Comparing Verbs across the Indigenous Language Families of the Caucasus

In my short attempt to explain why Georgian verbs are so difficult, a few readers asked which is harder: Georgian verbs or verbal systems in Northwest Caucasian, which includes Kabardian, Adyghe, Abkhaz, Abaza, and the now extinct Ubykh. Unfortunately, there is no quantitative method for measuring “difficulty” in a language; it will always be a qualitative judgement and therefore relative. That said, the notion that all languages are equally “hard” is also problematic; for example, most would agree with John McWhorter’s assertion that creole languages are less difficult than non-creole languages.

Thus, rather than seek answers to that question, I thought it may be interesting to compare aspects of the verbal systems across the indigenous language families of the Caucasus: Georgian (which will represent the Kartvelian language family), Adyghe (Northwest Caucasian), and Chechen (Nakh-Daghestanian). Hopefully, these brief comparisons will show that verbs in all three unrelated languages are highly complex, albeit in different ways.

But first, let’s talk about pronunciation. All three languages have a large inventory of consonants; all three make extensive use of ejectives. Yet, while Georgian — and especially, Adyghe — have a relatively small inventory of vowels, Chechen has a plethora of diphthongs — similar in fact to English. Nonetheless, it is safe to say that Adyghe, with a sixty-six-letter alphabet — fifty-six of which are consonants — is on a different level. For a sense of how Adyghe sounds, here is a helpful video from my amazing teacher, Larisa. (Note: I am not the expert in these languages; my teachers — all native speakers — however, are. If you are interested in taking formal classes in Chechen, Georgian, or Circassian, message me and I’d be happy to put you in touch!)

Now to the verbs: ergative structures are present in all three, unrelated languages. Ergativity is another term for “ergative-absolutive alignment.” Languages that use such alignments mark the subject of a transitive verb (a verb which takes a direct object) in the ergative case, while the direct object of the verb is unmarked. This is opposed to languages with “nominative-accusative alignment.” Let’s look at a few examples:

Russian:

John id-yot domoi.

John-nom go-3-pres home

John is going home.

Ya vizh-u John-a.

I-nom see-1-pres John-acc

I see John.

The case ending -a shows that John (as an animate noun) is the direct object of the verb, see. English has lost its case system, outside of a few frozen phrases, and word order is therefore paramount for understanding which is the subject and which is the object of the sentence.

Now let’s look at Chechen. I use the transliterated alphabet from Johanna Nichol’s Chechen-English dictionary, with a few modifications for simplicity (ts instead of c, for example).

John ts’a v-oedu-sh v-u.

John-abs home male-go-pres-pred male-be-pres

John is going home.

And:

Suo ts’a v-oedu-sh v-u.

I-abs home male-go-pres-pred male-be-pres

I am going home.

But:

Asa noxh-iin mott b-yyts.

I-erg Chechen-gen language-abs classVI-speak-pres.

I speak Chechen.

As we can see, when John is the subject of an intransitive verb, it is unmarked. Notice as well in the second two sentences that the first person singular pronoun, “I” changes from the nominative (or absolutive case) suo to asa. This is because unlike “to go,” “to speak” is a transitive verb, hence asa is in the ergative case, and language, mott, is in the nominative or absolutive.

In Adyghe we find the ergative as well:

Aar wna-m maa-k’w.

He/she-abs home-erg/obl 3-pres-go-intr

He/she is going home.

But:

Aaɕ aar jə-ɬaʁʷ.

He/she-erg/ob he/she-abs 3-pres-see.

She sees him.

(Note: I try to use as little linguistic terminology as possible, but Adyghe is especially difficult to transcribe without using the International Phonetic Alphabet (IPA). See here for explanation and recordings of these sounds.)

In the present tense Georgian does not employ the ergative:

John saxl-shi mi-d-i-s.

John-nom house-in preverb-go-pres/future stem-3.

John is going home.

And:

John mas xed-av-s.

John-nom he-acc/dat see-pres/future stem-3

John sees him.

But the ergative appears in the past tense:

John-ma is da-i-nax-a.

John-erg he-nom(abs) preverb-preradical-see-3past

John saw him.

So it’s true that ergativity exists across each unrelated indigenous language family of the Caucasus, but clearly it takes different forms. Now let’s try to tackle verb agreement — one of the most difficult aspects of all three languages.

In Georgian, the verb must not only agree with the subject, but the direct and indirect object as well. Often, though, the direct object is implied. The Georgian verb uses a combination of prefixes (first person v; second and third person zero marking) and suffixes (-s third-person singular; -t plural for first and second person; and -en for third person plural) to show subject agreement:

So far so good, so let’s introduce direct object pronouns:

There is overlap in agreement. For example, gxedavt can mean “I see you all,” “he sees you all,” or “we see you all.”

Here is the same verb in Adyghe, which also requires for the direct object to be incorporated:

As the table shows, the Adyghe verb must agree with subject and object. The subject here is in the ergative case, the object in the absolutive (as we saw in the sample sentence prior). Third person object takes zero marking. The order of incorporation is as follows: object-subject-verb root-3rd person plural (ta-w-ɬaʁʷ, us-you-see, or you see us; ta-0-ɬaʁʷ-x, we-3-see-3plural, or we see them).

Now “to see” in Chechen.

For “to see” in the present tense there is no need for agreement. Yet for the first person singular we don’t see asa, but rather suna. The objects are still in the nominative (or absolutive case), however. Suna is in the dative case; for sensory verbs or verbs of feeling the subject is often in the dative, the object in the nominative. Indeed, in Chechen the subject can be in the nominative, ergative, dative, and even a few others. Thus:

Suo ts’a v-oedu.

I-nom(abs) home male-go-pres

I go home.

But:

Hwuna go suo.

You-dat see I-nom(abs)

You see me.

Now let’s turn it up a notch and incorporate indirect objects as well. Here is the verb “to give” in the present tense in Georgian:

And in Adyghe:

In the above, qə operates as a directional prefix, which can also lend aspect (completed vs. incomplete, habitual, etc.) to the verb. Hence the standard third person direct object conjugation lists the q as more optional. For third person subject and object r is inserted, to differentiate between the two vowels.

Chechen shows far less incorporation, in fact in the present tense no incorporation. Notice that the ergative case is the subject of the verb “to give,” the dative case the indirect object, and the direct object (iza) is in the nominative case.

You give it to me in Chechen is: Ahwa suna iza lo; we give it to them: oxa tsarna iza lo, etc.

Now, let’s change to the past tense and see what happens. Here’s the Georgian for “to give” in the past, or “gave.”

In Georgian “to give” is highly irregular. The root for “to give” in the present dzl changes entirely in the past, to ts. This phenomenon is called suppletion, and Georgian verbs are rife with it — which is what makes learning the language so difficult. Notice as well that the ergative comes into play, which means subject and object take on new markers depending on the tense.

Irregular verbs are also a fundamental feature of Chechen, and even the regular ones are distributed, according to Johanna Nichols, across more than thirty conjugation patterns. This is in part because ablaut, which we discussed in the previous blog on Chechen and Ingush, plays a fundamental role in tense marking. Let’s take a look at “to give” in the past for Chechen:

“He gave it to me” in Chechen is: tso suna iza della; again, iza is in the nominative case, tso ergative, and suna dative.

But why was della in the above sentence written as d.ella in the table? Because Chechen has — depending how one counts — six noun classes, only the first two of which correspond to human gender. The participle d.u is often used to describe the six classes. Here they are below:

Outside of the few words where gender is obvious (boy, girl, mother, father, etc.), and a few rules, the class to which a noun belongs must be memorized. Many of the most commonly used verbs must agree with the noun class — with the object of a transitive verb, and the subject of an intransitive verb as we saw in the previous examples:

Suo ts’a v-oedu-sh v-u.

I-abs home male-go-pres-pred male-be-pres

I am going home.

But:

Asa noxh-iin mott b-yyts.

I-erg Chechen-gen language-abs classVI-speak-pres.

I speak Chechen.

So, “I gave the child to you” in Chechen would be:

Asa hwuna ber della.

But “I gave the boy to you” would be:

Asa hwuna k’ant vella.

So Chechen is difficult because one must navigate the relationship between noun class agreement, verb, and ergativity. Georgian is complex when it comes to the relationship between subject and object and the overwhelming number of irregular verbs.

That leaves Adyghe, which has neither noun class, nor gender, nor that many irregular verbs. In fact, outside of a few phonetic changes due to stress, forming different tenses is relatively easy, just add ʁ (gh), for example, for the past tense.

qəwsatə

I give it to you.

qəwstəʁ

I gave it to you.

Yet Adyghe is capable of incorporating a vast amount of information into its verbs, far more than Georgian. For example, here is a sentence elicited in Kabardian (Adyghe’s sister language) from John Colarusso:

0-q’ə-ɕə-s-ħʷə-w-bʁʷ-a-də-ʁa-a-t-r-q’m

3hor-deixis-I-able-you-side-dat-flat space-cause-conn-stand-pres-not

I cannot make him stand there beside you.

Hopefully, the Adyghe verb will be the subject of a future blog once I can wrap my own head around it — likely in twenty years or so.