About the category of version in georgian verbs

Key words: Georgian, verb, version, inversion, reflexive, polypersonal verbs.

Стаття присвячена дієслівній категорії варіативності в грузинській мові. Варіативність розглядається разом із пов'язаними із нею категоріями персональності, рефлексивності, інверсії, кон'югації та ін. Висвітлені особливості узгодження грузинського поліперсонального дієслова з його актантами. На достатньому мовленнєвому матеріалі доведено, що у грузинській мові лише поліперсональні дієслова здатні виражати дієслівну поссесивну категорію варіативності.

In the A and C tabs we have the two-personal transitive verbs with the subject and the direct object. The B tabs expose the three-personal transitive verbs with the subject, the direct object (the III person singular for all forms) and the indirect object. The A-I tabs with a- prefix vowel are the three-personal verbs. The same changes are in causative conjugation in terms of version.

As we see, the forms of the III series never show version, although they are produced by the version markers. But these markers don't show the category of version in the III series. All forms of version have the same forms in the III series. They don't show any differences. Compare:

Neutral v- Ш-khat'-av - da-m-i-khat'-av-s Subjective v-i-ikhat'-av - da-m-i-khat'-av-s Objective v-u-khat'-av - da-m-i-khat'-av-s

These markers in the third series are so called non-functional forms of version. Generally the version is the category of an indirect object. In the third series where this indirect object goes outside of the verbal morphology, these forms lose the meaning of version although they still have the markers, which are left in the verb in order to keep the destination or possessive semantics of the verb. The indirect object outside of a verb takes the post position -tvis - to keep the possessive semantics (For example: dam-i-khat'avs me(S) surati(Od.) dedis-tvis(Oind.) - I have painted the picture for the mother). In such forms the semantics are the same, but the morphology has changed. Possessive semantics choose the other morphological forms (with the postposition -tvis) to expose its content, but still it leaves its markers (markers of version) in the verbal forms. These markers are out of function as they stay without a real verbal indirect object.

Usually languages don't have any non-functional markers - any marker synchronically or diachronically has the proper meaning and function, which have been changed or lost later. As a matter of fact the morphemes are two sides of a coin: the proper semantics and their formal references. We speak about nonfunctional morphemes when these functions are changed or their semantic fields become narrower. The semantic field of the word is changeable. It can be changed in different ways.

The forms of the III series of transitive verbs are not plain in the sense of morphology and semantics, as several grammar categories meet here. The row of markers conveying the meanings of a many categories, such as: version; aspect; perfect; tense and mood; the thematic markers; maybe causation, split ergativity; inversion - exchanged marking for the verbal persons (when the subject is marked by the markers of the object and the object is exposed by the subject markers: da- m(S)-i-khat-av-s(Od.) - I have painted it. The prefix m- usually is a marker of the object for the first person singular form, but in this case it is marking the first person singular subject and the suffix -s is a subject marker for the third person singular form, but now it is marking the third person singular object). Besides the III series exposes the semantic meaning of "turme" (it seems); the indirect object is outside of the verb (formed as an object syntactically with the postposition "tvis" - for) and the direct object is a diachronically conversed verbal person form the former subject. verbal georgian language personality

The role of the thinking subject (of a human class subject) is less clear in the III series, as the forms of turmeobiti don't have any clear action. This is not the same kind verbal action of the thinking subject with the direct object as it is in the second series. This "it seems" - action is not that categorical as it is in the aorist and it is formed with the ergative-dative. Besides, such attitude of "it seems" creates the proper semantic basis for the subject - not to be a real clear participant of the verbal action and thus becomes an additional reason for appearing of a verbal category of inversion.

Historically the forms of the third series came from static verbs and the appropriate diachronic analysis are also different. The indirect object becomes the subject while producing the forms of the third series from static verbs. That also partly explains the fact of inversion in Georgian. But today these forms of the third series are on a different plane paradigmatically and syntagmatically. These are the opposition forms for the forms of the first and the second series and they are far from their original static semantics. Now they act with their modern semantics. Realizing this in a new way when a new verbal person appears - the indirect object, which is the only legal indirect object in the first and the second series. In the third series we have the historical indirect object with its markers (in the case systems of the nouns and in the personal markers of the verb), although now this is the subject for these forms. The verb cannot accept another indirect object and puts it outside of the verb. Thus we receive a simple object with the postposition "-tvis" with its destination /possessive meaning. In many other languages such objects are considered as verbal arguments. As a matter of fact Georgian verbs may have two indirect objects in one form (mich 'mie - you give food-it to him/her me-of mine; mismie - you give drink-it to him/her me-of mine), but usually this is an issue of another category. The causative are relations exposed by these forms. Besides, in such forms only one indirect object from these two has the priority of being marked in the verb.

It's significant that even without this new indirect object which can't realize itself in the verbal morphology we already have the markers of version from the original static forms. These markers were the functional markers in static verbs and therefore these markers can be considered without the new indirect object. But of course only diachronic analysis wouldn't be correct as the language creates even poly-semantic forms to express the new meanings. The tendency and reality is that new forms have new semantics.

On the other hand static verbs have forms of version: 'ts'eria - uts'eria - ats'eria. As there was a demand to outline the destination semantics in the third series, the referenced forms with objective version were taken as a basis for new forms. This choice between the forms of static verbs seems absolutely clear and understandable. On the synchronic level these forms are non-functional forms as they stayed without the indirect object and no objective version can exist without indirect verbal relation.

"In many languages exposition of the perfect tenses are based on the possessive verbs" (Melikishvili 2002: 121). Typologically the perfect is always with possessive semantics (I have done, She has painted, etc.) and in Georgian transitive verbs the perfect forms of the III series expose these semantics by version markers. In Georgian the analytic forms with the verb "to have" could be considered as the parallel forms for the Perfect, for example: dauts'eria - dats'erili akvs (I have written), gamik'etebia - gak'etebuli makvs (I have done), amishenebia - ashenebuli makvs (I have built), etc.

Finally the appearance of the markers of version in the forms of I turmeobiti has two important reasons: the first is the semantic content of the indirect object that left the verb, its destination or possessive character - as we already mentioned above and the second is the possessive semantics of the perfect tenses. Such a combination base is absolutely enough for one not to consider the vowel markers non-functional in these forms. Exposing the verbal semantics of possesive-destination content is natural for Georgian verbal morphology. In any case, if such semantics are to expose, of course the markers of version will be used, as there is a well exposed three-dimensional model of verbal possessivity in Georgian with its proper referents. There was no need to create new markers for the same semantic content and the language exposed this content with the existing model of the verbal possessive category.

The prefix vowels in Georgian verbs are poly-personal affixes with synthetic semantics outlining concrete semantic in concrete morphological forms. This fact (together with some other facts like that) separates the Georgian language from other agglutinative types of languages. Acad. A. Shanidze in his grammar wrote that the verbal prefix vowels in passives were from the category of version by their origin with the unified functions of version and mood in one marker (Shanidze, 1980: 323). Actually mood and version have the same ranges in the verbal morphemic row and this identical position of these two categories complicates the final answer for e-i opposition. In modern Georgian we have the poly-functional verbal vowels in the prefix positions.

The appearance of the vowel prefix "e-" is not an accidental affix in the II turmeobiti and the III k'avshirebiti rows. This is a relational vowel, which is considered as the indirect version marker in intransitive forms in opposite with the "i-" prefix in so called absolutive (one-personal) verbs. Finally we consider that the prefix "e-" was absolutely correct and a clear grammatical choice of the language for the forms of the II turmeobiti and the III k'avshirebiti to expose such combined or mixed semantics: the possessive semantics of the perfect form on one hand and the possessive destination of the former indirect object on the other hand.

It's a very important item that inversion takes place in the III series. We consider that the actualization of the object as a demand or a request from the ergative construction becomes the fundamental basis for inversion. "The third series is the medial mixed version of the first and the second series in light of ergativity. From the one hand we have no obligations for thematic markers and from another hand we have the ergative-dative. But still this is not a good exposition for the ergative construction and the language tries to outline the role of the object. That's why inversion takes place in these rows of conjugation. Inversion can be explained by the current process of ergativization. This construction process recalls inversion." (Makharoblidze, 2010: 142) Taking into consideration all the above-mentioned, we can say that the forms of the III series in transitive verbs are exposing the inversional possessivity.

Meeting of the different categories and the different semantics in the III series is well-exposed by the poly-functional prefixes. The prefix vowels in Georgian verbs are poly-personal affixes with synthetic semantics outlining concrete semantic in concrete morphological forms. This fact together with some other events in the Georgian language are the samples for breaking the frames of agglutination.

The system of version has its irregular forms as well. The existence of a few anomalies together with the regulations is considered a normal fact in languages. These are asymmetries towards the language system. The category of version has the following peculiarities:

1. The difference between transitive and intransitive verbs by the category of version is systemic. The three dimensional model of Georgian version is available only in transitive verbs. Intransitive verbs may have only the two dimensional model, because there is no subjective version in intransitive verbs. This is the systemic difference of principle character showing the importance and origin of the two-dimensional model of version in the Georgian language.

2. The verbs that lack some forms of version. There are the two groups of such verbs:

A. The lack of the forms inside the one row of conjugation with the missing opposition forms of version, for example: we have v-a-tsnob - v-i-tsnob, but there is no objective version form - v-u-tsnob.

B. The lack of the forms in the verbs that are missing the forms in the conjugation paradigm, for example we have the opposition of the static verbs 'ts'eria - u-ts'eria and they both have ets'era in the second series. The three personal verbs also lose the meaning of version in the forms of the third series. When version creates new forms of conjugation rows, having the function of flexion affixes in the verbal forms, then these affixes lose their destination meaning.

3. Sometimes there are two indirect objects in these verbs. Of course this doesn't mean that both of them are exposed in the verb.

A. Two indirect objects in intransitive verbs - in passives: mitsvia (He/she me-of mine wears it) mimep'q'ar (You treat him/her me-of mine); in medial verbs: mimidevs (He/she follows him/her me-mine), mit'k'iva (It pains him/her-mine). One from of these two objects is always a possessor object and it could have the alternatively descriptive syntactic forms with a two-personal verb and the argument in the possessive object case - genitive or dative (outside of the verb); for example: "mitsvia me tsols k'aba" compare with "chems tsols k'aba atsvia" (My wife wears a dress).

B. Two indirect objects in transitive verbs creates four personal forms: mimits'era (He/she wrote it me him/her) mimikhat'a (He/she painted it me him/her). In such cases mostly causatives are used: gamik'etebine (you make him/her do it me), damikhat'vine (you make him paint it me). All types of such forms are objective version forms without any exceptions.

The adding the indirect object has two ways and reasons:

I. The causation needs to have the executive object or

II. Possessive beneficiary semantics need to be outlined.

The forms with two indirect objects mainly have the semantics of objective version.

4. The poly-semantic forms of version.

There are a few verbal forms in the Georgian language with the conjugative direct object and makes it possible to have the forms of the subjective version in the objective rows of these verbs. These forms are mostly poly-semantic (damiq'ena/dagiq'ena). The poly-semantic form damiq'ena (I. He/she made me stand; II. He/she me made to him/her stand) expresses the subjective version with a conjugative object, loses its reflexivity. The best expression of reflexivity is in the forms where the object is in the subject and additionally can be given by adding the form - tavi; for example: ikhat'avs tavs (He/she paints himself), ilamazebs tavs (He/she makes beautiful herself), etc.

5. The verbs without the category of version. Such verbs have no opposition forms and therefore have no category of version: uq'vars (He/she loves), itsis (He/she knows), akhsovs (He/she remembers), hgonia (He/she takes into), uch'iravs (He/she holds), sdzuls (He/she hates).

These verbs have the two groups as well:

A. The forms without the markers of version: sch'ams (He/she eats), cdz'uls (He/she hates), hkvia (He/she /it is named) and etc. These verbs have no markers, nor understanding of version.

B. The forms with the markers of version: itsis (He/she knows), uq'vars (He/she loves), akvs (He/she has) and etc. These verbs have the version markers without destination or possessive semantic, as they have no opposite forms.

As we already talked earlier about how no marker can appear in the forms without the proper meaning and such forms show diachronically different pictures. This is proved by some forms of comparisons from the other Katrvelian languages, for example the opposition by version is restored in the following forms: I know it - vitsi (Georgian) is a form of a subjective version while the forms mikha-khokha are a clear opposition forms of subjective and objective version in Svan.

Version is a very important category of Georgian verb, which exposes the polypersonal relations of the verbal persons.

References