Name Constructions In Gaulish

by Tangwystyl verch Morgant Glasvryn (Heather Rose Jones,
© 2001 Heather Rose Jones, all rights reserved


The following is a survey of name structures found in Gaulish inscriptions of a variety of types, locations, and dates. The major sources are:

I have not dealt with Celt-Iberian data here (the Celtic language found on the Iberian peninsula), partly because the sources are significantly more difficult to work with, partly because it is distinct enough from Gaulish to be worth treating separately.

Unless otherwise identified, the following examples are in the Gallo-Latin group. This article deals primarily with native Gaulish naming practices and does not include the much vaster amount of material on Gauls who bore Roman-style names (which came to predominate as the region became more Romanized during and after the first century CE).

Only a small percentage of the known Gaulish names occur in the structures covered here. Two much more extensive sources for Gaulish given names are Evans and Whatmough (see the references below), although both need to be used with care as they cover a great deal of time and space.

Men's and women's names use the same types of structures in general, with a few variants that may be due to different usage or simply to the vagaries of the data. Rather than separating the two into different sections, I've treated them together to show the parallels.

The major theme in name structures is the combination of an idionym (the given name or equivalent identifying the individual) plus some sort of relationship identification. (I generally try to stay away from less familiar terms like "idionym", but I've tried to use it here to distinguish the given name of the person being identified from other given names that occur in the reference; for example, in patronyms.) Occasionally, other types of identification may also be present, or may substitute for the patronym. The relationship byname may be simply the genitive (possessive) form of the parent's given name; it may be an adjective with a possessive sense derived from that element (similar to the <X-ius> forms of Roman nomina); it may be an adjectival form (with genitive sense) made from the parent's given name with an added suffix <-ac-> before the inflection; or it may add the suffix masculine <-(i)gnos>, feminine <-(i)gna> (meaning roughly "offspring") to the parent's given name. Most typically, the relationship byname comes after the idionym, but the reverse order is sometimes seen.

There are some significant differences between the repertoire of men's and women's bynames in the available data, but this is probably skewed by the nature of the richest source for women's names: a magical text produced by a women's religio-magical cult. Many of the women named in this document are identified by their relationships to other women (including, in some cases, other women named in the document), but it may be a mistake to assume that the patterns seen here are typical for women's names. (At least one scholar is of the opinion that the relationships -- and perhaps even the names themselves -- are of cultic rather than everyday significance. [Koch in Meid et al. 1996]) In addition to identifying the women by their relationships to other women, the names in this document include separate words indicating the nature of the relationship (daughter, mother, spouse), a pattern I did not find in men's names (although I would be unsurprised to turn up an example of a man's patronym with an explicit marker).

In addition to bynames indicating relationships, there are a very few examples of descriptive bynames or ones indicating location of residence or tribal affiliation, or where the nature of the byname is unclear.

(Format note: the actual cited version of the name is given first, transcribed very closely into the Roman alphabet if necessary. Following is a normalized version that puts the name into the nominative, if necessary, and adjusts the spelling to a more standard form. After that, I give the nominative forms of the name elements, with notes on the inflectional class. An acute accent in a normalized form represents a macron (long-mark) -- these would not normally have been indicated in the original, but affect pronunciation.)

Idionym + Genitive Patronym

To form the patronym, the father's name is simply put into the genitive case of the appropriate declension.

Vectit[...] Biraci (m) [Lambert p.92]
normalized: Vectitos Biraci
nominative: Vectitos (m) o-stem, Biracos (m) o-stem
Martialis Dannotali (m) [Lambert p.98]
normalized: Martialis Dannotali
nominative: Martialis (m) i-stem (but probably a Latin borrowing), Dannotalos (m) o-stem
Doiros Segomari (m) [Lambert p.135]
normalized: Doiros Segomari
nominative: Doiros (m) o-stem, Segomaros (m) o-stem
Banona Flatucias (f) [Koch p.3, Meid et al. p.75]
normalized: Banona Vlatucias
nominative: Banona (f) a-stem, Vlatuca (f) a-stem

Genitive Patronym + Idionym

This is exactly the same as the previous construction, but with the order of the elements reversed. Quite probably, in the current example, this is due to the listing of two sons of the same father (i.e., "X's sons: Y, Z") and may not reflect normal name format.

Esanekoti Anareuiseos, Tanotalos (both m) (Gallo-Etruscan) [Lambert p.72]
normalized: Essandecotti Andareuiseos, Dannotalos (i.e., Essandecotti Andareuiseos, Essandecotti Dannotalos)
nominative: Essandecottos (m) o-stem, Andareuiseos (m) o-stem, Dannotalos (m) o-stem

Idionym + Genitive-Adjective Patronym

To form the patronym, the stem of the father's name is given the suffix <-I-> and then the appropriate o-stem declensional ending to match the idionym. Because the patronym is behaving grammatically as an adjective, it must agree in case and gender with the idionym. (This is crucial to remember with women's patronyms of this type.) See the appendix for information on how to identify what form to use for the stem.

ourittakos elouskonios (m) (Gallo-Greek) [Lambert p.83]
normalized: Vrittakos Eluskonios
nominative: Vrittakos (m) o-stem, Eluskú (m) n-stem

Bimmos litoumareos (m) (Gallo-Greek) [Lambert p.83]
normalized: Bimmos Litumarios
nominative: Bimmos (m) o-stem, Litumaros (m) o-stem

Frontu Tarbeisonios (m) [Lambert p.92]
normalized: Frontú Tarbeisonios
nominative: Frontú (m) n-stem, Tarbeisú (m) n-stem

ouenitoouta kouadrounia (f) (Gallo-Greek) [Lambert p.31]
normalized: Venitoutá Quadrunia
nominative: Venitouta (f) a-stem, Quadrú (m) n-stem

Eskengai Blandouikuniai (f) (Gallo-Greek)
normalized: Eskenga Blandouikunia
nominative: Eskenga (f) a-stem, Blandouikú (m?) n-stem

Idionym + Genitive-Adjective Patronym with <-(I)AC->

Another way of forming an adjective from the father's name adds the suffix <-(I)AC-> to the stem, and then the appropriate o-stem or a-stem inflectional endings to correspond with the gender of the idionym. (This suffix is cognate with a common adjective-forming suffix found in later Celtic languages.)

kabiros uindiakos (m) (Gallo-Greek) [Lambert p.82]
normalized: Cabiros Vindiacos
nominative: Cabiros (m) o-stem, Vindios (m) o-stem

...artaros illanouiakos (m) (Gallo-Greek) [Lambert p.86]
normalized: [...]artaros Illianuiacos
nominative: [...]artaros (m) o-stem, Illianus (m) u-stem

Idionym + Patronym with <-(I)GNOS>

The most common patronymic formation (in the data I found) adds the suffix <-(I)GN-> to the stem of the parent's name, followed by either an o-stem or a-stem inflectional suffix to match the gender of the idionym. (In both this and the previous construction, the "i" appears in most phonetic contexts: when the stem ends in a consonant, and evidently sometimes when the stem ends in "u" as well.) The suffix is a reduced form of the root <GEN->, cognate with Latin gens, and having the sense "family, offspring".

missukos siluknos (m) (Gallo-Greek) [Lambert p.82]
normalized: Missukos Silugnos
nominative: Missukos (m) o-stem, Silus (m) u-stem

Aneuno Oclicno (m) [Lambert p.94]
normalized: Aneunos Oclignos
nominative: Aneunos (m) o-stem, Oclos (m) o-stem

Lugurix Aneunicno [Lambert p.94]
normalized: Lugurix Aneunignos?
nominative: Lugurix (m) k-stem, Aneunos (m) o-stem

Iccauos Oppianicnos (m) [Lambert p.96]
normalized: Iccauos Oppianignos
nominative: Iccauos (m) o-stem, Oppianos (m) o-stem

Andecamulos Toutissicnos (m) [Lambert p.97]
normalized: Andecamulos Toutissignos
nominative: Andecamulos (m) o-stem, Toutissos (m) o-stem

Bratronos Nantonicn[os] (m) [Lambert p.105]
normalized: Bratronos Nantonignos
nominative: Bratronos (m) o-stem, Nantonos (m) o-stem

seuerim tertionicnim (f) [Koch p.3, Meid et al. p.75]
normalized: Severa Tertionigna
nominative: Severa (f) a-stem, Tertiú (m) n-stem

Diligentim Vlationicnom (f) [Meid et al. p.75]
normalized: Diligenta Vlationigna
nominative: Diligenta (f) a-stem, Vlatiú (m) n-stem(?)

(there is also an inscription where this format is paralleled by a Latin formula: IDIONYM + GENITIVE PATRONYM + F[ILIUS])

Ateknati Trutikni (m) (Gallo-Etruscan) [Lambert p.74]
normalized: Ategnatos Drutignos
nominative: Ategnatos (m) o-stem, Drutos (m) o-stem
compare Latin: Ategnati Druti f[ilius] (also Ategnatei Drutei f[ilius])

Koisis Trutiknos (m) (Gallo-Etruscan) [Lambert p.74], brother of the preceding
normalized: Coisis Drutignos
nominative: Coisis (m) i-stem, Drutos (m) o-stem
compare Latin: Coisis Druti f[ilius] (also Coisis Drutei f[ilius])

Patronym with <-GNOS> + Idionym

The same construction, but with the order of elements reversed. In this case, there is no particular reason in context for this order and it may simply be a rare but possible alternative.

Tanotaliknoi Kuitos (m) (Gallo-Etruscan) [Lambert p.72]
normalized: Dannotalignoi Quintos
nominative: Quintos (m) o-stem, Dannotalos (m) o-stem

Idionym + duxtir + Genitive Metronym

This and the following two constructions are found only in women's names and only in a single source (that I've been able to identify so far). There seems no inherent reason that men could not have a patronymic construction with an overt word meaning "son" -- there are certainly Latin versions of Gaulish names that use filius -- but I have yet to find one at this time.

In this and the following two constructions, the idionym and relationship word appear in the nominative case, followed by the relative's name in the genitive case.

Aia duxtir Adiegias (f) [Meid et al. p.75]
normalized: Aia duxtir Adiegias
nominative: Aia (f) a-stem, Adiega (f) a-stem

Idionym + matir + Genitive Daughter's-Name

Here the women are identified as the mother of another woman -- in two cases, the daughter also appears identified as the daughter of her mother. I find it somewhat unlikely that this would be a normal naming construction, but I may be prejudiced by the types of relationship bynames typically found in Europe in the medieval period. There are certainly cultures (e.g. medieval and later Arabic culture) that have name-constructions that identify a woman as the mother of her child (normally of a son).

Potita matir Paullias (f) [Meid et al. p.75]
normalized: Potita matir Paullias
nominative: Potita (f) a-stem, Paulla (f) a-stem

Adiega matir Aiias (f) [Meid et al. p.75]
normalized: Adiega matir Aiias
nominative: Adiega (f) a-stem, Aia (f) a-stem
compare with <AIA DUXTIR ADIEGIAS> above, likely the same individuals

Aucitionim materem Potiti (f) [Meid et al. p.75]
normalized: Auctiona matir Potiti[as?]
nominative: Auctiona (f) a-stem, Potita (f) a-stem

Vlatucia matir Banonias (f) [Meid et al. p.75]
normalized: Vlatucia matir Banonias
nominative: Vlatucia (f) a-stem, Banona (f) a-stem
compare with <BANONA FLATUCIAS>, likely the same individuals

Idionym + dona + Genitive?/Instrumental? Given-Name

The same text that includes the mother-daughter relationships also identifies several women as dona of another individual. The reading is not at all certain, further complicated by the unusual case-form of the following name. Lambert (p.53) suggests interpreting the word as "nurse", with the following name in an archaic instrumental/associative plural case, with an overall sense of "nurse to the children of So-and-so". Schmidt (in Meid et al. p.29) interprets dona as a borrowing of Latin domina with a spousal sense, although this leaves unanswered questions about the case of the supposed spouse's name. One of the attractions of Lambert's hypothesis is that it is interpretable as retaining the focus on relationships between women in this text -- three of the four stems found after dona occur in the same text as women's idionyms (Potitius << Potita, Paullius << Paulla, Banonus << Banona; only Primius does not have a corresponding Prima elsewhere in the text). The other three examples are listed in the section on multiple bynames.

Paulla dona Potitius (f) [Meid et al. p.75]
normalized: Paulla dona Potitius
nominative: Paulla (f) a-stem, Potita (f) a-stem or Potitos (m) o-stem

Non-Patronymic Constructions

There isn't much data on what sorts of bynames other than patronyms were used. There are a few examples of locative or ethnic bynames. There are also a couple of examples where the meaning and even the nature of the byname is unclear.

Balaudui Makkariui (m) (Gallo-Greek) [Lambert p.82]
normalized: Balaudos Makkarios
nominative: Balaudos (m) o-stem, either Makkarios io-stem or Makkaros o-stem
It's unclear whether this is a genitive-adjective patronym based on Makkaros, or some unknown adjectival byname from Makkarios.

Sacer Peroco[...?] (m) [Lambert p.95]
normalized: Sacer Peroco[...?]
nominative: Sacer a borrowed Latin name (m) r-stem, possibly Petrocorios(?) o-stem
Lambert discusses several possibilities for the byname (hampered by the possible erosion of letters at the end of the word) and suggests an ethnic byname Petrocorios (i.e., from Perigord). As such it would take either an o-stem or a-stem case form as appropriate to the idionym.

Boudilatis Lemisunia (f) [Lambert p.92]
normalized: Boudilatis Lemisunia
nominative: Boudilatis (f) i-stem, Lemisó n-stem
"From Lemiso." The locative byname is an adjectival formation and so agrees in gender with the idionym.

Licnos Contextos (m) [Lambert p.96]
normalized: Licnos Contextos
nominative: Licnos o-stem, Contextos presumably o-stem but the meaning is unknown.

Combinations Of Bynames

It is also rare for a person to appear identified by more than one type of byname. The following show various combinations. If Gaulish names follow the same element-order as we find in later medieval Celtic names, it is no accident that we see a pattern: idionym > descriptive byname > blood relation byname > other relation byname > locative or ethnic byname.

segomaros ouilloneos tooutious namausatis (m) (Gallo-Greek) [Lambert p.84]
normalized: Segomaros Villonios Toutiús Nemausatis i.e., "Segomaros [son] of Villú, of [the] tribe of Nemausus (Nimes)"
nominative: Segomaros o-stem, Villú n-stem, toutios (m) o-stem (presumably the feminine equivalent would be an a-stem <TOUTIá>), Nemausus (m?) u-stem but the form in the name is the derived Nemausatis, an i-stem noun.
In addition to an adjectival patronym, this man has a locative/ethnic byname, using the phrase "toutiús Nemausatis", as Lambert suggests "associated with the people [of] Nimes". I'm uncertain how to interpret the specific grammatical forms, but it appears that a feminine form of this ethnic byname would be identical to the masculine one.

Severa duxtir Valentos dona Paullius (f) [Meid et al. p.75]
normalized: Severa duxtir Valentos dona Paulius
nominative: Severa (f) a-stem, Valens nt-stem (but probably a borrowing from Latin), Paula (f) a-stem or Paulos (m) o-stem
The overall format for names in this source calls for Valentos to be a feminine genitive, but the form is wrong for that. (However Vallentas would fit.) See the discussion above on dona -- depending on the interpretation, the name underlying Paullius could be masculine or feminine. The overall format is presumably two independent relational bynames, both modifying the idionym.

Potita dona Primius ... Abesias (f) [Meid et al. p.75]
normalized: Potita dona Primius ... Abesias
nominative: Potita (f) a-stem, Prima (f) a-stem or Primos (m) o-stem, Abesa (f) a-stem
The inscription is damaged and the relationship with Abesias is not clear. See above for the discussion of the uncertainty of dona bynames.

Rufena Casta dona Banonus (f) [Meid et al. p.75]
normalized: Rufena Casta dona Banonus
nominative: Rufena (f) a-stem, Casta (f) a-stem, Banona (f) a-stem or Banonos (m) o-stem
Casta is presumably an adjectival byname. See above for the discussion and uncertainties of dona bynames.

Stems and Inflections

(This section will be easier to follow if you've ever studied Latin or Greek or a similar language.) Gaulish nouns and adjectives, in addition to coming in different masculine and feminine forms, belong to an inflectional class (identified as X-stem) that determines what the form of the stem will be (the thing to which endings are added) and which set of endings it takes for the different cases.

In the most common noun classes -- the masculine o-stem and feminine a-stem nouns -- the stem is formed by removing the inflectional suffix (either <-OS> or <-A> respectively). In other classes, the relationship may not be as straightforward. N-stem nouns (both masculine and feminine) end in <-U> (long or short) in the nominative case, but this changes to <-ON> for all other cases and inflectional suffixes are added to this. K/g-stem nouns end in <-X> in the nominative (i.e., <-KS>) and in either <-K-> or <-G-> for the other cases. The stem (i.e., the form ending in the "combining stem") is the form of the name to which the other suffixes discussed above are added.

The cases indicate what grammatical function a noun or adjective is serving. The "basic" form is the nominative -- and this is the form that the idionym will appear in. Under normal circumstances, the only other case needed for personal names is the genitive, which indicates possession. (In the data above, names may appear in other cases because of the context of the original inscription.) The following table shows these two inflectional endings for each of the declensions found in the names above, with representative names showing how they work.

NominativeGenitiveCombining stem
o-stem -os -i-
a-stem -a -(i)as-
u-stem -us, -eos -u-
i-stem -is -eos -i-
Consonant stems (have in common a consonant at the end of the stem that may not appear in the nominative)
n-stem -oneos -on-
nt-stem -ns -ntos -nt-
k/g-stem -x -ges -g-

Examples (some forms are hypothetical rather than being found in the data)


1. Evans, D. Ellis. 1967. Gaulish Personal Names. Oxford University Press.

2. Koch, John T. ed. 1995. The Celtic Heroic Age. Celtic Studies Publications, Malden.

3. Lambert, Pierre-Yves. 1995. La Langue Gauloise. Editions Errance, Paris.

4. Meid, Wolfgang et al. 1996. Die Grösseren Altkeltischen Sprachdenkmäler. Institut für Sprachwissenschaft, Innsbrück.

5. Whatmough, Joshua. 1970. The Dialects of Ancient Gaul. Harvard University Press, Cambridge.

HTML formatting by Wendi Dunlap, 1 July, 2001.