Optimizing Display Other Formats:      PDF - 736k

Volume 10 Issue 2 (2012)        DOI:10.1349/PS1.1537-0852.A.411

Note: Linguistic Discovery uses Unicode characters to represent phonetic symbols. Please see Optimizing Display for requirements to accurately reproduce this page.

Towards A Full Description of the Focus System in Tundra Yukaghir^[1]

Mark Schmalz

University of Amsterdam

This paper is meant to be a contribution to a better understanding of the focus system of Tundra Yukaghir. It recapitulates the major findings that were made with respect to the topic discussed and presents a number of new empirical facts some of which contradict or form an essential addition to the preexisting views.

1. Preliminaries

1.1 Intoduction

Tundra Yukaghir (henceforward TY) is one of the two survivors, the second being Kolyma Yukaghir (henceforward KY), of a group of closely related languages spoken by the peoples (wadul/odul) that once populated the vast area of the Russian Empire between the cost of the North Polar See and the upper reaches of Yana, Indigirka and Kolyma spreading longitudinally as far as the lower reaches of Lena and upper reaches of Anadyr (Donskoj 1996:22). The ethnic designation wadul/odul, as one can learn from e.g. Kurilov (1996:112), was a cover name for at least 12 closely related tribes, comparable with the term “Slav” referring to Russians, Poles, Serbians etc. as a whole. Although the common ancestry of those tribes is apparent, their representatives normally consider themselves different peoples (Nikolaeva 2002:1). This, together with the pronounced differences in the lexicon resulting in a very reduced inter-intelligibility between the extant varieties of Yukaghir, TY and KY, makes it more appropriate to regard them as two separate languages than two dialects as they were viewed earlier (e.g. Krejnovič 1982:3). Nowadays only about 10% of all Yukaghirs master their language to some extent (Kurilov 1996:113) because the major means of the daily communication have been for decades Yakut and since more recent times also Russian. Even some elderly speakers of TY are so accustomed to speaking a lingua franca of the area that they resort to it even when communicating with each other. At present only five persons have full command of KY (Prokopyeva 2011, personal communication^[2]), two of them living in the Republic of Sakha (Yakutia) and the other three in the neighboring Magadan region. TY is in a slightly better situation in that respect with not less than a half-dozen of competent fluent speakers of TY living in the village of Andryushkino, the settlement with the greatest number of TY speakers. At least as many can be found in the nearby tundra, settling or herding. There are, however, hardly any true TY monolinguals. Efforts to revitalize the language are undertaken. Some of the most important measures are the establishment of a national Yukaghir municipal entity in Andryushkino in 2008 and the introduction of TY lessons at the local school in the late 70’s (Kurilov 1996:113).

TY is conventionally regarded as a language isolate but factually it is remotely affiliated with Uralic languages (Nikolaeva and Chelimskij 1996:155). The focus system of TY is undoubtedly its most peculiar feature, which has many times been an object of scholarly research.

As the title of this article suggests, its aim is to give an account of the TY focus system. The system has already been described by Krejnovič (1958, 1968, 1982). His works along with the contributions by Maslova (2003c) and Kurilov (2006) form nowadays a considerable body of the descriptive sources on TY generally and on its focus system in particular. A number of scholars addressed the issue of (morphological) marking of the pragmatic category of focus in TY after it had been described by Krejnovič (1958) for the first time^[3] (Harms 1977; Ard 1982; Comrie 1992; Fortescue 1996; Maslova 1989a, 2005, 2006). The theoretical contributions that appeared before the turn of the century mainly drew on the data made available by Krejnovič (1958, 1982). However, just as the presence of ‘blank spaces’ on the descriptive ‘map’ of TY is true for TY grammar as a whole, it holds for the pragmatic category of focus specifically.

The ultimate goal of this contribution is to give a more accomplished picture of the whole focus system of TY providing new data for further research. Filling the gaps in its description and reconciliation of apparent controversies in different descriptive sources are thus two main concerns of this contribution. They can be defined more precisely in terms of the concrete research goals which are grouped in three larger blocks. These are as follows:

• means of expressing focus (studying the distributional behavior of the particle me with the view of more precise definition of its functions; focalizing A in general and particularly in sentences carrying a direct object; encoding of focal topics; focalizing verb modified by a modal adverb
• focus in grammatical context (focus in sentences with multiple focal constituents, focus in negative sentences, focus in passive)
• meaning of focus markers (studying contrast as a separate pragmatic function; answering the question whether the focus markers are focus markers s.s. or a device with a wider pragmatic function)

These research goals have received a varying degree of attention in the relevant literature on TY. The distributional behavior of the particle me and focus in sentences with multiple focal constituents can be counted among those that have been dealt with in a rather detailed manner in Matić and Nikolaeva (2008) and Maslova (2005) respectively. Nevertheless, as the contribution of some new interesting data in this article shows, the discussion points of these two topics have not been exhausted yet. Some other research topics were discussed either for KY, e.g. the functional scope of the focus markers (Maslova 1997), the other Yukaghir language, or presented equivocally, e.g. focalizing A in Krejnovič (1958, 1982), Kurilov (2006) versus Kurilov (2001). Yet other issues constituting the research goals of this presentation have merely been touched upon or hardly addressed at all. This concerns especially the focus in the passive and in negative sentences. As for the latter, it was wrongly concluded that the focus oppositions were neutralized (Comrie 1992), while, quite on the contrary, the focus system operates in them in an almost unchanged way as is demonstrated by the examples contained in the present article. A special issue of the TY focus system brought to the attention of linguists here, to my knowledge, for the first time, is the variation of its usage in individual speakers of TY.

Before the research goals are treated, an overview of TY focus system is given in section 1.2. Section 1.3 presents the basic notions of information structure as they are employed in this article. The approach chosen to study the information structure of TY is described there as well. A short presentation of other methodological issues of rather technical nature is given in section 1.4. Section 2 deals with SAs^[4] having the single pragmatic function of focus. It is designed as if consisting of three parts: sections 2.1 to 2.3 illustrate the focus system in sentences with a single focal constituent; sections 2.4 to 2.6 describe the situation in those with multiple focal elements and section 2.7 forms a conclusion containing generalizing remarks on the focus articulations of TY. Section 3 is dedicated to SAs with multiple pragmatic functions: focal topics (section 3.1) and instances of contrast (section 3.2). One of the most enigmatic devices of the TY focus system, the verbal clitic me, becomes the subject of discussion in section 4. This is followed by section 5 on the focalization of verbal predicates modified by qualitative adverbs. In section 6 the functional scope of the focus markers of TY is discussed. Sections 7 and 8 describe focus in negative sentences and focus in passive respectively. Paradigmatic deviations encountered in modern speakers are discussed in section 9. The results of the present study are summarized in the conclusion in section 10.

1.2 Overview of the TY focus system

The focus system of TY is fascinating for several reasons. Its primary linguistic devices are morphological and their employment is highly grammaticalized. The nominal focus markers show a very unusual ergative distributional pattern. The predicate participates in focus marking of the arguments by taking ‘focal agreement suffixes’ (Matić and Nikolaeva 2008:2) and focus on the predicate is overtly marked. The combination of these features, each of which by itself could be sufficient to attract the linguist’s attention, makes the TY focus system an intriguing research object, especially from the typological perspective. Some of the most basic morphosyntactic properties of this language, i.e. the encoding of the core syntactic relations S, A and O, depend on the given focal pattern of a sentence. Therefore, for understanding TY grammar as such, it is crucial to fully understand its focus system.

The focus system of TY is characterized by a number of formal and distributional features. The focus markers as such are the verbal proclitic mer=, the nominal suffix – leŋ and the pronominal suffix –ek^[5]. The focus markers show complementary distribution, thus only one constituent in a sentence can be marked for focus. The focal constituent is reflected in the verbal agreement endings different for intransitive and transitive verbs. Focus on the subject leads to a dramatic loss of finiteness in the verb as out of the 6 personal endings only the one distinguishing 3PL is left. The nominal^[6] focus markers show an ergative distributional pattern with marked S and O on one hand and unmarked A on the other hand. The ergativity is thus split, the conditioning factor being the pragmatic function of the argument, i.e. ±focus.

The following overview of the TY focus system is briefly sketched after Comrie (1981a:259) and Krejnovič (1958:131, 133, 146, 152)^[7].

1.3 Information structure

Every sentence has, apart from its syntactic structure, an information structure. Syntax operates with more or less abstract grammatical categories like subject, object, predicate etc. and describes the laws according to which these categories interrelate. The universal status of some of these categories is questioned sometimes or at least they might have to be defined differently depending on the language under scrutiny. However, the notions of information structure have to do with knowledge and awareness, which rather pertain to the realm of psychology and are characteristic of all humans.

Information structure is basically the way in which an utterance is organized by the speaker with respect to the presumed state of knowledge and/or awareness of the addressee. As Lewis (1979:339) rightly notes, “at any stage in a well-run conversation, a certain amount is presupposed”. Therefore, in a discourse a sentence normally begins with a constituent carrying a piece of information that is shared by the interlocutors to establish a point of departure, or ‘common (back)ground’ in terms of Stalnaker (1974:199), and continues with the constituents that carry information that is presumed by the speaker (S) to be unknown to the addressee (A).

These two types of information – given and new – are analyzed in FDG (Hengeveld and Mackenzie 2008:89, 92) as belonging to two different dimensions of discourse: Topic vs. Comment and Focus vs. Background.

Dik (1978:130) defines constituents with Topic function as those presenting “the entity ‘about’ which the Predication predicates something in the given setting”. There is a strong link between the notions of ‘givenness’ and ‘topic’ although they do not coincide. The given information is normally contained in topics, or subjects which statements are made about, and constitutes the background of statements. “Given the basic function of Topics of relating the Communicated Content to existing information in the Contextual Component” (Hengeveld and Mackenzie 2008:92), it is not surprising that, as in the present work too, the term ‘topical’ sometimes substitutes for the term ‘given’.

The new information contained in the comment is, according to Dik (1997:326), the difference between the pragmatic information of the speaker himself, or P_S, and the speaker’s picture of the pragmatic information of the addressee, or (P_A)_S. This difference constitutes the Focus of an utterance as opposed to the Background, which is presupposed, and the constituents expressing this difference are called ‘focal’^[9]. Hengeveld and Mackenzie (2008:89) add that Focus represents “the Speaker’s strategic selection” of new information, which implies that whenever focus is involved, there is a possibility of an alternative. The indication of “the presence of alternatives that are relevant for the interpretation of linguistic expressions” is seen by Krifka (2008:248) as the main functional and definitive characteristic of focus.

Another important dimension of information structure is the dichotomy Contrast vs. Overlap. Contrast, its by far more important member, “signals the Speaker’s desire to bring out the particular differences between two or more Communicated Contents or between a Communicated Content and contextually available information” (Hengeveld and Mackenzie 2008:96). In this article only the former, marked members of the Focus-Background and Contrast-Overlap dichotomies will be studied because topics remain unmarked^[10] in TY.

Syntactic and pragmatic structures do not coincide in many languages. If analyzed syntactically, the sentence “I am writing an article on the focus system of TY.” can be said to convey only one message, which becomes fully clear when the reader arrives at the dot following the linear order of the constituents of the sentence. But from the point of view of information structure this very sentence can be organized in different ways depending on the packaging of information, or on the way “in which a speaker accommodates his speech to temporary states of the addressee’s mind” (Chafe 1976:28). A pitch accent on the italicized constituent in the following sentences clearly signals that the corresponding constituent is focal and serves to point out to the addressee that he was unaware of a certain relation, adding to or correcting/replacing his pragmatic information.

I am writing an article on the focus system of TY.

I am writing an article on the focus system of TY.

I am writing an article on the focus system of TY.

I am writing an article on the focus system of TY.

I am writing an article on the focus system of TY.

I am writing an article on the focus system of TY.

Unlike the message derived from the syntactic analysis, the message conditioned by information structure becomes clear, at least partly, even before the addressee hears the end of the sentence if the focal constituent occurs early in the sentence: the addressee becomes aware of its focal status as soon as he hears it.

TY presents a relatively rare example of a language in which syntactic and informational structure do coincide. In many languages a certain information structure can be encoded with differential morphosyntactic means (e.g. cleft-sentences in English). However, in TY a certain information structure must be encoded with the corresponding morphosyntactic means. The formal realization of this obligatory interdependence is to be explored in the following sections.

The easiest way to determine how a syntactic constituent is focalized is to design a question that would imply an information gap that can be filled by naming a referent of the constituent fulfilling the same syntactic function in the answer as the question word in the question.^[11] In fact, this, along with the construal of the contrastive meaning, is the only constellation in which the focus marking is obligatory in TY (Maslova 2005) and thus represents a reliable way to study the focus patterns of TY. Therefore, to accomplish the above stated research goals, a number of specific questionnaires as well as isolated sentence pairs or single sentences with a particular structure were used. Before the specific research goals could actually be dealt with, a basic questionnaire (see Appendix) had to be applied in interviews with a number of potential informants in order to study what is believed to have been well established, namely the basic focal patterns of TY already described by Krejnovič (1958, 1968, 1982). This was necessary for two reasons. First of all, at some points quite a few informants deviate considerably from what could be expected according to Krejnovič’s (1958, 1968, 1982) descriptions (see section 9). This presents a major source of interpretational difficulties. The discrepancies could have various possible reasons: influence of the lingue franche of the area, dialectal differences, shift in the focus system or even its deterioration conditioned by the decreasing level of mastery of the language. The latter was the second reason to go with every available informant through the basic questionnaire in order to assess his or her value and reliability as a potential informant in the future. Working with the basic questionnaire was useful also because some of the research goals could either be directly studied with its help (focalizing A) or be shed some light upon (usage and function of the clitic me(r)-^[12]). The basic questionnaire was designed on combinatory basis as having all possible focus articulations^[13]. The obtained focus articulations were grouped as follows:

As far as the study of contrast is concerned, different approaches were combined. Contrast has been singled out following the newest edition of FDG by Hengeveld and Mackenzie (2008). Contrast, as it can be inferred from the presentation by Hengeveld and Mackenzie (2008:99) always occurs in combination with either topic or focus as two other major pragmatic functions. The corresponding (contrastive) focus types identified by Dik (1997, 1:330-334) in his FG form the subcategories of these combinations. They are only one parameter for describing contrast in languages (“communicative point” in Dik’s terminology (1997, 1:330)). The other parameter is the contrast scope which was studied in terms of contrast articulations, not unlike the ‘focus articulations’ of the basic questionnaire. For the study of contrast a simplified version of the questionnaire was employed which contained only the contrast articulations that involved a single constituent (either the predicate or a core argument) that would be contrastive or, if there were more than one contrastive constituent in the sentence, they had the same syntactic function (again, either that of a predicate or argument).

Since contrast does not occur as a single pragmatic function, contrastive constituents can be labeled in terms of FDG as ‘Subacts’ (SAs) with multiple pragmatic functions. The rough structure of the questionnaire for studying contrast is thus as follows (see the more detailed structure in the Appendix):

SAs with multiple pragmatic functions

Focus/Contrast

Replacing

Expanding

Topic/Contrast

Rejecting

Restricting

Selecting

Combinations of SAs with multiple functions

Topic/Contrast + Focus/Contrast

Parallel

Rejecting + Replacing

Expanding (fuller form)

Topic/Contrast + Topic/Contrast

Restricting (fuller form)

To cover certain research goals mentioned in section 1.1 satisfactorily, the work with questionnaires was supplemented with the analysis of the available textual material.

1.4 Methodological issues

The examples listed in sections 2.1 to 2.6 represent, unless indicated otherwise, the most frequently obtained translations of the answers given to the questions from the basic questionnaire which were devised to trigger the corresponding focus articulations. These answers can thus be regarded as standard usage of TY as it is spoken today. Throughout this work, the examples were mainly obtained during elicitation sessions. Otherwise they are given with a reference to their source. The morphemic analysis is in all cases done by me.

The morphemes involved in marking focal status are glossed in accordance with the existing descriptive literature. This makes it easier for the reader to spot the deviant usages.

To illustrate a given focal articulation fully, the answers had to be given in full forms as well. This is not quite natural for TY, though, because theconstituents carrying given information are normally omitted in it. Nevertheless, sentences carrying constituents redundant under normal circumstances are imaginable in certain contexts and can be found in the available published primary data. Apart from this consideration justifying the elicitation of full answers, care was taken to point out the differences, if there were any, between the focus patterns employed in such explicit sentences and sentences in which topical constituents were only implicit. If one of the main constituents is not present in the surface structure of a sentence, the letter indicating it in the abbreviated representation of the focus articulation under scrutiny is included in brackets. The resulting different representations are called ‘configurations’ (of a given focus articulation).

In order to collect the necessary data, a three months fieldwork was conducted during which native speakers of TY residing in Andryushkino, Cherski and Yakutsk (all in the Sakha Republic of the Russian Federation) were questioned. Altogether 13 informants were consulted with varying degree of thoroughness, which was mainly conditioned by the availability of the informants and the setting in which interviews took place. Four informants were male, nine – female. The majority of them were over 50 years of age. The main informants were well over 60. All informants were permanent residents of a multiethnic settlement.

Russian served as the medium language in all elicitation sessions.

2. SAs with the Single Pragmatic Function of Focus

As it is mentioned in section 1.3, several pragmatic functions can be assigned to a single constituent: focus + contrast, topic + contrast etc. In this section only constituents with the sole function of focus are systematically looked at.

2.1 Focal predicate

This section illustrates the morpho-syntactic structure of the sentences carrying focal predicates expressed by intransitive (SV_F) and transitive (AOV_F) verbs.

The pragmatic function of focus in (1-2d) is unambiguously marked by the verbal clitic me. The predicate is the only constituent in the answers which does not contain given information.

AOV_F

According to Krejnovič (1958 :43) the answers with a focalized transitive verb are prompted by the questions with the structure “What did A do?”. In fact such questions regularly produce answers with the pattern characteristic of sentences with the configuration AO_FV_F (OF pattern) exemplified in (8a) and (8b). In order that just a focalized verb form be obtained in the answer, both A and O have to be mentioned in the question. This makes them the constituents carrying given information in the answer and thus deprives them of the status of the potentially focalizable ones.

For the answer to the question from (2b) to be understood as it is intended according to its translation, a short pause would have to follow the personal pronoun in order that its function be disambiguated, i.e. that it be interpreted as the subject of the sentence. Otherwise it could be taken for a possessive pronoun modifying the word lalime(ŋ), the object of the sentence, which would result in the meaning: “[I] broke my sledge”. This functional disambiguation can take place in an alternative manner, by employing a marked word order as in (2c):

The ambiguity of this kind does not arise if an independent demonstrative pronoun functions as the object:

2.2 Focal core argument

The single focal constituent of a sentence can be represented by a core argument. With intransitive verbs it can be only S while transitive verbs can have either the focal subject (A) or the focal direct object (O). The morphological equipment of the focal arguments and their predicates encoded in accordance with the syntactic function of the focal arguments is shown in (3a-6b).

All informants over 50 years of age form the sentences with the focal S in a uniform and predictable way demonstrated in (3a). The assignment of the function of focus to the subject takes place through attaching to it the nominal focus marker, the suffix –leŋ:

In the short answer with the configuration S_F(V) the subject is invariantly accompanied by the nominal focus marker –leŋ irrespective of the age of the informant:

The greater degree of uniformity in the answers with the reduced configuration of this focus articulation could be explained by the supposition that the focal marker –leŋ is reinterpreted by the speakers as the homonymic copula, which must not be missing in a TY sentence with a nominal predicate.

In both of his major treatises of TY Krejnovič (1952, 1982) gives hardly any example with the full configuration of this focus articulation. However, his presentation does give an impression that a direct object can be used in the sentences with focal A occupying the position either in front of the group A_FV or behind it. The scarcity of available information about this focus type makes it especially interesting. Another intriguing and controversial feature of this focus articulation is the fact that in Kurilov’s (2001) view its predicate is encoded in exactly the same way as is the predicate in the focus articulation S_FV, which contradicts completely the accounts by Krejnovič (1958, 1982) and Kurilov (2006).

The focus articulation A_FOV shows the greatest variation from speaker to speaker and within single speakers. It is especially true for its full configuration. In spite of the relatively high degree of divergence, three common encoding patterns can be distinguished and certain tendencies in encoding the constituents of the sentences with this focus articulation can be observed. In discussing this matter it is reasonable to differentiate between the full configuration of this focus articulation and the reduced forms thereof: A_F(O)V and A_F(OV).

In the sentences with the configuration A_F(O)V the AF pattern prevails. It is characterized by the fact that both the subject and the predicate are deprived of any inflexion, the only exception being the 3PL form of the verb that has a plural affix –ŋu.

With the configuration A_F(OV), the number of the informants who encode the only overt constituent, the subject, as SF form^[15] rises. It is regarded as an alternative even by those informants who reject the SF form of the subject in the sentences with the configuration A_F(O)V. The employment of the nominal focus marker, characteristic of the SF pattern and unexpected in (4b), can probably be explained by the same logic as in (3b).

In the full configuration, the expected AF pattern is sometimes replaced by the BC (‘basic conjugation’, see 2.3) form of the verb. For this reason the form of the subject in such sentences, though formally identical with that of the subjects in the sentences formally encoded as having AF, cannot be called AF form any longer.

Two out of 10 consulted informants employ the OF pattern with the full configuration.

Possible reasons for the deviant patterns in (4c) and (4d) are suggested in section 2.7.

From Kurilov’s remarks (personal communication) it follows that there are basically two strategies of encoding the focus articulation A_FOV: the basic form of the subject is accompanied by the AF or SF verb form. The former is normally employed in utterances in which the object is not expressed like in (4a). The latter, known from KY, can occur also in sentences containing a direct object. Note the absence of the focus marker on the subject^[16]:

While the subject and the object in (5a) can switch places, the former strategy of expressing AF in sentences with an overt object is associated with the fixed marked word order OAV:

According to Kurilov (personal communication), (5b) is more categorical: “the dog and no-one else”.

(5a) represents a mixed syntactic structure as the morphological equipments of the predicate and the subject are not congruous: the predicate is encoded as focalizing S while the subject lacks the focal ending characteristic of S-arguments. The mix-up can have the opposite constellation as in 5c where the focal ending – ek of the question word that normally should occur in sentences with the SF verb form is accompanied by the AF form of the verb.

With ditransitive verbs, only the strategy in (5a) appears to be possible:

AO_F V

All but one elderly speaker encode this focus articulation in the same way, fully conformant with the previous descriptions.

The younger speakers have different deviations from this pattern.

2.3 Focal peripheral constituent

When a sentence contains a focal peripheral constituent, it is regularly accompanied by the omission^[17] of the verbal focus marker, the clitic me, but no special marker is employed. The verbal endings in the answers are those of BC. Thus, this focus articulation is marked negatively in affirmative sentences^[18], as far as the morphological means are concerned. A few examples follow to illustrate that.

• oblique cases

• adverbials

A comparison of the message conveyed in (7c) and (7d) is very illustrative too.

The presence of a modal adverb accounts predictably (see Krejnovič 1958:151) for the lack of the verbal focal clitic me in the first sentence of the answer. Just as logical is the presence of that clitic in the last line of (7c), represented by the verb-only sentence in which the predicate carries the new relevant information. But this changes if a focal peripheral constituent is added which is marked by its preverbal position in the sentence and the intonation:

This pattern was recognized by Comrie (1992) who proposed the label ‘neutral focus’ for it because he regarded the corresponding forms as basic. Since the word ‘focus’ very generally implies concentration of attention on something, the term ‘neutral’ can hardly be applied here. Neither can the designation ‘unmarked indicative’, proposed by Fortescue (1996:21) be adequate for these verb forms. First of all, they are, contrary to what Fortescue (1996) believed having analyzed Krejnovič’s (1958, 1982) material, not confined to the indicative mood s.s. Second of all, verb forms as in 7d serve to express the narrow focus on a peripheral constituent and are in such contexts anything else but functionally unmarked. However, they seem to represent a ‘default’ option in certain other contexts (e.g. the use of the BC form in APSs, see 3.1) To my mind, it is therefore more preferable to designate the verb form of the predicate in 7d simply as the ‘basic conjugation’ form, or, abbreviated, the BC form of the verb.

Focus is the main communicative point to be conveyed by the speaker to the listener. The constituent(s) effecting such transfer of information is/are called focal. Focal constituents are encoded, or focalized, in TY with morphological means. Since morphological markers of TY have complementary distribution, only one constituent in a given sentence can be actually focalized morphologically. So far only cases with a single focal constituent have been dealt with. Therefore, it is interesting to find out now which of the multiple focal constituents receives the corresponding morphological marking. This is done in sections 2.4 to 2.6 that follow.

2.4 Topical argument

The absolute majority of the informants encode this focus articulation employing the OF pattern:

This quite regular pattern is just as regularly broken if the sentence pair is designed in a certain way:

One’s tasks equal one’s work, therefore it does not surprise that the word wielmorawγane is degraded in the answer in its focal status. It simply is not new. It actually carries implicitly given information despite not having been explicitly mentioned in the question. Instead the speaker concentrates his attention and the attention of the inquirer on the predicate whose content cannot be foreseen by the latter, thus PREDF.

This focus articulation is not covered in Maslova (2005) and the following description is thus an attempt to close this gap. Considering the high degree of divergence attested for the focus articulation A_FOV, it is not surprising that this focus articulation shows varying patterns of encoding, sometimes even in one speaker. To illustrate it, a few examples are quoted:

The interpretation of the alternative verb forms in the answer is additionally made harder by the informant’s claim that neither of them emphasizes a constituent, but all of them rather just state a fact. It is conspicuous that in spite of the indifference of this particular informant she does not use the PREDF form of the verb in this context. When the configuration is reduced to A_F(O)V_F, the OF form of the verb, naturally, does not occur any longer but both others do. The PREDF form of the verb does occur when the subject of the sentences with such reduced configuration is a pronominal one, especially when it is omitted. The latter is probably the reflection of the normally obligatory usage of the clitic me in verb-only sentences:

Some other speakers are more liberal with the usage of me attaching it also to predicates with nominal subjects.

A specific attested usage is represented by application of exclusively BC form of the predicate for this focus articulation, which could be a sign of deterioration of the focus system. In a more competent speaker this usage is spontaneous as long as the subject is nominal. Here it might be an expression of a compromise in a conflicting situation.

Some elderly speakers are very clear about the usage of the verb forms in the two following alternative answers with reduced configuration to the question “What has happened to the teacher?”, (9c) stressing the subject and (9d) emphasizing the action.

This confirms that there is a high degree of confidence in the competent speakers regarding the choice of the focal pattern when redundant constituents, here O, are absent. It is only natural that no informant with the exception of two^[19] uses the OF form of the verb with the focus articulation A_FOV_F.

The occurrence of the AF focus pattern in the focus configuration A_FOV_F as exemplified in (9c) corrects Maslova’s (2005:606) claim that “the AF construction … marks the remainder of the clause … as its pragmatic supposition and is available only if q( x, O) can be assumed to be known by the listener(s)”. That the AF construction, contrary to what Maslova (2005:607) concludes, need not encode “only narrow focus on A”, is confirmed also elsewhere. The following sentence pair is taken from a story in which a glutton approaches different entities trying to find out who is the strongest one. In (9e) it asks this question of a cloud.

The same story contains an instance of this focus articulation being encoded as having predicate focus, which very clearly shows that both patterns are equally possible:

The choice of the particular focus construction in (9e) and (9f) is most probably dictated by the focus construction employed in the questions: the subject and the predicate are focalized respectively. Normally the same focus construction is used in the anticipated form of an answer to a question. Besides, the subject and the predicate in the answers in (9e) and (9f) are used contrastively with respect to their counterparts in the questions, which naturally triggers the corresponding focus constructions.

In some situations, even if the answer is not given in the anticipated form, certain choices seem to be favored. The absolute majority of the informants answer the question in (9g) with the PREDF verb form:

No one forms this sentence with AF. This is not surprising in the given context because once the speaker has chosen this somewhat unusual form of the answer judging Lasu’s work by how much he is praised for it by others, it is the chief whose opinion and attitude is most important. The chief is therefore is an expected participant of the state of affairs described in the sentence, or an inferred topic, whereas his opinion cannot be foreseen by the listener, thus the emphasis on the predicate.

In the second sentence of (9h) all but one informant put the predicate in its BC form, the “stray” variant being PREDF, impossible by a judgment. It is claimed (Kurilov, personal communication) that the subject is emphasized in the speaker B’s utterance.

While the second speaker’s choice of the particular verbal lexeme in this short dialogue cannot be foreseen by the first speaker, it is not really something conceptually new to the latter if the second speaker agrees with him in assessing Lasu’s personality. Since the second speaker’s opinion does coincide with that of the first one, the second speaker does not consider it necessary to focalize the predicate. On the other hand, it is quite reasonable to stress that literally everybody likes Lasu. The subject is, in such a case, emphasized here by syntactic means only (immediate preverbal position), the morphological AF marking lacking altogether. It seems to be a property of the pronoun jawner that it carries the focus of the utterance by default and the predicate need not be in the AF form.

Unfortunately, clear generalizations can hardly be made with respect to the full configuration of this focus articulation. With the reduced articulation lacking an explicit object, which is more frequent in natural speech, the choice of either AF or PREDF verb form in competent speakers depends predictably on which of the constituents they wish to emphasize.

2.5 Topical predicate

The absolute majority of the informants chose OF while encoding this focus articulation.

It appears that the object has a priority of being formally assigned the focal status in the focal pairs A_FO_F and O_FV_F.

2.6 Thetic sentences

S_FV_F

This focus articulation was studied in two types of dialogue. In the first dialogue (11a) a rather extraordinary or unexpected event is reported in an excited manner. In the second dialogue (11b) the events that are probably expected or at least natural and in no way extraordinary, are reported in a peaceful manner.

While the majority of elder informants apply pf with this focus articulation, the younger speakers show the tendency to drop the clitic me, which transforms the verb form into BC.

In the dialog of the second type the speakers show much more uniformity encoding the non-modified predicates as PREDF:

These findings seem to contradict Maslova’s (2005:605) observations about thetic sentences involving intransitive verbs, namely that “the SF construction subsumes both … S-focus and sentence focus”. This, however, is only an apparent contradiction because in all three examples that Maslova (2005) gives in order to substantiate her claim the S-argument fulfills not only the function of focus but also that of the topic thus being a part of a presentative sentence which behave differently from the categorical sentences regarding the encoding strategies of the pragmatic function of focus. Presentative sentences are discussed in section 3.1. If the S-argument of a thetic sentence is only assigned the function of focus, in the vast majority of cases the sentence is encoded as having PREDF and not SF.

Notwithstanding the above-mentioned, in the dialogues of both types one informant respectively prefers an SF verb form:

This deviation from the common encoding choice shows that the choice becomes relatively free once a discourse goes beyond simple question-answer constellation. In the particular context of the second dialogue the SF choice is logical in a sense. The verb “come” had become topical by the time the speaker reached the point at which he uttered the sentence in (11c). He had been speaking about his son’s coming back from the army and now it is his wife who is coming. So, notwithstanding the fact that there is often a preferred focal encoding pattern, depending on the speaker’s perspective and provided that the context allows it, alternative encoding patterns can be chosen, as could be expected of a living language.

A_FO_FV_F

This focus articulation was studied in the same types of dialogues as the preceding one. For the first type of dialogue the translations could be obtained only from 5 informants and thus can hardly have any statistical value. But even those few translations again indicate how much variation in the encoding of the predicate there is when A is a focal constituent. Three informants choose BC, one – OF, and yet another one uses AF to encode the predicate of the affirmative sentence conveying the new information. The more common choice, the BC, is exemplified here:

The picture is not less colorful in the second type of dialogue. However, the encoding of the predicate(s) as PREDF prevails numerically and is observed in speakers of different ages. The encoding pattern CVB+PREDF (see footnote 20) is employed here to illustrate this focus articulation:

It should be noted that despite of a clear tendency to prefer PREDF verb forms some speakers employed other encoding patterns which are presented below for the sake of completeness. These can be divided in two groups: both predicates are represented by finite verbs vs. only the last predicate is finite. The informants’ choices are then as follows:

This is in clear disagreement with Maslova’s (2005:606) generalization that this focus articulation is subsumed under the OF construction.

The A_FO_FV_F focus articulation obtains also when a sensible answer simply does not contain any constituent of the question. With the following question-answer pair elderly informants construct the answer with the OF form of the verb and the younger ones – with its BC form.

This answer obviously carries contrastive meaning. Since it is the object that is contrastively used, the occurrence of the OF verb form in this somewhat unusual answer is but natural. This sentence is an indication that focus and contrast are expressed by the same morphological means in TY.

2.7 Focus articulations of TY: generalizing remarks

On the basis of the examples in sections 2.1 to 2.6 an overview of the focus system of TY can be given:

Ultimately, it is the particular context of the state of affaires described that decides over the speaker’s choice of a focal pattern as has clearly been shown in (8c), (9g), (11c) and (12c).

Several hierarchies for focalizing a constituent in sentences with multiple focal constituents can be established. (7d) and (11b) suggest that a peripheral constituent carrying new information is perceived by the speakers as more salient than a predicate with the same information structure, which results in the BC form of the predicate, thus: X_F > V_intrF, where X represents the peripheral constituent. There is only one comparable example with a sentence containing a transitive verb in my material but it clearly indicates that the opposition transitive/intransitive is irrelevant. In no single elicitation of it do informants use PREDF form of the verb, which makes it possible to state the hierarchy more generally as X _F > V _F. It must be admitted, however, that the matter is additionally complicated here by the presence of a direct object which also ranks rather high in the focal hierarchy. The choice of the particular focus pattern in sentences with a transitive verb goes hand in hand with the placement of the focal direct object or the focal peripheral constituent in the immediate preverbal position, a privileged position for assigning the function of focus. If the object stands in it, the sentence is encoded as having OF, if that slot is occupied by a peripheral constituent, the BC form of the verb is chosen assigning the focus function to that peripheral constituent. It is one of the clearest proves of the influence of the syntax in the domain of pragmatics in TY, a language which is famous for its morphologically marked focus.

The other two hierarchies are as follows:

• in non-thetic sentences: O_F > A_F/V_F
• in thetic sentences: V_F > S_F and less consistent but tendentially V_F > A_F/O_F

The above-mentioned observations allow to equate in non-thetic sentences the direct object with the peripheral constituent with respect to their rank in the focal hierarchy: O_F = X_F.

It is apparent that among the speakers of TY there is a substantial degree of variation in encoding different focus articulations. As could be expected from an endangered language, the speakers pertaining to the older generation, on the whole, show more conformity with the previous descriptions of TY in employing focal patterns. The deviant usages in some of the younger speakers can definitely be taken as signs of the deterioration of the focus system in TY. For instance, the youngest informant consulted does not use the morphological patterns of SF, AF and of^[22] switching only from BC to PREDF. Another informant used the verbal endings of AF, of and BC interchangeably denying their emphatic value. Some informants provide translations that are contradictory not only to the existing descriptions of TY focus system but to their own comparable translations made on other occasions, which is indicative of the focus system being unstable in them (see also section 9.1). On the other hand, there seem to be instances of deviant usages that are of a more systematic nature. One good example of this is the use of the BC form of a transitive verb to stress the object of the sentence as long as the subject is 3^rd person (see section 9.2). In this latter case one probably cannot speak of the deterioration of the focus system but should consider those deviations as (systematic) modifications of that system, whatever their origin may be. Thus, alternative focus subsystems would have to be recognized for TY.

The least divergences are observed in the sentences involving a predicate expressed by an intransitive verb. In the sentences with transitive verbs the stability of the focal patterns is mostly maintained as long as A is not among the focal constituents. Out of 4 focus articulations involving a focal A only one, A_FO_FV, is stable. While the divergence in encoding focus in the focus articulations A_FO_FV_F and A_FOV_F is more or less natural^[23], its presence in A_FOV calls for an explanation. The instability of the focal patterns in the sentences containing a focal A could be explained by two reasons. Firstly, this type of focus is very rarely attested in the textual material of TY. If the same holds true for the spoken language, it simply can be the first victim of the deterioration of the focus system in TY speakers, whose traces are observable in younger speakers.

The other reason is a possible interfering influence of the topical constituents^[24] that are normally omitted in speech (see also Kurilov 2006:267-268), since they are readily retrievable from the context, but are present in the constructed examples of this work.^[25] What appears in the sentence, are the focal constituents and when, normally suppressed, topical constituents also turn up in the surface structure of a sentence, it leads to a cognitive clash in some speakers who apparently cannot interpret and treat them as topics any longer and sometimes assign the focal status to them, which is reflected in a corresponding encoding pattern. Thus, for instance, the most natural answer to the question triggering the focus articulation AOV_F– and it was the spontaneous answer with the majority of informants – is a verb-only clause. When an informant is prompted to give an answer with an overt object, it probably produces in him or her a signal that the object, since it is present, might be the constituent to be stressed. This leads to a reinterpretation of the answer as having focus on the object with the subsequent attachment of the OF verbal endings to the predicate. The fact that the answer is not an immediate reaction to a question any longer but rather a translation, being thus perceived as somewhat detached from the question, additionally induces such a reinterpretation. A similar effect is manifested in (4d) where the redundant direct object must cause the choice of the OF pattern. This is, in fact, harmonious with the overall tendency that the object is preferred over the other focal constituents in being focalized.

One may object that a comparable reinterpretation and its consequences do not take place in the answers in which the object is missing but the subject is present, i.e. А(О)V_F. This possibly has to do with the status of the predicate in the sentences with AF. While predicates that display PREDF and OF forms can be called true predicates as having full-fledged inflexional paradigms and lacking nominal properties whatsoever, predicates representing AF that are, with the exception of 3PL, zero-ending forms, resemble possessa of possessive NPs (see Krejnovič 1958:134) and express secondary features of the focalized subjects. This led Krejnovič (1958:133) to call them “non-predicative form of the verb” and later group them together with SF predicates under the label “rank II predicates” (Krejnovič 1982:204) characterized, among other things, by the fact that they cannot occur on their own.

This is to show how PREDF and OF predicates are cognitively perceived as more similar to each other than in comparison to AF predicates. If the constituents, here predicates showing AF and OF, are not readily comparable, their arguments, the subject associated with a predicate encoded as the AF form and the direct object associated with a predicate encoded as OF are not comparable either. Therefore the occurrence of A in the surface structure of a sentence does not produce the same effect as the occurrence of O. Indeed, when both A and O, or both candidates for an alternative potential focus assignment, are there in the answer, the only alternative assignment of the focus function that can occur in the context of the focus articulation AOV_F is the one signaling the presence of OF, never AF.

3. SAs with Multiple Pragmatic Functions

Almost all focal constituents in the examples in sections 2.1 to 2.6 had only this one pragmatic function. In this section the encoding of a combination of two pragmatic functions will be presented: focal topics (3.1) as well as the combinations Focus/Contrast and Topic/Contrast (both 3.2).

3.1 Focal topics (presentative sentences)

The combination of focus and topic obtains in so called presentative sentences meant to introduce a new topic (Hengeveld and Mackenzie 2008:99).

When discussing presentative sentences in TY, it is reasonable to divide them into two groups: absolute presentative sentences (APS) which initiate a text being thus (normally) free from any preexisting context and relative presentative sentences (RPS) which introduce a new topic into an already existing context within a text.

As for the RPSs, in the overwhelming majority of them the new topic is encoded as SF form which is expected for obvious reasons:

In very rare instances the predicate of a sentence carrying a new topic appears in its BC form and the noun representing the new topic, correspondingly, in its basic form, without the nominal focus marker:

In fact, the sentence in (13b) was the only example of this encoding strategy in the longest available text in TY. Since the encoding of focal topics in RPSs is otherwise very regular, one could presume that in (13b) another constituent is actually focalized. The morphological equipment of the constituents indicates that it is a peripheral constituent^[26], the word taŋnigine. But with this interpretation the sentence could not be regarded as presentative, which it undoubtedly is. Therefore an alternative explanation for the use of the BC form of the verb in this exceptional example should be sought.

Further instances of BC verb forms encoding focal topics in RPSs could be detected. Consider the beginning of a children story:

There is no doubt that the sole PURPose of the first two sentences is to introduce successively two new topics. In the third sentence they represent given information and are spoken about, which is signaled by their omission. The first two sentences are thus clearly presentative ones but the predicates in both of them are encoded as BC forms. What could be the reason for the use of the BC form of the verb in the RPS in this example? Given the status of this sentence as being presentative, i.e. one in which the functions of focus and topic are combined in one constituent, focalizing of the adverb “suddenly” appears impossible on pragmatic grounds, since manner adverbs are not established as topics in narrative texts. In other words, it would have to be the topic of this sentence that gets focalized. Besides, there are instances of very similar RPSs that contain SF verb forms, which shows that the choice of the BC verb form is not obligatory in presentative sentences containing a focal peripheral constituent:

The situation in (13c) could be explained in the following way. The status of its second sentence as RPS is relative because there has hardly been anything in the text that has already been spoken about. The narration is so far practically free of the preexisting context. Therefore the second sentence could be considered a counterpiece of the first sentence necessary to set the scene. Such interpretation allows regarding the second sentence also as APS, whose predicates as will be seen further are far less rigidly encoded as SF forms. The sentence in (13d), on the other hand, appears in a text after a young man that had been established as the topic in the introductory sentence of that text had been spoken about for quite a while. Therefore there is a need to mark the transition to a next topic in a very clear manner, of which seemingly only SF forms are capable.

Another potential explanation for the focal pattern of the RPS in (13c) is connected with the fact that, as will be seen in section 6, emphasis is expressed in TY with the same grammatical means as focus. In this sense focus and emphasis can be equated in TY. Given the strong overall tendency for focal peripheral constituents to be focalized and since focalization and emphasis can be equated with respect to the formal means of their expression in TY, one can assume that there is a competition between the focal topic of this sentence and the peripheral constituent for the formal assignment of the function of focus and emphasis respectively. Since intra-clausal focal patterns of TY are complementary, the speakers have to opt for one of these, either focalizing the new topic or emphasizing the peripheral constituent. With this interpretation, one would have to conclude that in this particular instance the narrator decided to give preference to the emphasis over the focus. The same reasoning could be applied to account for the corresponding focus pattern in (13b), which, unlike the RPS in (13c), by no means can be regarded as APS, which, as will be seen in the following, can in a number of instances display the BC pattern instead of the SF pattern.

Native speakers of TY (Kurilov, personal communication) confirm that for the sentence in (13b) to accentuate pragmatically the topic it would have to follow the SF pattern:

Analysis of textual material reveals that there is more variation in encoding focal topics in APSs. Numerically SF prevails as with RPSs but BC does not produce an impression of an exceptional phenomenon at all. In fact, these two encoding patterns occur almost evenly. Out of 11 APSs found in Kurilov’s book of children stories in 5 of them the predicate is encoded as BC form of the verb. Regularity can be observed there. The verb l’e- ‘to be’ always appears in its SF. Another existential verb, sa γ ane- ‘to live’ occurs both in its SF and BC form. All other verbs introducing a new topic, those with the meaning of coming into appearance, are used only in their BC forms. A few examples illustrate that:

With APSs even PREDF of the verb can be encountered ^[27]:

It could be argued that (14d) is not a genuine presentative sentence due to the lexeme choice for the predicate. But in my view the verb is semantically empty here. It is not essential whether the protagonists met, dined together or fought. The verb solely describes the circumstances in which the participants became important for the narration. They are clearly accentuated, not the action. The lexical emptiness of the verbal predicate in (14d) becomes especially clear when it is confronted with (14e), the first sentence of a story too, where the predicate delivers along with the subject an essential piece of information. For this reason the sentence cannot be regarded as presentative but should be viewed as an ‘all-new’ sentence whose predicate is predictably encoded as PREDF form.

The less consistent encoding of focal topics in APSs as SF may result from the fact that since they occur in the beginning of a text there is less pragmatic pressure to label them as such for the lack of competitive preexisting topics. They are so to speak the only available topics, which accounts for the choice of the default BC form of the verb and the basic form of the subject. The new topics within a narration cannot be recognized as readily as such by the listener unless the speaker uses a device, the SF marking, to signal that the subject spoken about has changed.

Considering a relatively widespread use of BC verb forms in APSs, it has to be postulated that this encoding pattern apart from focalizing a peripheral constituent has a default function and is chosen when no particular constituent of a clause needs to be emphasized (see also sections 4 and 6), which in view of the above-said probably is the case in APSs.

3.2 Contrastive foci and contrastive topics

These two categories are treated together because elicitations have shown that they are encoded mainly in the same way. The morphological devices employed are the same as those for marking focus. While their usage in contrastive foci can be conditioned by the necessity to mark focus, in contrastive topics they can only serve to mark contrast.

In terms of Dik (1997, 1:330-334) contrastive foci are represented by the focus types Replacing and Expanding while contrastive topics are exemplified by the types Rejecting, Restricting, Selecting. A sixth type, the so called Parallel (contrastive focus) is characterized by the simultaneous occurrence of two pairs of contrastive constituents in the same sentence. One of them is focal, the other one is normally topical. The fuller forms of Expanding and Restricting as well as the combination of Rejecting and Replacing necessarily involve more than one SA with multiple pragmatic functions.

The encoding of contrast is systematically illustrated with the example of contrastive predicates (expressed by intransitive verbs^[28]) because in them contrast is expressed in a most consistent way throughout all six types of contrastive focus within Dik’s (1997) classification. The encoding of contrast in intransitive verbs shows also the least degree of divergence from speaker to speaker.

The subscript capital letter _C stands in the following presentation for ‘contrast’.

SV_C

From (15a-15f) it becomes clear that the function of the verbal clitic me is not confined to encoding focus but covers contrast too. With intransitive verbs, there is only one instance of a contrastive predicate expressed by the affirmative form of a verb that does not carry the focus/contrast marker. It is found in the fuller form of Expanding:

A comparable sentence pair with transitive verbs repeats this pattern. It could be speculated, therefore, that in case more than one affirmative ascriptive SA with multiple pragmatic functions is present in a sentence, only that expressing focus gets the focus/contrast marker, which makes it reasonable to consider focus and contrast separate pragmatic functions despite the fact they are encoded in the same way in TY.

With contrastive arguments (S and O) the picture is not so uniform. Normally they carry the focus/contrast markers as in (16a) and (16b):

Replacing:

Selecting:

While focus/contrast markers are consistently attached to contrastive topics (16b) and to contrastive focal S (16a), this does not happen with contrastive focal O. It is unclear yet whether it is a systematic phenomenon.

The extreme is represented by speakers in whose speech all contrastive arguments systematically lack the nominal focus markers. Their predicate is then used in its BC form:

Replacing:

Selecting:

Given the existence of a specific way of encoding OF (some speakers consistently use the basic conjugation forms to encode OF as long as the subject is 3^rd person, see section 9.2 on paradigmatic deviations) and considering the observation that focus and contrast are encoded in the same way, the encoding of contrastive objects as in (17b) is not surprising. Unlike that, the failure to use nominal focus/contrast markers with contrastive S cannot be accounted for by assuming a subsystem because their omission is not systematic from speaker to speaker and even with the same speaker. Remnants of the expected patterns can be found in speakers, who otherwise use basic forms in sentences with contrastive S, in the combination Rejecting + Replacing and in the fuller form of Restricting of which the latter is illustrated in (17c):

S_CV

The fact that some speakers allow both SF and basic forms for the combination Rejecting + Replacing means that the shift from the focal/contrastive forms to the basic forms is ongoing. This shift can be viewed as a sign of the overall tendency toward simplification by generalization in younger speakers of TY.

An interestng pattern was observed in Parallel contrast of S:

In the second sentence in (18) only one of the two contrastive arguments carries the nominal focus/contrast marker and has the predicate in the SF form. The other appears in its basic form accompanied by the predicate in the PREDF form. It has yet to be confirmed, whether this is a systematic restriction.

As far as the encoding of contrastive A is concerned, there are two different strategies. It can be done by purely syntactic means switching from the unmarked AOV word order of TY to AVO. Speakers applying this strategy do it quite consistently. The only exception is the Restricting type of contrast where the usual AOV word order is observed. This may be conditioned by the fact that restrictive sentences are normally formed with the help of the words like ‘only’. Possibly this lexical device is considered sufficient for expressing contrastive meaning. Morphologically both the contrastive argument and its predicate are basic forms. Interestingly, speakers resorting to syntactic means to present A contrastively assign the function of focus to it in the same way, which is another indication that these pragmatic functions merge in the cognitive apparatus of TY speakers. Consider the following examples:

Selecting:

A_FOV

The other strategy combines syntactic and morphological means. The encoding pattern corresponds to the one expected for encoding focal A: zero-ending forms of the contrastive argument and the predicate. A occupies the preverbal position^[31]:

Selecting:

Putting A in the immediate preverbal slot can by itself testify of the contrastive usage of A as in the second sentence in (20b).

Parallel:

(19a) and (20b), with the generalized word order (O)AV(O), in which syntax is the only means to assign the pragmatic function of focus/contrast, might be indicative of the role syntax is acquiring in encoding information structure in modern TY. It appears that to enable syntax to determine the focal/contrastive status of the main constituents, is another function of the BC form of the verb.

Summarizing, it can be said that the pragmatic function of contrast is encoded with the same morphological means that are used do denote focus on the respective constituent. Deviations from this can be attributed to either language change or the decreasing linguistic competence of some speakers and are, apparently, additionally conditioned by the preference of syntactic means for encoding contrast over the morphological ones.

4. The Verbal Clitic me

Whereas the function of the nominal focus markers –le(ŋ) and –(e)k is quite obvious, the function of the verbal clitic me is not so transparent. In his early work Krejnovič (1958:39) called the verb forms with this clitic “affirmative/assertive-predicative conjugation”^[33] as opposed to the verb forms serving to emphasize the core arguments. In his account Krejnovič (1958:150-151) enlists a number of conditions in which the clitic me can be dropped. For the discussion of the focus system only two of them are important because they concern verbal forms participating in the morphological marking of the function of focus, namely finite verbs in affirmative^[34] indicative sentences. Only one of the two conditions is formal: the presence of a qualitative adverb. The other condition is the lack of the necessity to emphasize the verb. This condition is formulated and illustrated somewhat vaguely. In his later work Krejnovič (1982:187ff) does not take into account the omission of the clitic me any longer, grouping the verb forms with and without it in the same category labeled as ‘rank I predicates’ as long as they display the same person/number endings. The predicates from this category carry new information and characterize a known subject, while the rank II and III predicates (in this work they correspond to SF/AF and OF respectively) do not carry new information.

A decade later scholars showed renewed interest in this issue. One of the main points in Comrie’s (1992) article is that the omission of me takes place when a peripheral constituent is in focus. Fortescue’s (1996:22) analysis led him to say that the functions of the clitic me were the narrow focus on the predicate itself and the wider focus foregrounding the whole predication. The verb forms belonging to the rank I predicates in Krejnovič ’s (1982) terminology and not accompanied by the clitic me, Fortescue (1996:20) regards as default indicative forms in most cases unmarked for their focal status.

The latest academical contribution dealing with the clitic me, an unpublished article by Mati ć and Nikolaeva (2008), is concerned with the cliticized verbs only. The authors come to a similar preliminary conclusion Fortescue (1996) made, namely that the clitic me „signals that the main verb is focused, without further specification of the focus scope“ (Matić and Nikolaeva 2008:5). With the help of numerous examples they convincingly demonstrate that me can express broad focus that includes constituents other that the predicate itself and the narrow focus on the predicate alone. For determining the type of narrow focus they provide the following interpretational hierarchy: focus on the lexical content of the verb

   focus on the polarity

        focus on TAM

           pragmatic enrichment of the content of the verb

My own observations confirm the conclusions made by Mati ć and Nikolaeva (2008)^[35] and enrich them with a few new facts. My data (see section 2.3) also support Comrie’s (1992) proposal of an additional focus type for TY, manifest when the focal constituent is not one of the main constituents of the sentence.

I would like to begin the discussion with two phenomena I was able to detect in TY that up to now were considered ungrammatical in the literature. They concern the allegedly obligatory usage of the clitic me in verb-only clauses and the obligatory omission thereof in sentences with qualitative adverbs^[36].

It is true that all speakers normally use me in verb-only clauses and some of them reject verb-only sentences without it as incorrect. However, a considerable number of them, including elderly people^[37], accept verb-only clauses without me as in (21a) and (21b) or at least do not consider them as entirely incorrect.

This usage is definitely substandard, but the degree of its incorrectness does not even approach that of what would be considered by all native speakers a grave error, such as wrong agreement ending. One elderly speaker claimed that the cliticized forms and those without me could be used interchangeably since there was absolutely no change in the meaning:

There is even more tolerance toward the uncliticized form when the predicate is represented by a so called qualitative verb:

With verbs derived from nouns the clitic me is used in polarity questions to ask for a mere confirmation of already known information (24a). When the inquiring person does not anticipate the answer, the clitic is omitted (24b). The answer invariantly repeats the focal pattern of the question.

(24a) provides the first evidence that the clitic me can be detached from the function of focus and express emphasis alone (see also (26a) and (26b)).

Verbal forms expressing possession can also occur without me if it is not present in the question:

With the quantitative verbs, the clitic me is preferably dropped in verb-only clauses because otherwise the idea of limitation is stressed too much:

As one can see in (26a), verb-only clauses without the clitic me are not a new discovery. In fact, corresponding examples could be found already in Kreinovič ’s (1958) earlier work:

There are three instances of verb-only sentences describing seasonal changes in the text that contains (27). All of them lack the clitic me. One of them is quoted in (28).

The above quoted examples prove that there is a wide array of verb-only sentence types in which the clitic me does not occur with various degree of obligation. Lest a wrong impression arise, be it repeated that with verbs that are sometimes designated as ‘dynamic’ me does normally occur, but there is no absolute prohibition of its omission.

As for the obligatory omission of me in sentences with qualitative adverbs, the following sentences relativize it. They are taken from a short children story in which a girl named L’al’uu finds a snow crane with an injured wing and intends to treat it:

The search for similar examples in the earlier primary sources brought to light three comparable sentences two of which are quoted in (29b) and (29c).

Again, there is undoubtedly a strong tendency for the omission of me in sentences with qualitative adverbs because in most cases the narrow focus on the adverb is intended, but it has to be recognized that it is only a tendency, not an absolute rule.

It has been established in section 2.3 that the clitic me is dropped in sentences whose sole focal constituents are peripheral. There are, however, instances in which a focal peripheral constituent co-occurs with a focal cliticized verb. The following example illustrating that is the opening sentence of a children story.

Since it can be counted as proven that the presence of me normally indicates predicate focus and the emphatic function does not irresistibly suggest itself in the case of (30), the notion of broad focus postulated for TY by Mati ć and Nikolaeva (2008) is a very plausible solution. Broad focus that includes the verb explains probably, at least partly, also the presence of the clitic me in sentences like the ones in (29a), (29b) and (29c). The idea of the specific focus domain ‘predicate + peripheral constituent’ is particularly lucky because it reconciles the fact, that patterns in (29a), (29b), (29c), (30) and in many other instances are attested in TY, with the first hierarchy established for the choice preference when encoding focus in sentences with multiple focal constituents: X _F > V _F. The use of the clitic me in sentences containing focal predicate and focal X might be simply the way of neutralizing the otherwise obligatory narrow X focus reading, letting the focal interpretation of an utterance spread over the whole predication.

Another interesting issue is the usage of me in sentences that do not carry peripheral constituents. Here the particle me is normally present as long as the predicate is focal:

It is not ungrammatical though to omit it. Since there is no peripheral constituent to produce the corresponding narrow focus by such an omission, there has to be another effect. Informants claim that the omission of me in (31) and similar simple sentences, also with intransitive verbs, would produce an utterance made in a ‘matter of fact’ way, or in passing, or as if in hurry, often conditioned by the mood of the speaker. This lack of expressivity can be taken as another indication that me has some emphatic function.

Summarizing the above it can be stated that:

• me focalizes the predicate (but see section 5)
• me is omitted in clauses with the narrow X focus
• me is retained in clauses with focal X(s) if the predicate belongs to the focus domain; the presence of me indicates then that there is no narrow X focus in the clause
• me can be omitted in sentences without X, indicating a decrease of expressivity

The emphatic use of me will be an object of closer examination in section 6.

5. Narrow Focus on the Predicate Accompanied by a Qualitative Adverb.

Adverbs are verbal modifiers. In section 2.3 the narrow focus on them is described. The cliticized form of the verb has been recognized as testifying of the presence of the broad focus in the presence of a peripheral constituent, such as an adverb. The logical question to be posed is how the narrow focus on the modified member in this relation, the verb, is encoded. It is especially interesting to find this out for instances in which a manner adverb serves as the modifier, since manner adverbs show the least tolerance toward the presence of the clitic me, that is the device for marking predicate focus. Elicitations have shown that it is achieved through the nominalization of the verb which becomes the subject of the sentence while the adverbial peripheral constituent is verbalized and promoted to the position of the predicate:

Instead of the nominalized form of the verb, its converb form can be used. This means, in effect, that the predicate and the peripheral constituent (a manner adverb) of the underlying sentence exchange their functions in the surface structure. In other words, peripheral constituents rank so high in the focus assignment hierarchy that the predicate accompanied by a peripheral constituent, in order to be assigned the function of narrow focus has to be converted into a peripheral constituent, pushing the peripheral constituent of the underlying sentence out of this privileged syntactic slot and making it fill the less privileged predicate slot:

The following examples demonstrate the narrow contrast on the predicate.

It follows from the examples presented in this section that peripheral constituents in TY are such strong focus and contrast attractors that a special construction has to be applied in order to focalize or contrast exclusively the predicate in a sentence containing a peripheral constituent.

6. Functional Scope of the Focus Markers in TY

It has been shown in section 3.2 that the focus markers can fulfill the function of contrast markers as well. In section 4 the possibility of use of the verbal clitic me as an emphasis marker has been pointed out. The following examples, in my view, further strengthen, if not decisively prove, that assumption.