A functional analysis of the analytic past tenses of the Udmurt language

In the case of the Udmurt language it is commonly accepted that evidential is part of the tense system of the language, to be more precise, part of the past tense. [Winkler's 2001, 49-51] states that evidentiality is a feature of the verbal mood system, but his opinion has been contested by other linguists, like [Kozmacs 2008, 168-170] discussed why it cannot be the part of the mood system. This tense is called тодмотем ортчем дыр in Udmurt `unknown past tense', in Hungarian it is often referred to as masodik mult ido `second past tense'.

As Kozmacs [2002, 48] also mentions, the Udmurt second past contains many morphological items. To generate these forms the base form is the infinitive form. In Udmurt infinitive forms end in -ыны or -аны. The endings of the paradigm can be grouped into pairs based on the suffixes they get: the first group contains the 2Sg and 3SG forms, the second contains the 2PL and 3PL forms, and the third contains the 1SG and 1PL forms. As it is also noted by [Kozmacs 2008, 170-171] the closest language to Udmurt (Komi) the second past form only appears in second and third person. Maybe this could be connected to the fact that the first- person forms are formed in a different way than the others. However, there is a suffix that can be found in all forms, the -м suffix, which can also appear in the form of -ем or -эм (the difference between them is only orthographical, depending on whether the stem is palatalized or not), and in the case of the 3^rd group it is always in the form -ем. The past participle marker has developed into the second past ending [Kozmacs 2008, 101]. This suffix is optionally followed by the possessive suffix¹, except for the 1Sg and 1Pl forms, where it is obligatory to mark the person and the number [Timerhanova 2011, 182-183].

The verbs from the second group ending in -ыны, and all the forms ending in -аны the whole infitive ending is dropped, but in the 1st and 3rd group, if the verb ends in -ыны the ы sound stays at the end of the verb, and turns into an -и or -Й. To the stem, the following suffixes are added (respectively to their order): - ськем(е), -ем(ед)/-эм(ед), -ем(ез)/-эм(ез), ськеммы,-ллям(ды)Possessive suffixes are the ones appearing between brackets after the -м suffix where the evidential forms are discussed. Although there are different opinions on whether the -и/й is the part of the stem, or it belongs to the suffixes, the aim of this study is to show, which forms appear, so I consider it as the part of the stem for a practical reason: this way the number of the possible endings is smaller, and this way the system is less vague. In Timerhanova's tables [2011, 182-183] we can find the -эмды and -емды forms as well, but these ones are starting to fade away, which is shown by that fact as well,that in these tables they are at a secondary position, and in other books, they are not even mentioned, like in Ganejev and Perevozchikov [2005], and my own experiences during my fieldworks also show that in the 2Pl forms the -лля- suffixes is used, so I will use this form as well., -ллям(зы) [Timerhanova 2011, 182-183].

In this list it is visible, that there is another suffix that appears in front of the -м suffix. In the case of the second group it is -лля- which is the marker of the plural, and in the case of the 3^rd group it is -ськ-. The - ськ- ending has many known functions [Kozmacs 2008], but in this case it refers to the lack of control of the speaker over the event. udmurt language speaker

The negation of the Udmurt second past can be carried out in an analytic and a synthetic way as well. The use of either one has no effect on the meaning, they are in free variation, both being the part of the standard Udmurt. The only difference may be that the analytic form is much more frequent in the northern dialects, and the synthetic is more frequent in the southern dialects [Kel'makov-Hannikeinen 1999, 181]. The analytic form is formed by combining the second past form of the verb and the negation of the verb to be (овол), and the synthetic form is made by changing the -м- suffix to -мте [Kubitsch 2017, 12-13]. In the 3^rd group the -ськмт- consonant cluster is solved by adding an -ы- sound. In case of negation the marking of person or number becomes fully optional. The synthetic negational forms are the following in respect of the order: -ськымтэ(е), -мтэ(ед), -мтэ(ез), -ськемтэмы, -ллямтэ(ды), -ллямтэ(зы) [Timerhanova 2011, 182-183].

Among the analytic past forms in most cases the auxiliary verb is in second past form, and in one case the main verb has this form. So, the question rises, whether these second past forms also express any kind of evidential meaning, and if there is any difference between them in function.

In the Udmurt language there are 4 synthetic tenses (present, first past, second past, future), and there are 4 analytic ones as well. These latter ones are often barely discussed by textbooks and studies focusing on Udmurt tenses, even the latest work on morphology, written by native speakers of the Udmurt language, discusses all these four tenses on only five and a half pages [Timerhanova 2011, 184-189]. It also does not really help the situation of the reader of this study that these barely half page long sections about the different second past forms are mainly morphological descriptions and examples, whereas a functional description is almost completely missing. During my research, except for one informant, all of them could identify these forms, but not all these forms are used these days.

Furthermore, the functional differences are not discussed even in descriptive sections [Timerhanova 2011, 184]: „Тодмо но тодмотэм ортчем дырын каронкылъёсын но вал, вылэм юрттйсь кылъёсын чош верам каронкылъёс кылкабтодос пуштроссыя синонимъёс луо (пуштроссыя коня ке портэмгес ке но).” (Translation: The ones formed by the first and second past forms of the verb and the вал, вылэм auxiliary verbs, in their meanings are synonymous to each other (although there is a small difference in their meaning).”).

If they are really synonyms, then the following question rises: Why does Udmurt have three (four?)Although Timerhanova [2011] differentiates 3 different forms, as I will show, it is possible, that maybe there is a fourth form as well different forms (first past + вал, first past + вылэм (?), second past + вал, second past + вылэм), if they are interchangeable, and there is no difference in meaning? Although there are redundant phenomena in the languages, but it seems too much for a language to have three (or four?) forms to express the very same mean- ing/phenomenon/function. Furthermore, even though the author mentions, that there is a small difference between the meanings, she does not discuss what exactly this difference is, although it would be really helpful for the reader to differentiate these two tenses.

In the followings I will describe these analytical past forms based on the morphological work edited by [Timerhanova 2011, 184-189], because this is the most up-to-date, and one of the longest summaries of these past tense forms, which was written by a native speaker of the Udmurt language.

Altogether six sentences were created by changing only the berb to all possible analytic past tense forms, including the one considered non-existent, the present tense form of the main verb and the вылэм auxiliary verb. Then I asked native speakers to describe the differences in the meaning of these 6 sentences. I asked 6 native speakers to answer my question (one of them got the sentences in a printed form, the other five were contacted via internet). My informants were all females. I chose to ask only female native speakers as the language use of women tends to be more conservative than the language use of the men [Labov 1991, 206], so this way there is a much higher chance that they will be able to differentiate these forms from one another. These six informants were chosen in a way that the different Udmurt dialects should have been represented. Moreover not only their sex was the same, but they were all fluent speakers of the Udmurt language, and they were all university students, or young Udmurts, who had finished their studies at university not so long ago. Although all 6 people assured me to send an answer, in the end only 4 of them answered my question. The lack of answers from that two people could occur, because they could not differentiate the meaning of the sentences, but they did not want to admit it. This assumption, and the fact that one of them admitted that she could not differentiate them supports the hypothesis that these analytic past forms (at least a part of them) are not part of the active language use anymore. Although the number of answers is quite small, we still have answers from the speakers of the northern and southern dialect of the Udmurt language, and one of the informants is from Izhevsk, the capital city of Udmurtia, which is linguistically a Russian dominant city. The answers were congruent, but before I expose the results, I would like to mention three important comments. First, there was only one person who could not fit the first plusquamperfect + вылэм auxiliary verb form (the form that according to the grammar is non-existent) into the system, but not even this person said that it would be an incorrect form, she just said that this form is not used in her dialect. The second thing to mention is that, as I mentioned above, there was one person to whom there was no difference in meaning between these sentences („мон понна соос вообще ваньмыз огкадесь” 'As for me they are all the same'). She only found some difference between the two different forms of the simple plusquamperfect, but even the difference between them is neglectable. The third interesting thing is that according to the answers of the these native speakers the second plusquamperfect + вал auxiliary verb, and the simple plusquamperfect + вылэм have the same meaning, although according to Timerhanova's morphology [2011, 184] the interchangeable forms are the first and second plusquamperfect forms.

The answers were summarized in a table based on whether the event happened once, multiple times, or it is irrelevant how many times it happened, and based on whether the speaker knows how the event passed, or she has only non-firsthand information about it, so the evidentiality has a role in the differentiation of the meaning.

The results

The congruent answers of my informants can be summarized in the following table (table 4).

Table 4 The results

As it is visible the вылэм and вал variants of the simple plusquamperfect are in a complementary relationship. The variant with the вал auxiliary verb expresses an event that happened multiple times, while the variant with the вылэм auxiliary verb expresses an event that happened only once, and the information is considered non-firsthand from an evidential point of view. The following question rises: Is it possible to express somehow that an event happened multiple times, while the information is considered non-firsthand? The multiple time meaning appears in simple plusquamperfect with the вал auxiliary verb, and the first plusquamperfect with the вал auxiliary verb expresses that it only happened once and the speaker has firsthand experience about it. The frequentative past with the вал auxiliary would be a good guess, but it carries a habitual meaning as well, which is not necessarily the case in this situation.

It is also visible, that both in first and second plusquamperfect if the вылэм auxiliary verb is used, it becomes irrelevant how many times the event happened. We can also conclude that in these tenses the form of the main verb shows if the speaker has first- or non-firsthand information about how the event happened. The only exception is the first plusquamperfect with the вылэм auxiliary verb, because in that case both firsthand and non-firsthand experience is marked. How can it happen? It can be explained the following way: the speaker has firsthand knowledge about the fact that this event happened, but she is not familiar with the exact details about the way it happened. To put it in another way: in all the other cases the fact that the person went for the committee to ask for permission is handled as one single event, no matter if the speaker is familiar with the circumstances or not, but in this case the asking for permission, and the way it is asked is separated from each other. So, this form (which according to the grammar does not exist) fits into the system, as the speaker is aware of the fact that the event happened, just has no knowledge about how it happened.

There are more results that can be found if we put those forms next to each other, where the auxiliary verbs are the same, but the main verbs stand in different forms. In this case it becomes visible that the simple plusquamperfect with the вал auxiliary is in a complementary relationship with the second plusquamperfect formed with the same auxiliary. As the auxiliary verbs are in the same form, it should be the main verb that carries the difference in meaning. So the one in which the main verb is in present tense form, the main verb expresses that it happened multiple times and the speaker has firsthand experience about the event, and the one in which the main verb is in second past form expresses that the event happened only once, and the speaker has only non-firsthand experience about the event. But if the form of the main verbs remains the same and we change the auxiliary to its second past, then the auxiliary verb will be the one showing that it is non-firsthand knowledge of the speaker and the form of the main verb will express whether the event happened only once, or it is irrelevant how many times it happened.

In the case of putting the forms with the same forms of the main verbs next to each other with the different auxiliaries, then we can see that compared to the previous complementary situation, here the auxiliary verbs will be the ones carrying all the meanings, as the forms of the main verbs are the same, so it cannot be the result of the differences. In this case in the second plusquamperfect the main verbs second past form expresses that the speaker has non-firsthand information, and the auxiliary verb expresses that the event happened only once, or it is irrelevant how many times it happened. In the case of the first plusquamperfect the situation is a bit more complicated. The first past form of the verb expresses that the speaker has a first hand experience about the action, and the вал auxiliary expresses that it happened only once, but the вылэм has a dual meaning, as it expresses that it is irrelevant, how many times it happened, and also that, that the speaker does not have any firsthand information about how the event happened.

Summary

Based on the results we can state the following. Although in the morphology edited by [Timerhanova 2011, 184-187] there is only one form of the plusquamperfect, based on the results, the speakers could easily fit the one considered non existent in the system, and it was fitted in the same way. What is even more interesting about fitting it in in the same way is that this form has a much more complicated meaning than the others (as it separates the event from the way the event happened), so it is interesting, why it is not included in the descriptions. We can also state that according to the results the simple plusquamperfect with the вылэм auxiliary and the second plusquamperfect with the вал auxiliary express the same thing, namely, that the event happened only once, and the speaker has non-firsthand knowledge about it. So, based on this, and the fact that in Timerhanova's work on morphology [Timerhanova 2011, 184] the meanings of the first and second plusquamperfect forms are interchangeable, we could assume that in this case, except for the simple plusquamperfect with the вылэм auxiliary, all the other examined forms are interchangeable. But this would mean, that this is a heavily redundant system in the Udmurt language. We can stick to this statement, and then we can assume that these analytic past forms start to lose their original meanings, because the speakers do not really feel the urge to use these forms anymore, or it can occur because of the constant contact with the Russian language which does not have these kind of analytic past forms. Maybe this decline of the system is really happening in the present-day Udmurt language, but based on the results we got, we can still find the possible core meaning of these forms:

• simple plusquamperfect with the вал auxiliary: expresses an event in the past that happened multiple times, and the speaker has first-hand experience about it;

• simple plusquamperfect with the вылэм auxiliary: expresses an event in the past that happened once, and the speaker has non-firsthand experience about it;

• first plusquamperfect with the вал auxiliary: expresses an event in the past that happened once, and the speaker has first-hand experience about it;

• first plusquamperfect with the вылэм auxiliary: expresses an event in the past for which it is irrelevant, how many times did it happen, and the speaker has first-hand experience about the event taking place, but does not have any information about how it happened;

• second plusquamperfect with the вал auxiliary: expresses an event in the past that happened once, and the speaker has non-firsthand experience about it;

• second plusquamperfect with the вылэм auxiliary: expresses an event in the past for which it is irrelevant, how many times did it happen, and the speaker has non-firsthand experience about it;

Also, it is fair to state that for someone who studies Udmurt it can cause a lot of problems to differentiate these forms, because it is not regular, whether the main verb causes the differences or the auxiliary verbs.

My aim for the future is to investigate this topic further, to support the theory I presented above, and also to make further investigations in the Udmurt corpus to find out, whether the forms having the evidential meaning in them have the potential to express politeness [Aikhenvald 2004, 241-242]. However, based on the system I presented above, only the non-firsthand meaning appeared in those cases where the second past form was used. which strongly limits the possibility to express politeness, because for that, the lack of control over the event is much more suitable.

References

1. Aikhenvald A. Y. Evidentiality, Oxford, Oxford University Press, 2004. 452 p. In English.

2. Csucs S. Chrestomatia Votiacica, Budapest, Tankonyvkiado, 1990. 223 p. In Hungarian.

3. Ganeev I., Perevozcikov J. Marym les'a. Izhevsk, Udmurtia, 2005. 203 p. In Udmurt.

4. Kel'makov V., Hannikainen S. Udmurtin kielioppia ja harjotuksia, Helsinki, Suomalais-Ugrilainen Seura, 1999. 335 p. In Finnish.

5. Kozmacs I. Udmurt nyelvkonyv, Szeged, JATE Press, 2002. 104 p. In Hungarian.

6. Kozmacs I. Az -sk- kepzo az udmurt (votjak) igekepzes rendszereben, Nyitra, 2008. 196 p. In Hungarian.

7. Kubitsch R. Evidencialitas a mai udmurt nyelvben, Szegedi Tudomanyegyetem: MA thesis, 2017. 47 p. In Hungarian.

8. Labov W. The intersection of sex and social class in the course of linguistic change, Vol. 2. Issue 2., USA, Cambridge University Press. 1990. 205-54 pp. In English.

9. Siegl F. The 2^ndpast in the permic languages, Tartu Ulikool: MA thesis. 2004. 189 p. In English.

10. Timerhanova N N. (ed.), Udmurt kyllen kylkabtodosez, Izhevsk, Izdatel'stvo „Udmurtskii Universitet”, 2011. 408 p. In Udmurt.

11. Winkler E. Udmurt. In: Languages of the World: Materials 212. Munchen, Lincom Europa, 2001. 85 p. In English.

Размещено на Allbest.ru