevallier & Wichmann
2024). An online text file2 that follows the outline of the paper repeats all lines of code in the text and also
contains additional code allowing for figures to be recreated and results to be fully replicated.

Methodology
The tool used to carry out toponymic research in this paper is the R package ‘toponym’ (see Chevallier &
Wichmann 2024 for a practical introduction). The package provides an interface to the GeoNames data. Here
we provide an overview of the contents of this database, which is hosted at https://www.geonames.org/.
GeoNames (GeoNames 2023) is an evolving database of public domain geographical data. The version of the
database downloaded for this paper, as embodied in the allCountries.txt file (last modified at 03:52,
November 17, 2023), contains 12,492,864 entries pertaining to 253 countries and 9 feature classes. The
information is aggregated from more than 400 sources 3 and is entirely free for download. Figure 2 shows the
distribution of entries over the different feature classes, which are defined in the caption to the figure. As is
clear from figure 2, around two thirds of the database consist of names for the built or humanly defined
environment (P, S, A, L, R; ~8.5 million entries), with features of the landscape (H, T, V, U; ~4.0 million
entries) constituting the remaining third.

Figure 2: Number of Entries per Category of Toponym in GeoNames. P: City, Village, and so on; S: Spot,
Building, Farm; H: Stream, Lake, and so on; T: Mountain, Hill, Rock, and so on; A: Country, State, Region,
and so on; L: Parks, Area, and so on; R: Road, Railroad; V: Forest, Heath, and so on; U: Undersea

ans-names.pitt.edu
DOI 10.5195/names.2025.2616 Vol. 73 No. 2, Summer 2025 ISSN: 0027-7738 (print) 1756-2279 (web)
 NAMES: A JOURNAL OF ONOMASTICS 5

Mapping Place Names

We now briefly describe the columns of the GeoNames database, referring to column names in square
brackets. Of the entries, 100% contain an integer ID [geonameid], a toponym [name], geographical
coordinates [latitude, longitude], and a modification date [modification date]; 95.33%–99.99% of the entries
additionally contain an ascii version of the toponym [asciiname], a feature class [feature class] (see the
caption to figure 2), a feature code [feature code] indicating a subclass of the feature class, a two-letter
country code [country code], a code for highest-level administrative units (such as states in the US or
Germany) [admin1 code], a population figure [population], a digital elevation model (in meters) [dem], and a
time zone [timezone]; and 53.63%–67.20% of the entries contain alternate names [alternatenames] and
codes for second-level administrative units [admin2 code]. Finally, 00.10%–20.86% of the entries contain
third- and fourth-level [admin3 code, admin4 code], elevation (in meters) [elevation], and alternate country
conde [cc2].
In a few cases the contents of a column is not self-explanatory, calling for a few comments. The [name]
column contains toponyms in the standard spelling whenever the relevant writing system is Latin based.
When the writing system is not Latin based, the column will feature a Latin-based transcription and the name
rendered in the national orthography will instead normally be found in the [alternatenames] column—
possibly alongside other transliterations and/or variants. Non-Latin orthographic renderings of toponyms,
however, are provided less systematically than their counterpart Latin transcriptions. For instance, 11.1% of
all entries for Russia have empty cells in the [alternatenames] column, which holds Cyrillic versions. Similarly
4.4% of entries for China have empty cells in [alternatenames]. Thus, more complete results are expected
when accessing [name] than [alternatenames]. The relationship between the [dem] and the [elevation]
columns also invites comments. The [dem] column contains srtm3 (Shuttle Radar Topography Mission 3) or
gtopo30 (Global 30 Arc-Second Elevation) data, but the documentation does not specify the origin of the
[elevation] column data. The values are not identical, but correlate as r = .997 by a Pearson correlation, which
shows that they measure the same quantity but also that there is room left for large discrepancies in
individual cases. More importantly, only 19.14% of the entries have filled cells for [elevation], whereas [dem]
is near-complete. Hence, it would normally make little sense to use [elevation]. The GeoNames
documentation4 provides additional detail on the contents of the database. As far as we are aware, there is no
other single database of place names that might compete with GeoNames as a source of toponymic research.
Acheson et al. (2017) offer comparisons with the next largest global gazetteer, the Getty Thesaurus of
Geographic Names (TGN 2023), noting that GeoNames (at their time of writing) was about ten times as large.

Results

Tracking the Former Extent of Xincan Languages

In a book on Maya culture, Thompson (1970, 98–99) suggested that place names in eastern Guatemala and
western Honduras ending in -agua or -ahua, or their reduced forms -gua or -hua, may relate to a particular
non-Maya ethnic group which got partly displaced by the Ch’orti’ Mayas but had a stronger presence in
eastern Guatemala before the latter arrived on the scene. Thompson was not able to identify the ethnic group
more precisely. His presentation of the relevant toponymic data is worth quoting in extenso since it provides
for an interesting comparison with the output of the ‘toponym’ package
In eastern Guatemala and western Honduras, approximately the area of the Chorti at the
time of the Conquest, are Toquegua Indians, Motagua and Managua rivers; and the sites
of Tigua north of Camotan, Tigua east of Paraiso, Chanmagua or Chaumagua, Jagua,
Pasasagua, Anchagua, and Cocuyagua.
In central Honduras, approximately between 86° 50' and 88° 10', that is to say, east
of Chorti territory, are the rivers Jacagua, Jalagua, Chilistagua, Comayagua, Sasagua,
Chasnigua, and perhaps Ulua; and the sites of Silisgualagua, Manzaragua, Masahua,
Mulacagua, Tircagua, Chumbagua, Chapulistagua, Xagua, Xelegua, Eraxagua, Moncagua,
Teconalistagua, Laxigua, Talgua, Colomoncagua, Tiscagua, Apacilagua, Conchagua, as
well as Masaguara, Sicaguara, Yaguacire, and Yamaguare.
In El Salvador we find Comasahua, Atepammasagua, Quixnagua, Masahua, and
Moncagua, as well as Aguasarca and Guahtajigua.
One may also note other examples in territory even farther east and southeast, such
as Nicaragua, Managua, and even Veragua. (Thompson 1970, 98–99)
It can be verified that the majority of the particular place names mentioned are actually in GeoNames by
entering a search for the specific names. The result is displayed in figure 3. It shows 29 of the 49 place names.
ans-names.pitt.edu
ISSN: 0027-7738 (print) 1756-2279 (web) Vol. 73 No. 2, Summer 2025 DOI 10.5195/names.2025.2616
6
NAMES: A JOURNAL OF ONOMASTICS
Søren Wichmann and Lennart Chevallier

Figure 3: Attempt to Map All the Specific Place Names Hypothetically Associated with the “agua people” in
Thompson (1970, 98–99)
To replicate Thompson’s findings starting out from recurrent substrings rather than full names, the strings to
search for are final agua, gua, ua, ahua, guara, guacire, guare. Searching for these among names for
populated places and hydronyms in Guatemala, Honduras, El Salvador, and Nicaragua could, in principle,
return all the names mentioned by Thompson except Toquegua, which is an ethnonym, and Aguasarca,
where agua appears initially—which makes its inclusion somewhat dubious anyway. Additional names are
expected to appear. The output data frame contains 477 rows, but in many of these the target string pertains
to Spanish words: antigua ‘ancient’, legua ‘league’, yegua ‘mare’, Padua ‘Padua’, jagua ‘species of tree
(Genipa americana)’ [the last was included in Thompson’s list although the jagua is well-known name of a
tree in several varieties of Latin American Spanish]. After filtering we are down to 181 rows. The coordinates
of the revised data frame are plotted in figure 4.

ans-names.pitt.edu
DOI 10.5195/names.2025.2616 Vol. 73 No. 2, Summer 2025 ISSN: 0027-7738 (print) 1756-2279 (web)
 NAMES: A JOURNAL OF ONOMASTICS 7

Mapping Place Names

Figure 4: Trailing Strings agua, gua, ua, ahua, guara, guacire, guare in Toponyms in Guatemala,
Honduras, El Salvador, and Nicaragua, with Spanish Forms Removed
The maps in figures 3 and 4 greatly help to evaluate Thompson’s suggestions. Regarding the general
distribution of a(gua) names, the main insights from the extended search (figure 4) are that more of southern
Guatemala should be included than is apparent from Thompson’s specific examples. We also more clearly see
a radial distribution roughly centered on 14°N, 88°W, rather far from the area around the border between
Guatemala and Honduras that was Thompson’s concern.
Campbell (1978) cites the earlier suggestion of Thompson (1970, 98–99), but brings other relevant parts
of toponyms to the table, and his focus is on a historically known ethnic entity, namely Xincan. Xincan is a
small family of some four to five languages formerly spoken in southeastern Guatemala. Campbell’s latest
extensive discussion is in his 1998 publication, which is the study we will refer to here. The same ideas are
consistently repeated in multiple publications (Campbell 2013, 1998, 1978) as well as in a recent talk
(Campbell 2022), slides for which were generously shared with us by the author.
As in our handling of the data from Thompson (1970), we can first map specific names given by
Campbell (1998, 188) and, after that, the distribution of some specific strings supposedly pertaining to
Xincan toponyms. All the specific names mentioned by Campbell except four (Sanguayaba, Cerro Sansuque,
Sanyoyo, and Arloroma4) are found in the database, and we can provide a plot that was never provided by
Campbell but would represent his view (see figure 5). The map shows a compact area in southeastern
Guatemala, where the border seen is the one shared with El Salvador. This would seem to be an adequate
representation of Campbell’s view of the historical extension of Xincan as revealed through toponyms. In the
1998 paper he describes the area as extending “from the Motagua Valley to the Pacific Coast and from near
Guatemala City [e.g., San Pedro Ayampuc] to Honduras and El Salvador”, 5 and in the 2022 talk there is an
explicit statement that Xincan never extended into (what is currently) El Salvador.

ans-names.pitt.edu
ISSN: 0027-7738 (print) 1756-2279 (web) Vol. 73 No. 2, Summer 2025 DOI 10.5195/names.2025.2616
8
NAMES: A JOURNAL OF ONOMASTICS
Søren Wichmann and Lennart Chevallier

Figure 5: Specific Names Associated with Xincan According to Cambell (1998, 188)
Up to this point we have neglected some additional evidence that Campbell cites, namely some agua-type
names that Thompson (1970) had also discussed. Following the excerpt given above Campbell writes:
Thompson was on the right track; many of his examples turn out to be from Xinca ṣ̌aw̓i̵̓i̵
‘town’ (ultimately derived from the verb root ṣ̌awi̵ ‘to dwell’). [. . .] The following (many
from Thompson's list of “-agua” towns) appear to incorporate this Xinca term for ‘town’:
Pasasagua (compare Xinca ṣan-paṣaʔ ‘Pasaco’ [a town]), Jagua, Sasagua (Xinca ṣa(n)-
ṣ̌aw̓i̵ ‘in-town’), Xagua, Eraxagua (i̵ra ‘big’, i.e. ‘big town’), Conchagua, Comasahua,
Manzaragua, Anshagua, Anshigua, Xororagua, etc. None of these has a Mayan or Nahuatl
source, and the Xincan origin is clear for most. Campbell (1998, 188-189)
When we try to map these names the result is as in figure 6. As it turns out, there are multiple instances of
Conchagua in Honduras and El Salvador, but apparently none in Guatemala; Manzaragua is only in
Honduras; and Jagua is widespread, but this is a Spanish tree term as mentioned above, so we can simply
ignore that. The results in figure 6 suggest that names in -shigua, -shagua, -xagua, and (sometimes) -sagua,
but not those in -chagua and -ragua are good candidates for pertaining to Xincan, unless we want to accept
an extension way beyond southeastern Guatemala for which evidence is otherwise lacking.

ans-names.pitt.edu
DOI 10.5195/names.2025.2616 Vol. 73 No. 2, Summer 2025 ISSN: 0027-7738 (print) 1756-2279 (web)
 NAMES: A JOURNAL OF ONOMASTICS 9

Mapping Place Names

Figure 6: Specific Place Names Associated with the “agua people” in Thompson (1970, 98–99) and Taken by
Campbell (1998) to be Xincan
While the proposal of Thompson was pattern-seeking and that of Campbell mainly particularistic we hope to
have demonstrated that our software package was helpful in investigating both proposals through the maps
produced. Although a few specific names should be removed from Campbell’s list of Xincan toponyms his
proposal could be supported. 6 As for Thompson’s “agua” people it seems unlikely that the data supports a
hypothesis linking a toponymic ending similar to -agua to a specific ethnic entity, although the idea merits
further investigation. We now know more about the distribution of the toponyms containing the endings that
Thompson discussed, which is an essential first step in the further investigation of this proposal.

Slavic Toponyms in Germany

In this subsection we demonstrate how our package can be helpful in providing distributional overviews of
the most common Slavic toponymic elements in German place names. They pertain to West Slavic varieties
spoken in the eastern parts of what is now Germany before Germanic populations expanded eastward into
these regions causing assimilation of the Slavic populations and, to a large extent, loss of the Slavic languages
there.
It is with some trepidation that we approach the topic of place names of the formerly Slavic parts of
Germany, since this is one of the best—if not the best—researched geographical areas with respect to
toponyms in the entire world. This strand of research is part of a long tradition of toponymics in Germany, a
tradition which was already established by the middle of the 19th century (Förstemann 1863), and it
flourished particularly within the German Democratic Republic from 1949 to 1990 (Walther 1995), also
continuing vigorously after German reunification (Bily 2002; Eichler 1994-95). While it is challenging to
confront such a well-researched topic, the former Slavic area of Germany is an attractive target for
toponymics because here place names are the best and sometimes the only source of information about
language distributions in past centuries (Eichler & Witkowski 1985b, 51). Slavic groups moved into the thinly
populated area from the 6th century onwards (Bily 2003, 22; Plate 1998, 14–17). In the well-known
classification of Slavic languages into East, West, and South Slavic, these groups belong to West Slavic, but in
the historical sources they are referred to by a bewildering number of different names and are often referred
to as different “tribes” in the literature (Hermann 1985). It appears that there is not a one-to-one match
between ethnicity and language, so ethnic designations are of relatively limited use in a linguistic

ans-names.pitt.edu
ISSN: 0027-7738 (print) 1756-2279 (web) Vol. 73 No. 2, Summer 2025 DOI 10.5195/names.2025.2616
10
NAMES: A JOURNAL OF ONOMASTICS
Søren Wichmann and Lennart Chevallier

classification in this case. Sometimes they are simply designations derived from the area dominated by a
certain tribe (Eichler 1998, 276–277; Eichler & Witkowski 1985a; Hermann 1985, 9).
While we would like to develop a clear, data-driven, well supported, and tractable model for the
erstwhile distribution of Slavic varieties in Germany, it turns out that a blunt pattern-seeking approach will
not capture borders between Slavic varieties because differences are too subtle. Instead, some distributional
differences in toponyms that are due to German dialect differences tend to emerge. We will, however, at least
be able to delineate the extension of the former Slavic territory.
The suffix that is both most widely and most densely distributed is –itz (see figure 7). Some of the cases
of the trailing string -itz belong to -witz, which is also Slavic. They originate in Slavic suffixes -ici/-ovici or -
ica/-ovica (cf. Eichler 1976, 137). Not shown in figure 7, which is restricted to Germany, is the northernmost
extension represented by a handful of placenames in -itse on the southern islands of Lolland and Falster in
Denmark (Thorndal 1963). Outliers to the west mostly pertain to German names ending in a string -itz which
is not identical to the Slavic suffix, such as Titz (North Rhine-Westphalia), Merkenfritz (Hesse) or Olpenitz
(Schleswig-Holstein). But a closer look at each may reveal evidence for some isolated Slavic settlements
outside of the general Slavic area.

Figure 7: Distribution of Names of Populated Places in Germany Ending in -itz

Another typical Slavic suffix is -ow. Its distribution is mapped in figure 8. This originates in a Slavic suffix -ov
or -ava and may derive place names from either common nouns (appellatives) or proper (personal) names.
Typical shapes reflecting -ov/-ava in Low German are -ow and -o, while Middle and High German may show
-au, -a, -e or zero, according to Eichler (1976, 139). Thus, the distribution is somewhat deceptive because the
Slavic suffix is present throughout the “itz-area” in other guises even if the shape -ow does not continue
further south into Sachsen and Bayern. It is an example of German dialectal variation which interferes with
phonological shapes of the attested Slavic toponyms.

ans-names.pitt.edu
DOI 10.5195/names.2025.2616 Vol. 73 No. 2, Summer 2025 ISSN: 0027-7738 (print) 1756-2279 (web)
 NAMES: A JOURNAL OF ONOMASTICS 11

Mapping Place Names

Figure 8: Distribution of Names of Populated Places in Germany Ending in -ow

Interference from German dialectal variation also affects the third of the most common Slavic suffixes,
namely -in. This originates in the Slavic suffixes -in, -ina, -yn, and –ъn, and is reflected as -in or -en in Low
German and -en, -n, -a, -e in Middle and High German according to Eichler (1976, 13). In figure 9 we map the
distribution of -in, with only a little filtering, removing cases where -in belongs to some obviously non-Slavic
names. As in the case of -ow, the distribution of the original Slavic suffix is expected to cover the whole “itz-
area”, but we would have to analyze hundreds of toponyms individually to find the more extended
distribution—maps of -en, -n, -a, or -e without further filtering produce dots all over Germany.

ans-names.pitt.edu
ISSN: 0027-7738 (print) 1756-2279 (web) Vol. 73 No. 2, Summer 2025 DOI 10.5195/names.2025.2616
12
NAMES: A JOURNAL OF ONOMASTICS
Søren Wichmann and Lennart Chevallier

Figure 9: Filtered Distribution of Names of Populated Places in Germany Ending in -in

After the establishment of Slavic-German sister towns due to the eastward spread of German populations,
modifiers such as Groß-/Klein-, Alt-/Neu-, Nieder-/Ober-, and Wendisch-/Deutsch- were applied to
distinguish the two (Eichler and Walther 2001, Vol. 1, xxxiv). In the case of Groß-/Klein- it was the Slavic
town that received the modifier Klein- (Bily 2015, 133; Gringmuth-Dallmer 1990). All pairs of modifiers
except Wendisch- vs. Deutsch- are used in other contexts as well and will not help us to delineate the Slavic
area. In figure 10 we map Wendisch- vs. Deutsch-. Wendisch is a generic term which roughly translates to
“Slavic”. Since we are only interested in the string Deutsch- when it is potentially paired up with -Wendisch,
we do not include states where only Deutsch is found. And in order to better focus on the area of interest we
exclude from the map an unexplained single occurrence of Wendisch- in Baden-Württemberg in southern
Germany. Although thinly distributed, Wendisch- does roughly bring out the contours of the Slavic area.
Additionally, the map shows that Wendisch- can occur both alone and paired with Deutsch- (one pair in
Lower Saxony, one pair in Saxony-Anhalt, at least two pairs in Saxony).

ans-names.pitt.edu
DOI 10.5195/names.2025.2616 Vol. 73 No. 2, Summer 2025 ISSN: 0027-7738 (print) 1756-2279 (web)
 NAMES: A JOURNAL OF ONOMASTICS 13

Mapping Place Names

Figure 10: Wendisch- and Deutsch- in States Where Wendisch- Is Found (Minus Baden-Württemberg)
In figure 11 we combine the maps showing the three most widely occurring Slavic suffixes as they can readily
be identified in German place names, as well as a map of the modifier Wendisch (now without Deutsch). The
area in eastern Germany where the dots are densely distributed corresponds to the former Slavic territory,
whereas the more scattered outliers to the west are either individual incursions or, probably more commonly,
due to strings not pertaining to Slavic suffixes. There are, to our knowledge, not any Slavic suffix strings with
frequencies anywhere near those of -itz, -ow, and -in. A few could be added which are also found throughout
the former Slavic area, but they would be too thinly distributed to change the general picture. An example
is -ehna, which is characteristic of some Slavic place names according to Eichler (1976, 133) and which occurs
widely, albeit sparingly.

ans-names.pitt.edu
ISSN: 0027-7738 (print) 1756-2279 (web) Vol. 73 No. 2, Summer 2025 DOI 10.5195/names.2025.2616
14
NAMES: A JOURNAL OF ONOMASTICS
Søren Wichmann and Lennart Chevallier

Figure 11: Distribution of -in (filtered), -itz, -ow, and Wendisch-

We tried to find isoglosses providing evidence for linguistic subdivisions within the Slavic area. There is a
known major division between Polabian in the north and Sorbian in the south running about midway through
the area. The border roughly coincides with the southernmost extension of the “in-area” in figure 9 (see the
map in Eichler & Witkowski 1985b, 52; and cf. further Bathe et al. 1971; Fischer 1970). But, as we have
already pointed out, the suffix -in does appear further to the south, it just takes on other shapes, apparently
due to the influence of German dialects. Known differences between Polabian and Sorbian are too subtle to be
found in a broad, pattern-seeking approach using modern place names. A major difficulty is that the Slavic
names became blurred by German orthographic conventions as the area was appropriated by these non-
Slavic populations starting around a millennium ago. An even greater difficulty is changing pronunciations,
including many changes due to accommodation to German.
Instead, when we do find diagnostic traits they seem to reflect German dialect features rather than
Slavic ones. For instance, an area quite closely corresponding to Berlin-Brandenburg stands out by exhibiting
hydronyms containing the lexemes Fließ, Pfuhl, and Luch, as shown in figure 12.

ans-names.pitt.edu
DOI 10.5195/names.2025.2616 Vol. 73 No. 2, Summer 2025 ISSN: 0027-7738 (print) 1756-2279 (web)
 NAMES: A JOURNAL OF ONOMASTICS 15

Mapping Place Names

Figure 12: Distribution of the Lexemes Pfuhl, Luch, and Fließ in Hydronymic Compounds
Fließ means ‘stream, river, canal’. It comes from Proto-Germanic *fleuta, according to Greule (2014, 150),
who does not comment on the distribution. Pfuhl means ‘pond, puddle’. It is related to Low German Pool,
Puhl, Paul, Puel. In the 18th century larger bodies of water could be referred to as Pfuhl (cf. Greule 2014,
405). Luch means ‘swamp’. Its origin is West Slavic, ultimately reflecting Proto-Slavic *lǫg ‘meadow’ (Debus
& Schmitz 2002), and it apparently predates the split between Polabian and Sorbian. Thus, the distribution of
these lexical items seem to mainly be characteristic of German dialect differences. To claim that their
distributions somehow relate to a Slavic substrate influence would be speculative.

Discussion
Two case studies were presented. In the first one, on Xincan, we demonstrated how maps drawing upon
GeoNames data inform proposals concerning place names that have been taken up again repeatedly since
they were first proposed around half a century ago (Campbell 1978; Thompson 1970). Thompson proposed
that the ending –agua in place names of Guatemala, Honduras, El Salvador, and Nicaragua could be
connected to a specific ethnic group, and Campbell proposed that some of these names belong to the Xincan
family of languages. In both cases the proposals are made in the absence of maps. When mapping the names
discussed, Thompson’s proposal is brought out more clearly, and the widely scattered distribution of names
in –agua casts doubts on the viability of a hypothesis linking them to a certain language or language groups.
Campbell’s proposal is better supported, but this proposal also benefitted from mapping, which brought out
the need for some corrections.
In the second case study, we exercised the GeoNames data in the context of a better-studied toponymic
research area, namely that of Slavic place names in Germany. The package is only as good as the data for
which it provides an interface, which consists of modern place names. This puts severe restrictions on a
linguistic investigation where the shapes of toponyms as they first appear in attestations from the Middle
Ages would be crucial. Nevertheless, even for a study area privileged by masses of literature the software is
useful, especially as a companion for quickly producing distributional maps of morphological and lexical
elements brought up in the literature.

ans-names.pitt.edu
ISSN: 0027-7738 (print) 1756-2279 (web) Vol. 73 No. 2, Summer 2025 DOI 10.5195/names.2025.2616
16
NAMES: A JOURNAL OF ONOMASTICS
Søren Wichmann and Lennart Chevallier

Conclusion
In this paper we have illustrated uses of the new R package ‘toponym’, which is designed to select, display,
and manipulate data from GeoNames. The package is a research tool particularly useful in a pattern-seeking
approach to toponymy while the complementary particularistic approach that concentrates on individual
elements and their histories requires additional tools like etymological dictionaries and archives. Our case
studies on Xincan toponyms and Slavic toponyms in Germany were intended not only to illustrate the utility
but also the limitations of the GeoNames data.

Notes
1 GNIS is hosted at https://www.usgs.gov/tools/geographic-names-information-system-gnis
2 https://github.com/Sokiwi/ToponymPaper/
3 https://www.geonames.org/datasources/
4 http://download.geonames.org/export/dump/readme.txt
5There is in a sense one name ‘too many’ because Río Sansur is repeated in our output. We removed the
westernmost occurrence from the data.
6 The Motagua River runs just below 15° N.
7A more in-depth study of the earlier distribution of Xincan should also take into account records from the
colonial period (cf. Sachse 2010, 42–47).

Acknowledgements
We thank Bernard Comrie for comments on the manuscript.

Funding Details
This work was supported by the Deutsche Forschungsgemeinschaft (German Research Foundation) under
Germany’s Excellence Strategy, Grant number EXC 2150 390870439.

References
Acheson, Elise, Stefano De Sabbata, and Ross S. Purves. 2017. “A Quantitative Analysis of Global Gazetteers:
Patterns of Coverage for Common Feature Types”. Computers, Environment and Urban Systems 64:
309–320.
Algeo, John and Katie Algeo. 2000. “Onomastics as an Interdisciplinary Study”. Names 48, no. 3/4: 265–274.
Azaryahu, Maoz and Arnon Golan. 2001. “(Re-)naming the Landscape: The Formation of the Hebrew Map of
Israel 1949–1960”. Journal of Historical Geography 27, no. 2: 178–195.
Baca, Keith A. 2007. Native American Place Names in Mississippi. Jackson: University Press of Mississippi.
Bathe, Max, Reinhard E. Fischer, and Gerhard Schlimpert. 1971. “Die sorbisch-polabische Sprachgrenze und
das Problem der Entnasalierung im 10. bis 12. Jahrhundert” [The Sorbian-Polabian Language Border
and the Problem of Denasalization in the 10th to the 12th Centuries]. Forschungen zur slawischen und
deutschen Namenkunde [Research on Slavic and German Onomastics] (= Deutsche Akademie der
Wissenschaften. Veröffentlichungen des Instituts für Slawistik, 55), edited by Teodolius Witkowski, 32–
39. Berlin: Akademie-Verlag.
Bily, Inge. 2002. “Zu einigen namenkundlichen Projekten an der Sächsischen Akademie der Wissenschaften
zu Leipzig (1992–2003)” [On some Onomastic Projects in the Saxonian Academy of Sciences of Leipzig
(1992–2003)]. Onoma 36: 7–24.

ans-names.pitt.edu
DOI 10.5195/names.2025.2616 Vol. 73 No. 2, Summer 2025 ISSN: 0027-7738 (print) 1756-2279 (web)
 NAMES: A JOURNAL OF ONOMASTICS 17

Mapping Place Names

Bily, Inge. 2003. “Siedlungsnamen und slavische Besiedlung” [Settlement Names and Slavic Settlement].
Słowiańska onomastyka [Slavic Onomastics], Vol. 2, edited by Ewa Rzetelska-Feleszko, Aleksandra
Cieślikowa, and Jerzy Duma, 21–22. Warszawa-Kraków: Towarzystwo Naukowe Warszawskie.
Bily, Inge. 2015. “Deutsch-slawischer Siedlungs- und Sprachkontakt im Gebiet zwischen Saale und Neiße –
vorgestellt an ausgewählten Ortsnamen (Siedlungsnamen)” [German-Slavic Settlement and Language
Contact in the Area between Saale and Neiße – Presented through Selected Place Names (Settlement
Names)]. Linguistica 55, no. 1: 131–147.
Bright, William. 1998. 1500 California Place Names. Their Origin and Meaning. A Revised Version of 1000
California Places Names, by Erwin G. Gudde, 3rd ed. Berkeley: University of California Press.
Campbell, Jon C. 1991. “Stream Generic Terms as Indicators of Historical Settlement Patterns”. Names 39,
no. 4: 333–366.
Campbell, Lyle. 1978. “Quichean Prehistory: Linguistic Contributions”. Papers in Mayan Linguistics, edited
by Nora C. England, 25–54. Columbia: Museum of Anthropology, University of Missouri.
Campbell, Lyle. 1998. “The Linguistic Prehistory of Guatemala”. The Life of Language: Papers in Honor of
William Bright, edited by Jane Hill, P. J. Mistry, and Lyle Campbell, 183–192. Berlin/New York:
Mouton de Gruyter.
Campbell, Lyle. 2013. Historical Linguistics: An Introduction, 3rd ed. Edinburgh: Edinburgh University
Press.
Campbell, Lyle. 2022. “Las lenguas indígenas de El Salvador” [The Indigenous Languages of El Salvador].
Talk presented (online) at Tamachtiluyan Alan R. King, Escuela Virtual del Náhuat de El Salvador,
October 21, 2022.
Chen, Xixi, Tao Hu, Fu Ren, Deng Chen, Lan Li, and Nan Gao. 2014. “Landscape analysis of geographical
names in Hubei Province, China”. Entropy 16: 6313–6337. https://doi.org/10.3390/e16126313
Chevallier, Lennart, and Søren & Wichmann. 2024. “A Note on the ‘toponym’ R Package: A Practical
Introduction”. Names 72, no. 3: 77-83.
Coates, Richard. 2012. “Onomastics”. The Encyclopedia of Applied Linguistics, edited by Carol A. Chapelle.
Blackwell Publishing Ltd.
Cohen, Saul B. and Nurit Kliot. 1992. “Place-names in Israel's Ideological Struggle over the Administered
Territories”. Annals of the Association of American Geographers 82, no. 4: 653–680.
Copley, Gordon. 1988. Early Place-Names of the Anglian Regions of England. BAR British Series, 185.
Oxford: B.A.R.
Debus, Friedhelm and Antje Schmitz. 2002. “(Mikro-)Toponyme im slawisch-deutschen Kontaktgebiet
Norddeutschlands” [(Micro) Toponyms in the Slavic-German Contact Area of Northern Germany].
Onoma 36: 51–70.
DeRosier, Jr., Arthur H. 1970. The Removal of the Choctaw Indians. New York: Harper & Row.
Eichler, Ernst. 1976. “Sprachkontakte im Lichte der Onomastik” [Language Contact in the Light of
Onomastics]. Onoma 20: 128–141.
Eichler, Ernst. 1994–95. “Namenkundliche Forschungen in Leipzig 1990-1995” [Onomastic Research in
Leipzig 1990–1995]. Onoma 32: 23–25.
Eichler, Ernst. 1998. “Die westlichste Peripherie des slavischen Sprachgebietes” [The Westernmost Periphery
of the Slavic Language Area]. Zeitschrift für Slavische Philologie 57, no. 2: 269–280.
Eichler, Ernst and Hans Walther. 2001. Historisches Ortsnamenbuch von Sachsen [Historical Book of Place
Names of Saxony]. Vol. 1–3. Berlin: Akademie Verlag.
Eichler, Ernst and Teodolius Witkowski. 1985a. “Namen der Stämme und Landschaften” [Names of the
Tribes and Landscapes]. Die Slawen in Deutschland. Geschichte und Kultur der slawischen Stämme
westlich von Oder und Neiße vom 6. bis 12. Jahrhundert [The Slavs in Germany. History and Culture of
the Slavic Tribes West of Oder and Neiße from the 6th to the 12th Centuries], edited by Joachim
Hermann, 10–14. Berlin: Akademie-Verlag.
Eichler, Ernst and Teodolius Witkowski. 1985b. “Sprachen und Dialekte” [Languages and Dialects]. Die
Slawen in Deutschland. Geschichte und Kultur der slawischen Stämme westlich von Oder und Neiße
vom 6. bis 12. Jahrhundert [The Slavs in Germany. History and Culture of the Slavic Tribes West of
Oder and Neiße from the 6th to the 12th Centuries], edited by Joachim Hermann, 47–54. Berlin:
Akademie-Verlag.

ans-names.pitt.edu
ISSN: 0027-7738 (print) 1756-2279 (web) Vol. 73 No. 2, Summer 2025 DOI 10.5195/names.2025.2616
18
NAMES: A JOURNAL OF ONOMASTICS
Søren Wichmann and Lennart Chevallier

Endo, Mitsuaki. 2021. “Geographical Distribution of Certain Toponyms in the Samguk Sagi”. Anthropological
Science 129, no. 1: 35–44.
Fischer, Reinhard E. 1970. “Zur sorbisch-polabischen Sprachgrenze zwischen Elbe und Spree” [On the
Sorbian-Polabian Language Border between Elbe and Spress]. Beiträge zum Slawischen Onomastischen
Atlas. Theodor Frings zum Gedächtnis [Contributions to the Slavic Onomastic Atlas. In Remembrance
of Theodor Frings] (= Abhandlungen der Sächsischen Akademie der Wissenschaften zu Leipzig, Phil.-
hist. Kl. 61/2), edited by Rudolf Fischer and Ernst Eichler, 190–221. Berlin: Akademie-Verlag.
Förstemann, Ernst Wilhelm. 1863. Die deutschen Ortsnamen [The German Place Names]. Nordhausen: Ferd.
Förstemann’s Verlag.
Fuchs, Stephan. 2015. “An integrated approach to Germanic place names in the American Midwest”. The
Professional Geographer 67, no. 3: 330–341.
GeoNames. 2023. Accessed December 7, 2023. http://www.geonames.org/
Greule, Albrecht. 2014. Deutsches Gewässernamenbuch. Etymologie der Gewässernamen und der
dazugehörigen Gebiets-, Siedlungs- und Flurnamen [Book of German Hydronyms. Etymologies of the
Hydronyms and Related Names of Areas, Settlements, and Plots of Land]. Berlin: De Gruyter.
Gringmuth-Dallmer, Eike. 1990. “Deutsch und Wendisch - Groß und Klein. Zur Siedlungsgeschichtlichen
Aussage von Ortsnamen mit unterscheidenden Zusätzen in der Mark Brandenburg” [Deutsch and
Wendisch – Big and Small. On Information about Settlement History Contained in Place Names with
Different Additions in Mark Brandenburg]. Onomastica Slavogermanica XIX. Hans Walther zum 70.
Geburtstag [Onomastica Slavogermanica XIX. For Hans Walther on His 70th Birthday]. ( =
Abhandlungen der Sächsischen Akademie der Wissenschaften zu Leipzig, Philologisch-Historische
Klasse, Band 73, Heft 2), edited by Ernst Eichler, 77–89. Berlin: Akademie-Verlag Berlin.
Hermann, Joachim. 1985. “Wohnsitze der slawischen Stämme und Stammesgruppen” [Dwelling Places of the
Slavic Tribes and Tribal Groups]. Die Slawen in Deutschland. Geschichte und Kultur der slawischen
Stämme westlich von Oder und Neiße vom 6. bis 12. Jahrhundert [The Slavs in Germany. History and
Culture of the Slavic Tribes West of Oder and Neiße from the 6th to the 12th Centuries], edited by
Joachim Hermann, 7–10. Berlin: Akademie-Verlag.
Hunn, Eugene. 1996. “Columbia Plateau Indian Place Names: What Can They Teach Us?” Journal of
Linguistic Anthropology 6, no. 1: 3–26.
Lehmann, Friedrich. 1999. Der Wandel der Ortsnamen in den ehemals deutsch besiedelten Gebieten der
Tschechoslowakei. Gezeigt an über 300 Beispielen ausgewählter ehemaliger Landkreise [Changes in
Place Names in the Areas of Czechoslovakia Formerly Settled by Germans. Shown Through More Than
300 Examples of Selected Former Districts]. ( = Scripta Slavica, Band 6). Marburg/Lahn: Biblion Verlag.
Perono Cacciafoco, Francesco and Francesco Cavallaro. 2023. Place Names: Approaches and Perspectives in
Toponymy and Toponomastics. Cambridge: Cambridge University Press.
Petersen, Niels Matthias. 1833. “Bemærkninger om danske og norske stedsnavnes oprindelse og forklaring”
[Remarks on the Origins of and Explanations for Danish and Norwegian Place Names]. Nordisk
tidsskrift for oldkyndighed [Nordic Journal of the Study of Antiquity] 2.
Plate, Christa. 1998. “Die Besiedlungsgeschichte des Landes Ruppin” [The Settlement History of the Ruppin
District]. Brandenburgisches Namenbuch. Teil 11: Die Ortsnamen des Landes Ruppin [The Book of
Names of Brandenburg. Part 11: The Place Names of the Ruppin District], edited by Elżbieta Foster, 9–
35. Weimar: Verlag Hermann Böhlaus Nachfolger.
R Core Team. 2022. R: A Language and Environment for Statistical Computing. R Foundation for Statistical
Computing, Vienna, Austria. https://www.R-project.org/
Radding, Lisa and John Western. 2010. “What’s in a Name? Linguistics, Geography, and Toponyms”. The
Geographical Review 100, no. 3: 394–412.
Read, William A. 1994. Indian Place Names in Alabama. Tuscaloosa and London: The University of Alabama
Press.
Sachse, Frauke. 2010. Reconstructive Description of Eighteenth-century Xinka Grammar. Utrecht: LOT.
Tent, Jan. 2015. “Approaches to Research in Toponymy”. Names 63, no. 2: 65–74.
TGN. 2023. “Getty Thesaurus of Geographic Names: About the TGN”. Accessed November 7, 2023.
http://www.getty.edu/research/tools/vocabularies/tgn/about.html
Thompson, J. Eric S. 1970. Maya History and Religion. Norman: University of Oklahoma Press.

ans-names.pitt.edu
DOI 10.5195/names.2025.2616 Vol. 73 No. 2, Summer 2025 ISSN: 0027-7738 (print) 1756-2279 (web)
 NAMES: A JOURNAL OF ONOMASTICS 19

Mapping Place Names

Thorndahl, William. 1963. “Slawische Ortsnamen in Dänemark” [Slavic Place Names in Denmark]. Slawische
Namenforschung. Vorträge auf der II. Arbeitskonferenz der Onomastischen Kommission beim
Internationalen Slawistenkomitee in Berlin vom 17. bis 20. Oktober 1961 [Slavic Onomastics.
Presentations at the II. Working Conference of the Onomastic Commission of the International
Committee of Slavists from the 17th to the 20th of October 1961], edited by Teodolius Witkowski, 56–70.
Berlin: Akademie-Verlag.
Walther, Hans. 1995. “Namenforschung in der ehemaligen Deutschen Demokratischen Republik (DDR)”
[Onomastics in the Former German Democratic Republic (GDR)]. Namenforschung / Name Studies /
Les noms propres. Ein internationales Handbuch zur Onomastik / An International Handbook of
Onomastics / Manuel international d'onomastique. = Handbücher zur Sprach- und
Kommunikationswissenschaft / Handbooks of Linguistics and Communication Science / Manuels de
linguistique et des sciences de communication. Vol 11.1, edited by Ernst Eichler, Gerold Hilty, Heinrich
Löffler, Hugo Steger, and Ladislav Zgusta, 102–124. Berlin: Walter de Gruyter.

Notes on Contributors:
Søren Wichmann is a Postdoctoral Fellow at Kiel University (since 2021). He has previously held positions
at University of Copenhagen, Leiden University, Kazan Federal University, and Max Planck Institute for
Evolutionary Anthropology. He specializes in historical linguistics, descriptive linguistics, language typology,
quantitative methods, and Mesoamerican languages and writing systems.

Lennart Chevallier holds an MA in Language, Variation, and Political Science at Kiel University. He is
currently a PhD candidate and is a part of the “DiaSAL” project at Kiel University.

Correspondence to: Søren Wichmann, ROOTS, CAU, Leibnizstraße 3, 24118 Kiel, Germany; Email:
wichmannsoeren@gmail.com

ans-names.pitt.edu
ISSN: 0027-7738 (print) 1756-2279 (web) Vol. 73 No. 2, Summer 2025 DOI 10.5195/names.2025.2616