The Incredible Agglutinative Power of South Caucasian (Kartvelian) Languages

When we hear of agglutination, we think of crowds, cells or grains, but certainly not of languages. However, languages too can agglutinate things, and in a surprisingly complex way!
If a language is agglutinative, it means it can string together bits of significant grammatical information into the roots of words, in the form of affixes (prefixed if in front of a root; suffixed if at the end; infixed if inside; or circumfixed if around). While languages like English, French or German do have prefixes and suffixes (e.g. English play-s, French re-tourn-ait, German auf-nehm-en-d), this is not enough to consider these languages agglutinative because on the one hand, they can add only a limited number of affixes, and on the other hand these affixes can express multiple bits of grammatical information at the same time (e.g. in English, the -s in (he) plays expresses both the present tense and the third person singular).
In a truly agglutinative language, first, each affix must contain only one grammatical information, and secondly, – this condition following from the first – the number of attachable affixes must be greater than one or two.
Kartvelian languages (Georgian, Mingrelian, Laz, and Svan – also called South Caucasian as they constitute of the autochthonous language families found in the Caucasus) present clear examples of linguistic agglutination. In these languages, it is not uncommon to find half a dozen affixes in one word, in the four positions mentioned above. Thus, even the most common words of everyday language can be broken down into a significant number of affixes.
For example, the Svan noun qan-är-s ‘to oxen’ contains two suffixes added to the stem qan ‘ox’: the suffix -är which marks the plural and the suffix -s which marks the dative (case expressing the addressee). Similarly in Laz, the noun k’oč-epe-š ‘of men’ contains two suffixes: -epe which marks the plural and -š which marks the genitive (case expressing belonging).
However, it is especially in the verbs that the very agglutinative nature of Kartvelian languages is observed. Thus, to say ‘they give it to us’ in Svan, a single word is enough: gw‑a‑hwd‑i‑x, where the prefix gw- indicates the indirect object (recipient) of the first person plural, the prefix a- is a marker of transitivity, the suffix -i helps express the present tense, and the suffix -x marks the third person plural subject. Similarly in Mingrelian, the verb me-r-k’vat-ǝn-k means ‘I shall cut it for you’ and has four affixes: me- which indicates the accomplished and future aspect of the action, r- which indicates the second person indirect object, -ǝn which is related to the expression of the future and -k which marks the first person singular subject.
Thus, a Kartvelian verb can integrate a large number of affixes (most often prefixed and suffixed, sometimes circumfixed) used to mark the following grammatical information:
- person (of the subject, the direct object and the indirect object);
- tense, aspect, mood;
- transitivity, voice;
- valency (number and type of complements governed by the verb);
- affirmation/negation;
- direction, spatial orientation;
- speaker’s uncertainty in relation to the given information;
- various types of derivations that correspond to independent expressions or words in most languages.
To this chain of affixes, we can also add suffixes of reported speech, of interrogation, of subordination…
This abundance results in relatively complex natural forms, such as the form o‑č’ar-apu-an-d-u ‘he was making him write it’ (where o- and -apu indicate the causative, i.e. the idea of inducing an action, -an indicates the fact that the action is not yet accomplished, -d indicates the imperfect and -u the subject in the third person singular) from Laz and Mingrelian č’ar ‘to write’ or the form va-mo-p’-c’on-an-ia ‘he says that they do not like it’ from Mingrelian c’on ‘to please’ (where va- is the negative prefix, mo- is a prefix expressing the aspect, p’- marks the first person subject, -an marks the plural of the subject, and -ia indicates the reported speech).
It is thus interesting to note that for some verb forms which consist of several syllables, the root is represented in reality by only one sound, or even part of a sound! For example, in Svan ǯ-a-k’-u-x ‘you want it’, the stem is k’; in Mingrelian ko-mo-mi-γ‑i ‘bring it to me’, the root is γ; and even more peculiar, in the very common form p-’-un-d-es ‘we had’, the stem is just a simple occlusion of the glottis at the back of the throat, a barely noticeable sound that we utter in English, in the middle of the interjection of astonishment uh-oh!, which is noted by an apostrophe ’ when transcribing Kartvelian languages!
Owing to their typically agglutinating nature, Kartvelian languages very often express by a single word what we express in English by a group of words, or even an entire sentence.

Hélène Gérardin, INALCO and ILARA, lecturer of linguistics and Kartvelian languages
Hélène Gérardin is a graduate of ENS – Ulm and a member of the SeDyl research team. She defended a PhD dissertation entitled Primary and derived intransitive verbs in Georgian: morphosyntactic, semantic and derivational description, at Sorbonne Paris Cité.
For further information
Video by Hélène Gérardin focussing on Kartvelian languages, as part of the Introductions series of ILARA Online.
OpenEdition suggests that you cite this post as follows:
ILARA (March 24, 2023). The Incredible Agglutinative Power of South Caucasian (Kartvelian) Languages. ILARA, the Institute for Linguistic Heritage and Diversity. Retrieved May 21, 2025 from https://doi.org/10.58079/q3at