THE SYSTEM OF GRAMMATICAL DETERMINANTS OF VERBS IN ABZAKH DIALECT SPOKEN IN TURKEY-TENSE, ASPECT AND MOOD

Abstract

The traditional grammar proposes a repartition of grammatical determinants of verbs according to the temporality and mood, but this approach is barely representative in terms of the diversity of the languages which can possess the morphemes that are divided into different syntactic classes i.e. the tense, the aspect, the mood and also the voice. Nevertheless, none of these classes is universal.

When the dialects of the Adyghe language is concerned, Hewitt proposes, for Temirgoï, eight tenses (2004: 130), and Arkadiev (2017) explains that “Circassian languages possess largely isomorphic, even if formally divergent, tense systems distinguishing a perfective and an imperfective past tenses.”. For Kabardian, Kumakhov & Vamling (2009: 45) proposes ten morphemes for tenses. However, according to Catherine Paris, there are four primary tenses in Circassian; the past tense, the present tense and two future tenses (1984). Nevertheless, she still remarques that even if we talk about the temporality easily in the Circassian language, we cannot say that all the dialects are based on the same system. « (…) Par contre, et bien que chaque groupe dialectal possède ses propres distinctions temporelles, essentiellement le passé, le présent et deux futurs, la catégorie temporelle est inconnue du diasystème. » (Paris, 1984 : 39).

In my presentation, I would like to propose you an analysis of these grammatical determinants of the verbs; tense, aspect and mood, in Abzakh, the occidental dialect of Adyghe language (Caucasian languages) spoken in Turkey. As my approach in analyzing the Abzakh dialect follows the principles of the functional linguistics of André Martinet, the results that we have will be slightly different from the descriptions of the former linguists who worked on the dialects of the Adyghe language. In my presentation, according to the realities of the utterances that I collected during my fieldwork in Turkey, I would like to propose you an analysis of these grammatical morphemes that are divided into distinct syntactic classes in terms of these two strictly syntactic criteria: compatibility and mutual exclusion.

Introduction

Abzakh is one of the four western dialects of the Adyghe language. Like all the dialects of Adyghe as well as other West-Caucasian languages, Abzakh is an agglutinative language with a complex verbal system. The present paper sets out an analysis of the grammatical determinants of the verbs conveying tense, aspect and mood in the dialect of Abzakh spoken in Turkey.

Traditional grammar categorizes grammatical determinants of verbs based on two main notions, tense and mood. Such an approach, however, is not adequate in view of the diversity observed among the world’s languages: morphemes that associate to verbs belong to different syntactic sets such as tense, aspect, mood, or voice, none of which is universal. The approach proposed here is based on the principles of the functional linguistic theory of André Martinet. Under this functional approach, categories are based on language-specific considerations, and cross-language generalizations emerge on the empirical basis of comparisons across languages. This approach goes a long way towards preventing circular reasoning whereby universals are recognized in the data because they were in fact assumed from the start. The perspective set out here thereby differs noticeably from the descriptions of the linguists who worked formerly on this topic of Adyghe grammar, as these descriptions hinge on the notion of tense. So, the present approach may require some initial effort on the part of readers. Methodological choices include the decision to base the definition of distinct syntactic classes of grammatical morphemes on two strictly syntactic criteria: compatibility and mutual exclusion. My hope is that readers will find that the adoption of a functional perspective turns out to be worth the effort, and that it allows for fresh insights into the intricacies of a remarkably elaborate and sometimes paradoxical system. Issues of compatibility among theoretical frameworks fall outside the scope of this short article, but I definitely hope that different approaches (distributional, functional, structural, pragmatic, generative, statistical, stylistic, sociolinguistic, and more) can have a fruitful dialogue and are by no means condemned to mutual exclusion.

As a further methodological clarification: the analyses set out below are based on observations of utterances in a data set that I collected during fieldwork in Turkey. My own familiarity with the language (as transmitted through the less than fully intact channel of first-language inheritance, and nurtured in adulthood by the study of materials recorded from older speakers) provides some guidance in the interpretation of the data, as a complement to the tried-and-tested functionalist toolbox.

The structure of this paper is as follows. Firstly, I present a general view on the target morphemes according to the literature found on Adyghe language. Then I analyze first-hand fieldwork data in Abzakh dialect spoken in Turkey, basing myself on the mutual exclusion criterion as proposed by Martinet. Finally, in a concluding part I summarize proposals concerning the analysis of the tense, aspect and mood system of Abzakh.

1. Literature on the tense-aspect-mood markers

Hewitt proposes, for Temirgoï (a dialect of Adyghe), eight tenses (2004: 130): “In Temirgoi the Indicative markers are : Dynamic Present / Dynamic.Root.ø/ (unless the final vowel is to be interpreted as the tense marker); Imperfect /-ʃtə.ʁ/; Future I (immediate/obligative) /-n/; Future II (general) /-ʃt/; Aorist /-(:)ʁ(ɑ)/; Pluperfect has the ending /- ʁɑ:.ʁ(ɑ)/; Conditional I /-ʃtə.ʁɑ/ (or /-n. ʁɑ/); and Conditional II /-ʃtə.ʁɑ:ʁ/. The Stative Present, like its Dynamic counterpart, has no suffixal marker but also lacks any Dynamic prefix.”

In his paper, Arkadiev (2017: 19) explains that “Circassian languages possess largely isomorphic, even if formally divergent, tense systems distinguishing a perfective and an imperfective past tenses.” He also proposes that “Upon the “primary” tense forms “secondary” forms are built by simple addition of tense suffixes to already inflected forms.” (ibid.)

He schematizes his proposal in the following table (ibid.):

Table 1 : “Primary” tense systems of Circassian languages

For Kabardian, an eastern dialect of Adyghe, Kumakhov & Vamling (2009: 45) propose ten morphemes for tenses and seven morphemes for mood.

According to Catherine Paris, there are four primary tenses in Circassian: the past tense, the present tense and two future tenses (1984). Nevertheless, she still remarks that even if all dialect groups have these tenses, it does not make sense to talk about temporality in the Circassian language as if all the dialects were based on the same system.

« (…) Par contre, et bien que chaque groupe dialectal possède ses propres distinctions temporelles, essentiellement le passé, le présent et deux futurs, la catégorie temporelle est inconnue du diasystème. » (Paris, 1984: 39).

In her thesis, Paris proposes a past tense morpheme “-γɑ” for the western dialects and “-ɑ” for the eastern ones, a present tense morpheme “-r(e)” which means repetitive or continuous, one future tense common to all dialects whose morpheme is “-n” which has the meaning of intentionality, and finally a second future tense, which has a more generic interpretation. For the latter, she proposes the morpheme “-š´:t” for the western dialects and “-n:əw” for the eastern ones (ibid.).

2. The target Abzakh dialect and its tense, aspect and mood system

According to the corpus that I collected during my fieldworks in Antalya and Düzce, where I also worked with Abzakh speakers from Corum and Tokat, the Abzakh dialect shows differences probably due to dialectal contact within the diaspora. Therefore, the Abzakh spoken in Düzce and the Abzakh spoken in Antalya exhibit some differences. Nevertheless, when it comes to the tense, aspect and mood system, the informants are highly consistent.

2.1. Bare Form of the Verb

Abzakh uses the bare form of the verb, which is unmarked as we can observe in example (1), referring to an action being realized at the moment of speech.

Ex (1):

“I read a book.” / “I am reading a book.”

When we analyze the first example, we observe that the verb read is marked only by the personal pronouns, it has no morpheme which might be considered as a tense-aspect-mood morpheme. This sentence can have two interpretations: it can refer to an ongoing action at the moment of speech, and it can also have a generic interpretation. The intended interpretation can be underlined by using an adverb of time:

Ex (2):

“I am reading a book now.”

Ex (3):

“I read a book every evening.”

When we compare (2) with (3), we observe that the verbal syntagm shows no difference. Thanks to the use of the adverbs, we interpret the example (2) as an ongoing action and the example (3) as a generic action. We should bear in mind that the bare form of the verb can never have a future or past reference thus can never be used with an adverb of time having a past or future meaning. So the following example is ungrammatical :

Ex (4):

Intended meaning: “Yesterday, I read / was reading a book.”

For the past time references, we must analyze the aspectual category of the language.

2.2. Category of Aspect

According to Comrie (1985: 6), “There are basically two ways in which one can relate a situation to the time line. One is to locate the situation somewhere on the time line, necessarily in relation to some other specified point or segment of the line, since in one sense all time location is relative, there being no absolutely specified points. (…) This concept of time location is essential to the linguistic category of tense (…).” (Comrie 1985: 6). But what we call aspect is rather related to the internal realization of the action itself. Comrie proposes that (ibid. : 6) “The second possibility for relating situations to the time line is that one might be interested in discussing the internal temporal contour of a situation for instance in discussing whether it is to be represented as a point on the time line, or as a stretch of the time line. The internal temporal contour of a situation provides the conceptual basis for the notion of aspect, which refers to the grammaticalization of expression of internal temporal constituency.”

If we return to the past time reference in Abzakh, we find three morphemes: “-γ”, “-ʃtəγ” and “-γaγ”. Their differences can be brought out based on the following examples:

Ex (5):

“I read this book.”

Ex (6):

“I was reading this book.”

When we compare examples (5) and (6), we observe that in example (5), the morpheme “-(a)γ” that we name as perfective refers to an action which is accomplished at the moment of speech. At the exemple (6), the morpheme “-ʃtəγ” that we name imperfective refers to an ongoing action in the past. Whe should admit that they both have a past time reference. Yet, as they both give information on the internal realisation of the action itself rather than just having a past time reference on the time line according to the description of Comrie that we have mentioned above, they both belong to the aspectual category.

At the same aspectual category, we observe another morpheme “-γaγ”, which we name as Perfective 2 :

Ex (7):

“I had read this book.”

When we analyze the sentence in example (7), we understand that the morpheme “-(a)γaγ” refers to an action which was accomplished in the past. The difference between the examples (5) and (7) is that the morpheme “-(a)γ” refers to an accomplished action at the moment of speech while the morpheme “-(a)γaγ”, which is probably the reduplication of the marker “-(a)γ”, refers to an accomplished action in the past.

Basically, all these three morphemes have a past time reference since they refer to an accomplished action or to an ongoing action in the past. But it is still not possible to consider them as past tense morphemes. If we considered the morpheme “-γ” as a morpheme of past tense, in this case, it wouldn’t be used in a complex form which has a future reference as we have in the example (8):

Ex (8):

“Tomorrow, I will have read this book.”

As we can observe at the example (8), the use of the morpheme “-(a)γ” shows us that it is not a past tense morpheme but a perfective aspectual morpheme. If it were a past tense morpheme, then we would have a different form which would be used as perfective in the example (8).

It is also possible to oppose the bare form of the verb and the imperfective morpheme “-ʃtəγ” in order to compare the ongoing actions in different time span:

Ex (9):

“When I see the little child, he is going to school.”

Ex (10):

“When I saw the child, he was going to school.”

If we compare the example (9) and (10), we see that the bare form of the verb refers to an ongoing action at the moment of speech while the use of the morpheme “-ʃtəγ” refers to an ongoing action which was realized before the moment of speech. But they both refer to an ongoing action. In other words, they are giving information on the internal realization of the action either being realized or having been realized.

If we are to summarize this part on the aspect, we can conclude that Abzakh has three aspectual morphemes. Even though they have a past tense meaning, as they give more importance to the internal realization of the action itself, we consider them in the aspectual category. As they cannot coexist with one another, we classify them in the same class of the aspectual category according to Martinet’s criterion of mutual exclusion that we applied during our analysis. The table below summarizes the aspectual category of Abzakh:

Table 2 : Aspectual markers in Abzakh

2.3. Category of Mood

Abzakh has two morphemes referring to an action which will be realized after the moment of speech: “-ʃt” and “-n”. The difference between these morphemes is that one refers to an action which will certainly be realized and the other refers to an action which will probably be realized. According to Mounin’s definition (1974), mood is “Principe de classement des verbes selon les diverses façons dont le locuteur peut concevoir et présenter le processus exprimé par le verbe”. Therefore, we can propose that these two morphemes give the opinion of the speaker himself on the possibility of the realization of the action rather than the future time reference.

Ex (11):

“Tomorrow I might read this book.” / “Tomorrow I will probably read this book.”

Ex (12):

“Tomorrow I will (certainly) read this book.”

Both of the sentences at the examples (11) and (12) refer to an action which will be realized after the moment of speech, in that they refer to a future action. Yet, they both give a subjective interpretation of the speaker on the realisation of the action. Therefore, we consider these markers in the category of mood rather than in the category of tense as future marker. The fact that they cannot coexist at the same verbal paradigme forces us to consider them at the same grammatical class in the category of mood.

In Abzakh; however, we have a third morpheme in the category of mood; “-ʂ^wə”, which we name as habilitatif.

Ex (13):

“I can read this book.” / “I am able to read this book.”

At the example (13), by using the morpheme of habilitative, the speaker gives his point of view on his potential or ability to read the book. Therefore, it is obvious that the habilitative belongs to the category of mood. Its difference from the other morphemes of mood is that, habititative does not have a future reference; it is also possible that it refers to an action being realized at the moment of speech. Another difference is that, in addition to his ability to coexist with the aspectual morphemes, it can also coexist with these two modal morphemes that we name as probable and certain.

Ex (14):

“I could read this book.” / “I was able to read this book.”

Ex (15):

“I will be able to read this book.”

These two examples show us that the habitative can coexist with an aspectual morpheme as well as with a modal morpheme. According to the principle of the mutual exclusion of Martinet, there is no obstacle that a morpheme belonging to the category of  mood coexist with a morpheme belonging to the category of aspect as they already belong to different categories. But two modal morphemes belonging to the same class cannot coexist with each other. Therefore, the fact that the habilitative coexist with the morpheme of certain or with the morpheme of probable is a problematical case. Yet, as they all reflect the interpratation of the speaker on the realisation of the action, there is no ambiguity that they belong to the category of mood. What we can propose here is that, in Abzakh, there is one category of mood having two classes; one of which contains the morphemes of probable and certain; and one of which contains only the morpheme of habilitative. This classification seems to be plausible from semantical point of view too: The morphemes belonging to the first class of mood refer to a future action while the morpheme belonging to the second class of mood has no such reference.

The following table resumes the category of mood in Abzakh dialect:

Table 3 : Category of mood in Abzakh

2.4. Coexistence of Morphemes of Mood and Aspect

As also stated by Arkadiev (2017) and Paris (1984), we observe in our corpus that complex forms can be created based on these pricipal forms that we have mentioned above. A morheme from the aspectual category can coexist with a morpheme from the modal one to create a secondary form.

Ex (16):

“Tomorrow, I would read this book.” / “I was going to read this book.”

At the example (16), we observe that the morpheme of certain and morpheme of the perfective codetermine the same verb at the same time and together they refer to an action which was intended to be realized but then cancelled before the time of speech.

Ex (17):

“If you had eaten your food, I would have bought you a toy.”

At the example (17), we have a counterfactual proposition. The codetermination of the verb buy by the morpheme of probable and the morpheme of perfective is used in a conditional structure.

As we have observed at the examples (16) and (17), one morpheme from the category of mood and one morpheme from the category of aspect can coexist in the same verbal syntagm in order to create secondary and complex structures.

Conclusion

From the examples that we have analyzed so far, we can conclude that Abzakh lays emphasis on the internal realisation of the action and the point of view of the speaker upon the realisation of the action, rather than on its time reference. Therefore, we suggest (paradoxical as it may seem) that Abzakh has no such thing as a distinct category of tense. Instead, it uses aspectual and modal forms to reflect differences in time reference. Based on the principle of mutual exclusion, the characteristics of the tense, aspect and mood system of Abzakh can be summarized as follows:

• Abzakh uses the bare form of the verb to refer either to an ongoing action at the time of speech or to a generic interpretation of the action.
• It has one category of aspect, containing three morphemes : Perfective “-(a)γ”, Imperfective “-ʃtəγ” and Perfective 2 “-(a)γaγ”.
• It has one category of mood which contains two classes. The first class contains the morpheme of probability: “-n” and the morpheme of certainty: “-ʃt”. They both refer to an action which will be realized after the speech moment. The second class contains only one morpheme, namely habilitative “-ʂ^wə”.
• These primary morphemes belonging to the aspectual and modal categories can coexist and create secondary and complex forms.

Table 2 summarizes the proposals put forward in the present paper.

Table 2: Tense, aspect and mood system of Abzakh

LIST OF ABBREVIATIONS

1SG: First person singular

2SG: Second person singular

2SG.POSS: Second person singular – possessive form

3SG: Third person singular

CERT: Certain mood marker

DEM_prox : Proximal Demonstrative

DIR: Directional

HBLT: Habilitative mood marker

IMPERF: Imperfective aspectual marker

NOM: Nominalizer

OBL: Oblique case marker

PERF: Perfective aspectual marker

PERF 2: Perfective 2 aspectual marker

PROB: Probable mood marker

REFERENCES

• Arkadiev, Peter, Historical linguistics of the Caucasus, book of abstracts Paris 12 -14 April 2017, “When perfective and imperfective mean the same: Pluperfect and retrospective shift in Circassian languages”, Mahaçkala, 2017.
• Comrie, Bernard, Tense, Cambridge University Press, 1985.
• Hewitt, George, Introduction to the studies of the languages of the Caucasus, Lincom Europa, 2004.
• Mounin, George, 1974, Dictionnaire de linguistique, Quadrige / Presses Universitaires de France, 3^rd edition: 2000.
• Kumakhov, Mukhiddin; Vamling Karina, Circassian Clause Structure, Caucasus Studies 1, Malmö University, 2009.
• Paris, Catherine, Système du tcherkesse à travers ses dialectes : phonologie, syntaxe, lexique, thèse de doctorat, 1984.