Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Manifestations of areal convergence in rural Belarusian spoken in the Baltic-Slavic contact zone

Manifestations of areal convergence in rural Belarusian spoken in the Baltic-Slavic contact zone Journal of Language Contact 4 (2011) 184–216 © Koninklijke Brill NV, Leiden, 2011 DOI 10.1163/187740911X589280 brill.nl/jlc Manifestations of areal convergence in rural Belarusian spoken in the Baltic-Slavic contact zone* Björn Wiemer a & Aksana Erker b a) Professor für slavische Sprachwissenschaft, Johannes-Gutenberg-Universität Institut für Slavistik, Jakob-Welder-Weg 18, D- 55099 Mainz, Germany wiemerb@uni-mainz.de b) Johannes-Gutenberg-Universität Institut für Slavistik erker001@uni-mainz.de Abstract Th is article combines methods and insights from dialect geography, areal and contact linguistics. It focuses on a specifi c, yet heavily understudied mixed dialect of Belarusian spoken in the entire Slavic-Baltic contact region. It turns out that on practically all levels (from phonetics to syntax), the variation of features encountered in this dialect is, from all varieties of that region, the most representative for convergence phenomena characteristic of East Slavic, Polish and/or Baltic vari- eties in contact with each other. Th is convergence is so far-reaching that, for instance, it seems impossible to distinguish the rural Belarusian vernacular from regional varieties of Polish on the basis of structural properties alone, despite the fact that these varieties are clearly perceived as diff erent by both native speakers and fi eld linguists. Simultaneously, features of the Belarusian dialect – and, thus, of the whole contact region which it most accurately refl ects – should be judged under the perspective of larger areal clines (in particular, within the eastern part of the Circum Baltic Area); this view is pursued on the basis of Wiemer (2004) and more recent insights into the areal distribution of structural features crossing family boundaries. Such areal continua, in turn, intersect with inner-Slavic dialect continua and phenomena occurring in vari- ous locally restricted “pockets” scattered around in Slavic. On the background of this, we approach answers to the problem of determining the infl uence of contact (with Baltic and/or Finnic) over “genetic heritage” and the question of which features are more “immune” against infl uence from genealogically less close contact varieties. Keywords Baltic-Slavic contact zone ; Circum-Baltic Area ; Belarusian ; areal convergence ; mixed dialects * We owe our gratitude to an anonymous reviewer who scrupulously pointed out some short- comings of a previous version of this article. Even though we might not have followed his/her advices in every detail, we hope that the article has increased in accessibility for readers not familiar with facts of Slavic and Baltic. Of course, the usual disclaimers apply. B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 185 Introduction: basics about a particular kind of rural Belarusian mixed dialects Mixed Belarusian dialects are spoken all over the countryside along the over- lap region of (East) Slavic and Baltic dialects stretching from Northeast Poland in a half moon-like way up to Latgalia (i.e. Southeast Latvia). Th ese dialects are commonplace on the Belarusian side of the contemporary border between Belarus and Lithuania, but they are also still encountered in many spots on the Lithuanian side of this borderline (see Map 1 ). Th is region will henceforth be called the Baltic-Slavic contact zone (BSCZ). It belongs to the Southeastern part of the Circum Baltic Area (= CBA; cf. Dahl/Koptjevskaja-Tamm 2001 ). Th e geographical expansion of mixed Belarusian rural dialects more or less coincides with the region in which local Polish, more precisely the north- ern variety of polszczyzna kresowa (henceforth PolKres), has evolved (see section 1.2). It is shown on Map 1 . Hereafter we will refer to rural mixed varieties of Belarusian in the BSCZ simply as ‘rural Belarusian’ (or ‘vernacular’) if not indicated otherwise. Th e mixed Belarusian rural dialects are often referred to by their speakers as Mowa Prosta . Th is self-denomination cannot be regarded as a term (or lingu- onym) proper. It has arisen (and is normally used) in contrast to PolKres. It occurs most consistently in the southwestern part of the BSCZ where, until 1939, the density of representatives of the Polish gentry (Pol. szlachta ) was higher than in most of the remaining parts of the BSCZ. Polish was usually chosen as a means to underline cultural superiority 1 so that Belarusian found itself in sociolinguistic opposition to Polish 2 (see 1.2). Simultaneously and during all periods, Belarusian has probably been used ubiquitously in public life, and there is grounds to assume that only few speakers of PolKres had no or only a poor command of Belarusian, whereas speakers of Belarusian were often not fl uent in Polish (either PolKres or the standard variety), at least after World War II. Furthermore, during the interwar period, Polish was the language of instruction in schools even in the BSCZ with its predominantly 1 Th is tradition developped in the times of the Grand Duchy of Lithuania (15th-18th c.); cf. Wiemer ( 2003b : 109f.). 2 Mowa Prosta literally means ‘straightforward, easy, unsophisticated speech’. An alterna- tive expression – also in Polish – is język tutejszy literally meaning ‘our language’. Informants themselves often say that they speak pa prostu or po tutejšemu (with various phonetic variants between Polish and Belarusian). Th is seems to be a very common way of self-denomination serving also as a sign of intra-group solidarity in linguistic and/or ethnic minorities. In even other localities, inhabitants would rather call their speech “mixed” or “local language” (cf., e.g., Jankowiak 2009 : 70). 186 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 Map 1: Dispersion of Belarusian rural dialects and PolKres (from Grek- Pabisowa 1992 : 56) B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 187 East-Slavic speaking population, while Belarusian was being ousted from education. From the sociolinguistic viewpoint, speakers of contemporary rural Belarusian in the BSCZ certainly do not constitute a homogeneous group. We know neither how many of them are or have been bi- or multilingual (or multi-dialectal), nor how many are or have been illiterate. Diastratically, rural Belarusian is “fl at” and represents a decidedly “low” variety. From the diatopic viewpoint, it is very dispersed, but we do not even know whether (and to what extent) its subvarieties in diff erent parts of the BSCZ share the same features and/or amount of structural variation in detail, or in other words: to what extent their variation has been subject to infl uence from a partially diff erent set of contact varieties (see Table 1 ). In rural Belarusian within the BSCZ, we observe an extraordinary amount of idiolectal variation, especially in phonet- ics and morphology. As a consequence, it turns out to be particularly diffi cult to describe it in structural terms including all kinds of phonic and morphic variants, for which even the status as allophones or allomorphs, respectively, is often unclear. Eff orts to comprehensively capture some structural variation in the rural Belarusian of the BSCZ have so far been made by Wiemer ( 2003a ; 2006 ) and Erker ( 2009 ; forthcoming) as well as by Jankowiak ( 2009 ). Recently Smułkowa ( 2010 ) has dealt with dialect levelling and mixing on the back- ground of an advanced process toward an East Slavic (Belarusian + Russian) – Polish merger for which it is often diffi cult to establish a matrix code. Finally, we ought to mention Sudnik ( 1975 ) who, in her impressive work on phonetic variation and a phonological diasystem of multilingual speakers, took account of local Belarusian varieties. Th ere have also been a number of articles in which structural features of the dialects we are interested in here have been investi- gated de facto, albeit often under other names (cf. the surveys in Wiemer, forthcoming1; in prep.). Table 1 subsumes the partially divergent arrays of Slavic and Baltic contact varieties which one has to reckon with when doing research into rural Belarusian in the BSCZ. Th e columns from left to right are ordered iconically according to the geographical direction from southwest (near to the Polish border) to northeast. By ‘regional Russian’ we mean varieties of Russian that are more or less close to the Russian standard language but are themselves usually infl uenced by Belarusian traits to a considerable degree. To what extent Russian played a role in the Lithuanian part of the border region (and to what extent it continues doing so now) is not clear. Th e weak Belarusian standard does not exert any tangible infl uence on rural Belarusian speakers. Th e same obviously holds for other standards, i.e. Polish (West Slavic), Lithuanian and (in the northeastern 188 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 3 Cf. Hentschel ( 2008 ), Zaprudski ( 2008 ), Kittel et al. ( 2010 ). 4 Cf. the volume edited by Zemskaja/Šmelev (1984). 5 Th e much smaller group of Polesian dialects in the southwest, bordering with Ukrainian, is beyond our present concern. edge) Latvian or Latgalian. Whereas Belarusian provides the dialectal basis everywhere, Lithuanian has ceased to play a role beyond the contemporary state boundary with Belarus; it seems to have retreated at least at some spots on the Lithuanian side, too (cf. Wiemer 2006 ). Th e diatopic and diastratic characteristics of rural BSCZ-Belarusian mark- edly diff er from urban mixed Belarusian (or Belarusian-Russian) varieties denoted with the quasi-term Trasjanka , 3 although the results of the “mix” rep- resented in rural Belarusian become manifest in structural properties and variation reminding us partially of phenomena observed in Trasjanka or Russian Prostorečie . 4 It is anything but clear whether and to what extent dialects distinguished in Belarusian (or East Slavic) dialectology are still alive in the BSCZ. It is thus very diffi cult to say anything specifi c about the possible impact of former dia- lect continua on the formation of the mixed rural dialects, which are the topic of our contribution. On the basis of isogloss bundles, traditional Belaruthenist dialectology distinguishes a northern and a southern group, which more properly should be called north-eastern and south-western, respectively. A third, transitional group is situated between them (Avanesa ŭ et al. 1968 ). 5 From Map 1 we can easily see that the BSCZ, where mixed rural Belarusian is spoken, crosses these traditional Belarusian dialect zones almost vertically, although to a large extent it belongs to a special agglomeration of dialects traditionally called the ‘Northwestern dialect zone’ (Bel. pa ŭ nočna-zaxodnjaja dyjalektnaja zona ). Table 1 Varieties in contact with rural Belarusian in the BSCZ Hrodna – Lida Dieveniškės, Šalčininkai, Eišiškės Pabradė, Švenčionys, Adutiškis Pastavy – Brasla ŭ Latgalia regional Russian, PolKres regional Russian, PolKres, Lithuanian regional Russian, PolKres regional Russian, PolKres, Latgalian, Latvian (?) B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 189 6 Most of these features have been included in List 1 , but we will consider some others, too. Table 1 helps us understand why rural Belarusian plays a central role in the Baltic-Slavic border region: it occurs in virtually all areas regardless of what other languages are used. Regional varieties of these languages have – to diff er- ent degrees and depending on the specifi c region within the BSCZ – had an impact on the formation of Belarusian mixed dialects. It is, therefore, any- thing but astonishing that, in a sense, Belarusian dialects of the BSCZ have incorporated features and highlight tendencies that are common in most or even all contact varieties in this region. A great deal of its rise and seemingly chaotic variation can most probably be explained by the dialect mixing that was enhanced by migration (caused, e.g., by the abandonment of serfdom in Tsarist Russia 1861, by wars, and by inter-war Polish as well as post-war Soviet policy) and also by its low-prestige, uncodifi ed and only oral basis. It is exactly its “unembellished”, low-prestige status that makes rural Belarusian a typical representative of the BSCZ and an outstanding candidate for the study of contact-induced variation. By the same token, the features and the varia- tion we encounter in these dialects are nothing special in comparison to what we can observe in the contact varieties used in the same area. Th ese points will become central in the further course of this paper. In what follows, we will fi rst point out some methodological intrica- cies caused by the endeavor to evaluate the contact phenomena found in rural Belarusian from a broader perspective of the Eastern part of the CBA (section 1.1). Th en we will briefl y discuss features that give evidence for a tight interconnection of Belarusian rural dialects with PolKres (1.2) and spell out the reasons why this is so (1.3). In section 2, we will elaborate on the question of whether and how one can distinguish between the reasons and triggers for the structural phenomena (and their variation) encountered in rural Belarusian: whether they are to be sought after in internal development or belong to fea- tures of inner-Slavic dialect continua, or whether they are contact-induced and, if the latter is the case, whether we can distinguish inner-Slavic from Baltic infl uence. In particular, we want to discuss a couple of features in greater detail, which have had attention paid to them in the dialectological literature of the Slavic varieties spoken in the BSCZ. 6 Our guiding questions are: (a) Which features of rural Belarusian are atypical in terms of traditional Belarusian dialectology? Or otherwise: which features of rural Belarusian of the BSCZ testify to its mixed character violating boundaries between previously established dialect groups? 190 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 (b) Which features of rural Belarusian are manifestations of change, and which are conservative Belarusian features of the given localities? (c) To what extent can we distinguish contact-induced change from the out- comes of internally motivated evolution? It is plausible to assume that, in the BSCZ, both factors, more often than not, go hand in hand. We may thus ask more particularly: can we tell triggers and subsequent processes of propagation apart? (d) Does the degree of genealogical closeness matter and, if so, to what extent? More precisely: are there features more resistant to infl uence from Lithu- anian than others for which Lithuanian infl uence is very likely or even obvious? Section 3 summarizes our fi ndings and contains an attempt at answering whether the degree of genetic closeness between varieties in the Baltic-Slavic region is of any signifi cance for contact-induced results in the structure of rural Belarusian (and partially also other varieties). 1. Local signifi cance and areal linguistic background 1.1 Local variation as parts of areal clines Th e genealogical affi liation of the East Slavic (Russian, Belarusian) and West Slavic (Polish) varieties in the BSCZ is very close. Th e East Slavic varieties belong to a single dialect continuum, and some morphosyntactic features of Belarusian are, in turn, transitional between Russian in the (North)East and Polish in the (South)West; this is partially refl ected in dialect groupings accepted in traditional Belarusian dialectology (see above). As for Baltic, Lith- uanian and Latgalian show many features shared with local varieties of East Slavic and Polish although they are genealogically “farther” apart. Th is refl ects the traditional groupings into families within IE., among which, how- ever, Slavic and Baltic have always fi gured as particularly intimately related. Whether and to what extent this is due to common ancestry or caused by long-lasting contacts (after a split within IE. had occurred) has largely remained an unsolved question. Nonetheless, it is indisputable that, in loca- tions where speakers of Baltic and Slavic dialects have come into intimate and long-lasting contact, heavy traces of (mutual or one-sided) infl uence have become apparent. Th e basic problem consists of determining which conver- gent features have really been triggered and/or expanded by contact with varieties of the respective other group, and which features have become com- mon on the basis of the formation of dialect continua (transitional zones). B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 191 Wälchli’s comments (forthcoming: ch. 1) are applicable here: ‘Contact- induced change is (…) more likely to occur between closely related languages than between unrelated languages’. Among closely related languages, contact- induced change ‘is diffi cult to distinguish from inherited common innova- tions’. Th e problem is aggravated if we apply this question to features inside the Slavic family, or even only its Eastern branch, to which Russian and Belarusian belong. Th is basic problem has been detailed in recent work on areal typology and population linguistics, cf. again Wälchli (forthcoming: ch. 1): Even though neighboring languages with a history of language contact will always exhibit many partial structural similarities, language contact is only one of many factors leading to language change and so it is often impossible to prove diff usion in individual cases. (…) quirks are only a small proportion of all contact phenom- ena (only the tip of the iceberg). Contact-induced change is more likely to occur in cross-linguistically frequent features where it will always be claimed that the areal explanation is not necessary because such features develop easily without external causes in very diff erent areas. In other words, the less spectacular observed features (or changes) are, the more diffi cult it becomes to diagnose the reason for areal convergence. Further- more, Wälchli points out that conclusions drawn from the emergence of con- vergent features in contact languages seem to depend on the formulation (or hidden assumption) of what we may call the ‘null hypothesis’. Whereas areal linguists explicitly or tacitly assume null hypothesis (A), typologists tradition- ally have relied on null hypothesis (B): (A) Lack of contact phenomena even between neighboring languages. (areal linguistics) (B) Contact phenomena are omnipresent even across macro-areas. (typology) In either case the motif is easy to disclose: typologists are primarily interested in the regularities hidden behind structural variation and are eager to deter- mine the limits of variation which natural languages are capable of; to this end they try to achieve insights into the widest possible range of phenomena and have been keen to avoid any “disturbance” from external factors. Th is reason- ing yields the ground for sampling procedures in the Greenbergian tradition. By contrast, linguists interested in the geographical distribution of structural patterns are eager to discover areal biases and, moreover, to explain areal clines (continua or clusters across languages and language families) by contact. In either case, the respective null hypothesis should be easy to disprove but also serve as the basis for a kind of alpha-error (to maintain the analogy with 192 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 7 On this caveat concerning approaches from population linguistics cf. Wiemer ( 2003c : 216f.; 2004: 498-501). 8 Cf. Wiemer/Hansen (forthcoming: ch. 3) for a justifi cation of this discrimination with respect to contact-induced grammaticalization. statistics), i.e. to track down (and then exclude) cases that are able to run counter to the desired overall picture. Moreover, the reasoning typical for areal linguistics resembles dialect geo- graphy in principle whose representatives, following the founders of wave theories, have assumed that the spread of features (and isogloss bundles setting their approximate boundaries) comes about as a result of stepwise accommo- dation between speaker groups in restricted local areas: the accumulation of local areas chained one after the other sums up to more global eff ects. However, the analogy with assumptions in dialect geography is not perfect since “the question arises as to what extent micro- and macro-areal clines are really comparable” (Wälchli, forthcoming: ch. 1). For this reason, one has to be very careful in transferring methods applied by (and results achieved in) areal lin- guistics to more locally restricted areas such as the BSCZ. 7 With this caveat in mind, we will nonetheless take advantage of recent insights of areal linguistics insofar as our main concern will be to show that the assessment of what has been going on in the Baltic-Slavic region in general and with rural Belarusian in particular becomes more complicated and at the same time easier to judge if we account for its features from both a micro-areal and a more global perspective on (at least) the Eastern part of the CBA. Th is approach should be complemented by a further consideration. Namely, we have to distinguish between three phases (or steps) in the consolidation of structural features: (a) initiation of innovations, (b) propagation, (c) entrench- ment. If we do not (try to) do this, a discriminate treatment of system-internal factors vs. external (i.e. contact-induced) triggers proves impossible not only for the moment of innovation (initiation), but also for the spread of innovations. 8 1.2 Comparison with ‘borderland Polish’ (‘północna polszczyzna kresowa’, PolKres) As was remarked already in Wiemer ( 2003a : 232-235), it is almost impossible to distinguish between rural Belarusian and PolKres on the basis of structural features alone, simply because most features mentioned as typical for PolKres are of a clearly East Slavic provenance; often they are shared by practically all Slavic and Baltic varieties of the BSCZ or even farther beyond. Th is causes B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 193 9 Information on Lithuanian dialects by courtesy of V. Kardelis (Vilnius). 10 Varies with “Polish manner” (cp. [xodźit ɕ , dźeń] etc.) and “Russian manner” (cp. [xad’it’, d’en’, bat’ka]) to the extent that three degrees of original /d’, t’/ occur mixed up: Russ. [d’, t’], Bel. [dz", c"], Pol. [dź, t ɕ ]. diffi culties in tracing the roots of the feature in question. Th e fact that Polish scholars have mentioned such features as “peculiarities” of PolKres – although as such these features do not discern PolKres from Belarusian of the same region – partially arises from the specifi c focus of attention paid to PolKres, namely: these features have been highlighted in comparison to standard Polish. However, standard Polish (as well as standard Russian) is in many respects atypical; it is the standard varieties exterior to the BSCZ that deviate from the varieties within the latter, not the other way around! Th erefore, using these standards as a sort of gauge against which local varieties of the BSCZ are described distorts the picture (cf. Wiemer 2004 ). For instance, Grek-Pabisowa/Maryniakowa (1999: 19-41), in their intro- duction to a volume of representative texts of PolKres in this borderland region, pointed out features of PolKres, among which at least the 17 shown in List 1 coincide with what we encounter in local Belarusian. Virtually all of them are characteristic of other contact varieties in this region, too. List 1 : Features shared by PolKres and rural Belarusian (not exhaustive) 9 I. Segmental and suprasegmental phonetics, phonotactics (i) Word stress on last syllable; e.g., PolKres pa ˡ t ͜ ʃ ai ‘look!’. All varieties of the BSCZ, except PolKres (penultimate stress) and Latgalian (ini- tial stress), have movable word stress, which also allows fi nal syllables to be stressed. (ii) No penultimate stress in PPs if preposition and noun are monosyllabic; e.g., PolKres do i ̯ éi ̯ ‘to her’, n’e mám ‘I don’t have’. Th is phenomenon is just a consequence of the violation of clitic rules (together with penulti- mate stress) obtaining in standard Polish. In no variety of the BSCZ do clitic rules work. (iii) So-called C’ekann’e/Dz’ekann’e : the etymological dental plosives [d, t], when palatalized, become aff ricates [dz", c"]; e.g. PolKres xadz”ic” ‘to walk’, dz”en’ ‘day’, bac”ka ‘daddy’. 10 Th is is a very salient Belarusian fea- ture, but it also occurs in Lithuanian insular dialects and in Southern Aukštaitija (Dzūkian dialects), although probably for a diff erent reason. 194 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 (iv) Palatalized consonants can also occur before back and middle vowels; e.g., PolKres pouc”akal’i ‘(they) ran away’, zav’azu ‘I’ll tie up’, z”aml’u ‘land. acc’ , l’on ‘linen’. Th is feature is common to northern Belarusian and Russian as well; it is also attested for Aukštaitian dialects, although the reason is diff erent (disappearance of * j ). (v) Obviously free variation of [n, s, z, t s] vs. [n’, s’/ś, z’/ź, t s’/t ś]; analogous to feature (iii). (vi) Mostly voiced /v/ in the groups /kv, sv, t sv, xv, ʃ v, t ʃ v/; e.g., PolKres [kv’at, śv’at] ‘fl ower, world’, instead of standard Polish [kf ’at, śf ’at]. Lack of assimilation of /v/ after these voiceless consonants is common to all varieties of the BSCZ; the odd variety out is standard Polish. (vii) Obviously free variation of [x] and [ γ ]. Th e same can be observed not only in rural Belarusian, but also in some Lithuanian dialects of the BSCZ. (viii) Pro- and epenthetic (semi-)consonants; e.g. PolKres vušy ‘ears’, jon / jana ‘he / she’. Th is feature can be encountered in many non-standard varieties practically all over Slavic as well as in Lithuanian dialects. (ix) Akan’e. Th is term refers to the reduction of etymological /o/ in unstressed syllables. It exists in diff erent variants but is characteristic for the entire middle part of the East Slavic dialect continuum; this includes middle Russian dialects (and standard Russian, whose basis are the middle Russian dialects of the Moscow – Suzdal’ – Vladimir region) and the northern half of Belarus relevant for the BSCZ. In PolKres, akan’e occurs with no observable consistency, but examples like adná adnú ‘one the other’ (etymologically /o/), vajná tája pačalá s'e ‘that war began’ (cp. standard Polish: wojna ‘war’, początek ‘beginning’) are well known. A phenomenon similar to akan'e is also encountered in the Latgalian dialect (Seržant 2010). Although its rise in Latgalian has to be explained on a diff erent basis (cf. Seržant 2005 ; 2010), it is striking that, despite probably independent (internal) motivations for the appearance of this phenomenon in both East Slavic and Latgalian, similar lines of development converged at some time. As a result, at later stages, bi-/ multilingual speakers at a synchronic level are likely to identify akan’e in these varieties. (x) Apparently free variation of [v – ŭ – u]; e.g. PolKres [ f s’o p’eraval’i ŭ ] ‘(he) turned all over’ vs. [ ŭ s’o, ŭ s’ayo ], [ ŭ že ] vs. [ uže ] ‘already’, [ dz”ec’i u plat ʃ ] (vs. [ v pla t ʃ ]) ‘the children began to cry’. Th is, again, is also typical for rural Belarusian and Lithuanian dialects of the BSCZ. (xi) Palatalized pronunciation of /x/, i.e. [x’i, x’e] (instead of [x ɨ , xe]); e.g. PolKres [mux’i, mux’e] ‘fl y. gen/dat ’. Th is pronunciation is the norm in northern Belarusian and Russian and can as well, although much more rarely, be encountered in Lithuanian. B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 195 11 A list of abbreviations is given at the end. If not indicated otherwise, data are taken from tape-recordings made during fi eld work in 2000-2004. II. Morphology (xii) Ending {u} instead of {ov’i} to mark dat.sg of the most productive mas- culine declension. Th is conforms to {u} being the exclusive ending in Russian. It seems to preponderate in rural Belarusian, too. (xiii) Ending {ai ̯ } instead of {o ̰ } for ins.sg of the most productive feminine declension. Th is corresponds to the East Slavic ending {oi } (under con- ditions of akan’e, see feature (ix)). (xiv) Unifi ed ending {ov} (phonetically [of, af ] or [ou ̯ , au ̯ ]) for gen.pl in all declensions (cf. Erker, 2009 ; forthcoming). Th e unifi cation of one end- ing for the gen.pl regardless of gender and declensional class belongs among the most persistent features throughout the BSCZ (compare, e.g. Lith. {ų}). A tendency toward unifying {ov} as the single gen.pl - marker can furthermore be observed in many other varieties of north Slavic beyond the BSCZ (see 2.1). III. Syntax (xv) Use of the genitive to mark indefi nite quantity for subjects and objects (see ex. 1a-b). Th is is another of the most pervasive features of all varie- ties in the BSCZ (see 3.2). (xvi) Predicative use of indeclinable anteriority participles (ending in {vši}) leading to resultatives (perfects); see ex. (2). Th is is likewise a ubiquitous feature of all varieties of the BSCZ. (1a) xl’ éba b ýla / kart ófl ’e b ýlo // (Str.) 11 bread. gen be. pst.3sg.n potato. nom.pl (?) be. pst.3sg.n ‘there was bread / there were potatoes’ (1b) i an’ í tak’ íx sm’ešn ýx anegd ót’ikaf rask á zyval’i // (Ad.) and they. nom such funny. gen.pl joke. gen.pl tell: ipfv.pst.3pl ‘and they told (us) very funny stories ’ (2) dz’ádz’ka byu ̯ pryi ̯ éxau ̯ šy , nu i pai ̯ éxal’i (Br.) uncle. nom be. pst.3sg.m arrive. ptcp ptc ptc go: pfv.pst.3pl ‘(our) uncle had come [lit. ‘was arrived’], so that we went off ’; ma i ̯ é rúk ’ i paščapáu ̯ šys’a (Br.) have.prs.3sg hand. acc.pl become_rugged. ptcp:refl ‘(s/he) has (his/her) hands (which have become) rugged’ ; 196 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 navučý u ̯ šy nas bylá, ni klása ni kónčy u ̯ šy (Lid.) teach: pfv.ptcp we. acc be. pst .3 sg.f neg class.gen neg end: pfv.ptcp ‘(she) had taught us, (although she) hadn’t (lit. didn’t) fi nish a class’; nas byl ’ í spal ’ í u ̯ šy (Br.) we. acc be. pst .3 pl burn_down. ptcp ‘(they) had burnt us’. IV. Discourse (propositional modifi cation) (xvii) Use of epistemic particle mus’i (< 3. sg.prs of mus’ec’ ‘must’); see ex. (3). Another commonplace feature in Slavic and Baltic varieties of the BSCZ. See some examples from rural Belarusian: (3) vot / γ etyx γ rošau ̯ u m’en’a było / ptc this . gen.pl money . gen.pl at me . gen be . pst .3 g.n mus”i i ̯ ak’ix dv’enaccac” tys”ač must_be indf. gen.pl twelve thousand . gen.pl ‘well / I did have some money / probably some 12 thousand (Latgalia; cited from Jankowiak 2009 : 155) Th ese 17 features are by no means an exhaustive list of micro-areal coinci- dences, but they provide a good idea of what comes to a dialectologist‘s eye (or rather: ear) if one looks at the data from outside the BSCZ and restricts one- self to merely enumerating what has been encountered. If, however, we restricted ourselves in this way, we would arrive at the strange conclusion that Belarusian and PolKres do not diff er in the BSCZ, or only scarcely. Th is is in fl agrant contrast to what speakers of these varieties as well as fi eld workers know: there is a distinct, well-perceived diff erence between both varieties not only in sociolinguistic, but also in structural terms. Everybody who has been in the fi eld and has listened to recordings of speech can usually tell whether a particular stretch of discourse is in Belarusian or PolKres. While we are not going to dwell on sociolinguistics in more detail, the lesson to be drawn from this weird superfi cial conclusion is that, if we want to render a proper linguis- tic description of these two varieties, we must integrate frequency patterns, more precisely: proportions of occurrence of variants (“allo-realizations”) both in the paradigms (types) and on discourse level (tokens; cf. Wiemer 2003a : 236; forthcoming1). And we should probably distinguish salient features from less salient ones, on both the acoustic and the morphosyntactic level. 1.3. Reasons that have led to coincident features of PolKres and rural Belarusian After all, what is the reason for this overwhelming convergence of struc- tural features in Belarusian and PolKres? Th ough the historical and social B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 197 12 According to Morita (2006: 127), most specialists on PolKres even assume that the Polish- speaking inhabitants of the villages use Belarusian ( Mowa Prosta ) more often than PolKres in their everyday life. 13 Th is concerns at least its rural variety in the border region. Th e urban variety of Vilnius, which is closer to standard Polish, is beyond the scope of this paper. 14 Turska ( 1995 ) considered that, after the abolition of serfdom, many Lithuanians began to switch directly to Polish without the mediation of Belarusian. However, she herself also noted a faster expansion of Polish particularly among those Lithuanians who previously had already changed to Belarusian (1995: 52). No less important is the consideration that the Lithuanian peasants were most probably able to communicate in Belarusian, so that the latter has to be considered as an adstrate anyway. For a more detailed discussion cf. Wiemer ( 2003b : 124-127). background is very complicated, part of the explanation presents itself quite readily if we realize that most (all?) rural speakers of PolKres also use Belarusian, at least in the localities where we are adducing data. 12 Moreover, Belarusian and PolKres 13 evolved in the same region most probably because the Belarusian and Lithuanian speaking population was considerably infl u- enced by Polish or even switched to Polish during the course of a few genera- tions. According to Turska ( 1995 [1982]), this process started after serfdom had been abolished in 1861 and became rapid at the latest during the interwar period when, in 1920, the whole region found itself under Polish rule and schooling was conducted exclusively in Polish. In this period, many Belarusian or Lithuanian speakers did not accomplish a transition toward Polish but con- tinued using Belarusian vernaculars 14 as Belarusian had already most probably been serving the role of a mediator variety for several centuries, i.e. from the times of the Grand Duchy of Lithuania. For the same reason, Belarusian ver- naculars must have constituted an adstrate or substrate for all speakers at least in the countryside of the Vilnius region where Lithuanian and Belarusian dia- lect zones overlap in intricate ways. Th e contemporary Belarusian – Lithuanian border runs right through the middle of the half moon-like shaped stretch of land referred to in the intro- duction in our comment on Map 1 . Th e spots where PolKres occurs are also well-established spots of Belarusian usage. Th e single diff erence is that, in comparison to PolKres, Belarusian certainly occupies a more compact terri- tory with fewer “holes”. Given the assumptions about the genesis of rural PolKres in the BSCZ (see above), one can say that, to a large extent, rural mixed varieties of Belarusian developed at least partially as an off spring of the at times quite rapid switch from regional East Slavic and Baltic vernaculars toward Polish. 198 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 15 For East Slavic cf., among others, Kuznecov ( 1954 : 75; 1973: 250-260), Lipovskaja ( 1983 ). 16 Cf. Urbańczyk (1984 7 : 45f.) for Polish, Skudrzykowa et al. ( 2001 : 192) for Silesian Polish, Breza/Treder (1981: 112) for Kashubian, SRA (1975/xi: 218-223) for Sorbian dialects. 2. Th e Baltic-Slavic contact zone as part of larger areas Th e 17 features included in List 1 diff er in terms of their relevance for areal linguistics. Some of the features appear to be really restricted more or less to the BSCZ or even to its Slavic varieties alone. Other features can reasonably be compared to a distribution within Slavic as a whole as they belong to dia- lect continua or are “scattered” without a connection to the BSCZ (see 2.2.1- 2.2.2). Further features are also well known in non-Slavic languages (varieties) of neighboring regions, above all Finnic ones (see 2.3). When speaking about convergence in that relatively restricted area which forms part of larger areas (as in a model with concentric circles), we have to distinguish between these layers of diff erent range. Even though we still have quite insuffi cient know- ledge about Belarusian in the BSCZ, we see that it is in these mixed varieties that these layers become manifest with their probably densest concentration. From among the 17 features given in List 1 , let us fi rst choose features related to nominal infl ection in order to demonstrate the intricacy of the ques- tion formulated in 1.3 above, in particular for question (c). 2.1 On telling Slavic-internal developments and Lithuanian infl uence apart Let us examine feature (xiv), the unifi cation of the gen.pl -ending. Here we are dealing with the reduction of (allo)morphic variation for which an ana- logical process can be observed in Lithuanian (and Latvian); nonetheless no convincing answer can be given to the question of whether and in which way Baltic varieties have participated in this process. Th e ending {ų} for all declen- sion classes is the norm in Lithuanian, but there are a few circumstances that diminish the probability that Lithuanian impact has been a decisive factor. Moreover, it turns out to be very diffi cult to even prove that Lithuanian had a strengthening eff ect in addition to other, major factors. Th ese circumstances are as follows: (xiv.a) Th e simplifi cation of the gen.pl -ending can be encoun- tered at many places in North Slavic; it has been noticed in dialectological textbooks and special investigations not only in the BSCZ or its immediate neighborhood, 15 but also in West Slavic down through its most extreme vari- eties in the West. 16 (xiv.b) Th e generalization of {ov} as marker of gen.pl is a B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 199 17 Cf. Zemskaja/Kitajgorodskaja (1984: 75). Zemskaja ( 2000 : 779) noted that this general- ization is also known from fi rst-language acquisition and the speech of Russian emigrants in Western Europe and Northern America. Th ese observations speak in favor of a common under- lying cognitive propensity responsible irrespective of (or in addition to) language contact. commonplace feature in studies on so-called (gorodskoe) prostorečie (‘substan- dard (urban) speech’) in Russian. 17 Th is shows that this phenomenon also has a widely known diastratic dimension. Both (xiv.a) and (xiv.b) make it unlikely for Baltic infl uence to have been a trigger for the unifi cation process; it tends to occur in central dialects as well as on the edges of Slavic (at least in the west) where no phonetic or functional parallels with nominal paradigms exist in non-Slavic contact dialects. Lastly, for (xiv.c), it is tempting to explain the spread of {ov} (at least partially) in the framework of Natural Morphology. Consider that, in North Slavic, {ov} as an ending is used in no other slot of case-number paradigms in Northern Slavic; it is salient and therefore an excel- lent candidate for being generalized in this function. Only here is a point where Lithuanian infl uence could have occurred: Lith. {ų} as marker of gen. pl is almost equally isolated in the case-number paradigms of nouns and adjectives; only the u -declension (exclusively masculine) also has it in the acc. sg (e.g., žmog-us ‘man, human being’ → žmog-ų . acc.sg , al-us ‘beer’ → al-ų . acc.sg , gudr-us ‘clever. m ’ → gudr-ų . acc.sg.m ). In a diasystem of multilin- gual dialect speakers, this pattern might have been strengthened since it favors paradigmatic economy. But can we discern whether the Lithuanian pattern served as a trigger or only strengthened the propagation of a tendency that was obviously already inherent to the Slavic varieties themselves? Note, moreover, that even if Natural Morphology could be used as an explanatory tool in this particular case, it cannot be generalized for other cases. Consider feature (xii), which will be further treated immediately below. It runs counter to what Natural Morphology would predict: {ov’i} as marker of dat.sg (with masculine nouns) has been ousted by {u} in most varieties of East Slavic (or become the marked choice), although it is no less salient and exclusive for dat.sg .m than is {ov} for gen.pl . Obviously, considerations based on type and token frequency have to be taken into account, too. In addition, the spread of the ending {u} at the expense of {ov’i} seems to have started much earlier than the unifi cation of {ov} and in several Slavic lan- guages has more or less come to an end. Th ere are other “peculiarities” of rural Belarusian beside those included in List 1 , which are well attested either elsewhere in Slavic (2.2) or in a broader area encompassing not only Baltic, but also Finnic languages (2.3). Let us browse through some of them. 200 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 18 An alternative explanation might be that the ending of the instrumental of the same inher- ited paradigm has been transferred to the dative. However, to our knowledge, this kind of case syncretism is totally unattested in East Slavic leading us to readily dismiss this alternative. 2.2 Further features attested elsewhere in Slavic We can further subdivide this subsection into features that are “scattered around” in Slavic, at least in North Slavic, or, more narrowly, in East Slavic (2.2.1), and features for which dialectologists of East Slavic have established continua and isoglosses for quite a long time (2.2.2). 2.2.1 Features attested in various Slavic regions We may assume that morphological features like (xii-xiii) are clearly restricted to inner-Slavic variation. What is at stake here is a choice among allomorphs that belong to a specifi c slot in the case (+ number) paradigm of Slavic nouns of diff erent genders. Since what is relevant for the variation are the forms of the allomorphs (not their abstract representations), it does not make much sense to compare them to anything beyond the particular Slavic varieties involved in our borderline area. Baltic (Lithuanian) could have only infl u- enced the functional range of particular cases and the inventory of grams. On the contrary, there is hardly any evidence that Lithuanian had an impact on the phonological shape of declensional markers. Th e only case known to us is the allomorph {ai} as marker of dat.sg of the a -declension (mostly feminine nouns). Th is can be inferred from examples like (4a-b), in which {aj} and {oj} occur as allomorphs of the dat.sg of this declension class (instead of expected {e}); {aj} conforms to Lith. {ai} phonetically, used in a stressless syllable, while {oj} has to be interpreted as another realization under stress. Accidentally, this makes s’astroj a grammatical homonym of the instrumental form (which, how- ever, would not render any sensible interpretation), but actually (4a) could be understood as demonstrating a further step after the assumed borrowing from Lithuanian happened. Namely: {ai} must have become entrenched in the par- adigm to a degree that it cannot only replace the inherited ending {e}, but also even show phonological alternation according to the usual rules of Akan’e (for which see comment to feature (ix) in List 1 ): 18 (4a) a na što ž nam byló spláčyvac’ s’astr ój (Br.) and on what ptc we. dat be. past .3. n pay_off : ipfv.inf sister. dat.sg ‘and why did we have to pay off [the debt] to my sister?’ (4b) mal’čyk byu ̯ n’ev’al’íki, boy. nom.sg be. past .3. m small. nom.sg.m B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 201 19 According to Anan’eva ( 2008 : 77) they have also been attested, albeit rarely, in PolKres. 20 In Russian and Polish themselves {iva} and { ɨ va} are allomorphs. 21 A Russian and/or Polish infl uence seems to also be obvious in Lithuanian dialects of the BSCZ. In many of these dialects, especially on the territory of Belarus‘, proliferous use of mátk aj za rukú učap ’íu ̯ šys’ a išou ̯ (Lid.) mother. dat.sg at hand. acc.sg seize: pfv.ptcp:refl and go: ipfv.past .3 sg.m ‘the boy was rather small, he seized his mother’s hand and went (with her)’ (lit. ‘seized for (his) mother at (her) hand’; the dative codes an external possessor) Such instances only occur exceptionally in our records (cf. Erker, forthcom- ing: 2.3). 19 However, according to Grinaveckene/Mackevič (1979: 52-56), such forms have been attested in Belarusian dialects of the ‘Northwestern dia- lect zone’ (see section 1) where Lithuanian infl uence until the most recent past is plausible. Regardless of such rare instances, as such, features like (xii-xiii) are quite widespread within Slavic, also outside the BSCZ. Th eir variation arose from the interference with case-marking patterns whose allomorphs derive from diff erent declensional classes; these patterns have undergone redistribution for many centuries (probably since the late Common Slavic period, i.e. from the 6 th -7 th c. AD). Even in closely related varieties (like the East Slavic ones), these patterns could have “merged” in the course of usual dialect mixing, and no external, non-Slavic infl uence is required to explain this unstable variation in rural Belarusian. A similar remark concerns the realization of animacy patterns (see 2.2.2) as well as the allomorphs of the verbal suffi x {va} vs. {iva / ɨ va}. 20 Th is suffi x is a productive device for the derivation of imperfective stems. Th e allomorph {va} is characteristic of Belarusian and absent as a productive device in both Russian and Polish. According to Mackevič ( 1959 : 90), in traditional Belarusian dia- lects, the forms {va} and {iva} are distributed by a simple morphonologi- cal rule: the latter occurs only after consonant clusters ending in a sonorant (e.g. vykarm l ̱ -iva-c’ ‘to rear’, peradraž n ̱ -iva-c’ ‘ to tease (by imitating)’), in all other cases {va} applies (e.g. raskaz-va-c’ ‘ to tell (a story)’). However, in our Belarusian data {va} occurs only sporadically with {iva, ɨ va} being used instead; e.g. raskaz-yva-la ‘she told’ (Br.), raz γ avar-yva-jim ‘we talk (with each other)’ (Br.), abmán-yva-je deceives’ (Lid.), raskuláč-yva-li ‘(they) dispossessed the kulaks’ (Lid.). Th is can be explained by the infl uence of Russian or Polish where {iva} is very productive; on the other hand, this inner-Slavic infl uence certainly builds on the availability of {iva} as an originally morphonologically conditioned allomorph (beside {va}). 21 202 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 the suffi x {(d)inė} has been attested. Note that this suffi x is not borrowed but demonstrates a productive calque (or selective copy) of a derivational technique (with an unclear distribution of functions; for a critical overview cf. Kardelis/Wiemer 2003: 59-66). Furthermore, it is an open question as to whether this calquing process of a Slavic pattern was not sustained predominantly by Belarusian dialects, which previously might themselves have adapted to the Russian and Polish allomorph. Unfortunately, at the moment, such considerations have to remain specula- tions since no investigations of such issues have been produced. 22 Since the retreat of neuter nouns has been observed predominantly in dialects with akan’e, the “phonetic” explanation was often considered as decisive (cf. Kuznecov 1954 : 76f. and others). However, this explanation cannot apply in South Russian dialects (with Okan’e, e.g. in the group around Kursk-Orlov) and in other dialects, in which nouns ending in stressed [o] control agreement with the masculine or (more frequently) the feminine (e.g., molokò vytek : pf.past.sg.m ‘milk has fl own out’; kakaja . nom.sg.f molokò ‘which milk’, bol’šaja . nom.sg.f selò ‘large village’); cf. Avanesov/Orlova (1965 2 : 100), Kuznecov ( 1973 : 253) and others. 23 Th is loss also becomes obvious with other categories, e.g. with resultative participles (cf. Wiemer/Giger 2005: 29-33). Besides the gen.pl -ending, loss of neuter gender is another case attested in diff erent regions of East Slavic. In rural Belarusian (as well as in PolKres) this process has been noticed to variable degrees. It is plausible to consider Lithuanian (which has a two-gender system of the French-Italian type) as a supporting factor, but it is debatable whether it has been a proper trigger of this process (cf. Wiemer 2004 : 512). Th e point is that we fi nd too many spots in other regions quite far away from the Baltic-Slavic zone in which the same or similar processes have been observed. Th erefore, reduction or elimination of the neuter need not be explained by contact with Baltic. 22 On the other hand, a phonetic explanation which rests on akan’e (see comment to feature (ix) in List 1 ) cannot give an account of the whole story even for Belarusian, because neuter nouns take on the ending {i / ɨ } in the plural, not {a}. Th is looks like unifi cation of noun endings in the plural. In our data from Brasla ŭ and Lida, the ending {i / ɨ } of historically neuter nouns preponderates as well; e.g. sv’at-y ‘feasts’, vojsk’-i ‘troops’, saapščenn’-i ‘messages’. Th e original ending {a} occurs more rarely, e.g. sv’at-a ‘feasts’, čysl-a ‘numbers’, and adjectives in the singular can agree with these nouns in the masculine or feminine gender. Th erefore, it is plausible to assume that, at some stage of this process, analogy played a crucial role. Reduction or loss of neuter gender is also amply attested for in North Russian dialects (characterized by okan’e); cf., for instance, Čagiševa ( 1968 ). Yet another reason seems even more plausible for these dia- lects, namely contact with Finnic varieties. Finnic languages do not know gender at all; as a substrate or adstrate they have caused loss of agreement marking in Slavic dialects around Pskov and north to this region. 23 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 203 2.2.2 Features belonging to East Slavic dialect continua We have to bear in mind that, as a whole, Belarusian can, in structural and geographical respects, be regarded as an intermediary between Russian and Ukrainian (cf., among others, Kuraszkiewicz 1963 2 : 79-86), and also between Russian and Polish. As pointed out in section 1, the Belarusian dialects we are investigating are spoken on a stretch of land that, as it were, crosscuts the traditional division of Belarusian dialect groups. In line with this division, the patterns by which animacy is marked with case+number endings constitute a continuum on an East-West axis. Northeastern Belarusian is closer to (stan- dard) Russian whereas western Belarusian dialects “show patterns which are either totally consistent with standard Polish or are intermediate between Russian and the standard Polish pattern” (Wiemer 2004 : 510; consult for the areal and typological background). On the East-West axis the patterns shown by standard Russian and by standard Polish demonstrate a sort of extreme. Th ese can be exemplifi ed as follows: List 2 : Case marking (plural) for referential distinctions 1) In the nominative plural, standard Russian does not exhibit any animacy- related distinctions, i.e. animate and inanimate nouns are infl ected the same way; e.g. stol-y ‘tables’, student-y ‘students’, studentk-i ‘female stu- dents’, vnuk-i ‘grandchildren’, kot-y ‘tom-cats’. Standard Polish, on the contrary, distinguishes virile (= male adult) from the rest of the nouns; e.g. stoł-y ‘tables’, studentk-i ‘female students’, wnuk-i ‘grandchildren’, kot-y ‘cats’ vs. studenc-i ‘students’, pan-owie ‘gentlemen’. 2) In the accusative plural, standard Russian treats all animate nouns alike by using an ending which is identical with the ending of the gen.pl ( acc=gen ) , e.g. student-ov ‘students’, student(o)k-Ø ‘female students’, vnuk-ov ‘grand- children’, kot-ov ‘tom-cats’. Inanimate nouns have endings homonymous with the nom.pl (e.g. stol-y ‘tables’). Standard Polish, again, treats only virile nouns diff erently; the acc.pl is only identifi ed with the gen.pl with such nouns, all other nouns show acc=nom -homonymy. Compare, e.g., virile student-ów ‘students’, pan-ów ‘gentlemen’ vs. all other like stoł-y ‘tables’, studentk-i ‘female students’, wnuk-i ‘grandchildren’, kot-y ‘ cats’. Intermediate patterns in some Belarusian dialects demonstrate, for instance, acc.pl = gen.pl in nouns denoting women but not animals, and they lack a declensional distinction in nom.pl (cf. DABM 1963: 753-756 and the maps belonging to these comments). Th e acc.pl = nom.pl pattern for animals can be seen in the following instances: 204 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 24 u corresponds to v (as in Polish or Russian), its appearance is conditioned by phonotactic rules (see feature (x) in List 1). (5) us’é kúr-y ad nas pazab’irál’i (Lid.) all. acc hen. acc.pl from us.gen take_away: pfv .3 pl ‘they took all hens from us’; tréba byló karóv-y pás’ic’ (Lid.) be_necessary be. pst .3 sg.n cow. acc.pl shephard: ipfv.inf ‘it was necessary to shepherd cows’; kón’-i m’él’i (Br.) horse. acc.pl have. pst .3 pl ‘(they) had horses’. Against this background, it has, so far, been impossible to fi gure out whether speakers of rural Belarusian follow any clear pattern. Furthermore, in the records of the two locations, which we have systemati- cally studied, we can sporadically encounter inanimate nouns in the singular marked acc = gen : (6) máju star-ó γ a páspart-a (Lid.) have. prs .1 sg old. acc.msg = gen.sg.m passport. acc.msg=gen.sg.m ‘I have an old passport’; xto a γ urc-á máje, somebody. nom cucumber. acc.msg=gen.sg.m have .prs .3 sg xto pam’idór-a (Lid.) somebody.nom tomato. acc.msg=gen.sg.m ‘some have cucumbers, others tomatoes’ (lit. ‘who has (a) cucumber, who (a) tomato’); sad’is’ u 24 vóz’ik-a (Br.) sit: ipfv.imp.sg in(to) cart. acc.msg=gen.sg.m ‘sit into the cart’. One is tempted to explain such examples as calques (or even patent borrow- ings) from Polish, especially in Lida where PolKres is still more widespread than in the surroundings of Brasla ŭ . But we simply lack a fair enough number of similar attestations (including negative data) in order to be able to make any conclusions. Apart from this, the fi rst example in (6) is troublesome to explain as a straightforward polonism because the Pol. noun paszport does not allow for acc=gen . We thus would have to speculate that the speaker who produced this utterance must have further extended a pattern, which s/he might have taken over from some variety of Polish. B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 205 A further remark is in order here. We are unaware of any piece of evidence that suggests that the marking of animacy would have started (or been immi- nent) to develop in any Lithuanian dialect. As for Latgalian, Nau (p.c.) has noticed a few instances of masculine object nouns with animate referents whose case marking coincided with the genitive; however, whether this testi- fi es to an initial stage of animacy marking still remains to be investigated. Th us, as it seems so far, Lithuanian (and maybe Baltic in general) has been practically “immune” against this categorial distinction, even if all Slavic vari- eties that have ever been in intimate contact with Baltic dialects have displayed one or other ubiquitous and salient pattern of animacy marking. Looking at the phenomenon from this angle, we can therefore say that animacy (and, thus, diff erential object marking) has always been a purely inner-Slavic matter in the BSCZ. 2.3 Features attested in an area reaching both beyond Slavic and the BSCZ Th e last three features listed in List 1 are representative of this. We repeat them here for the sake of convenience: List 1 Features (xv-xvii), repeated syntax (xv) genitive of indefi nite quantity (‘pseudo-partitive’), for subjects and objects (see ex. 1a-b) commonplace (xvi) indeclinable anteriority participles (ending in {v ʃ i}) in predicative use ( → resultative or perfect); see ex. (2) commonplace (xvii) discourse (propositional modifi cation) commonplace epistemic particle mus’i (< prs .3 sg of mus’ec’ ‘must’); see ex. (3) Each of these features will have to be commented on separately. (xv) Genitive of indefi nite quantity Th is usage of the genitive as marking the Undergoer of either transitive ( → object) or intransitive ( → subject) predicates has been coined ‘pseudo- partitive nominal constructions’ by Koptjevskaja-Tamm ( 2001 ). Th is usage is widespread in the entire Eastern part of the CBA, and it increases from south to north (cf. also Koptjevskaja-Tamm/Wälchli 2001: 650-660). We can thus 206 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 25 Its non-occurrence in (standard) Latvian is due to a general decrease of genitival use. To this extent, Latvian presents an anomaly to this areal cline (to be explained on other grounds). only expect this pattern to also occur in Belarusian spoken in the BSCZ. We actually fi nd it the usual choice in Lithuanian. 25 (xvi) Predicative anteriority participles Th is feature is one of the diachronically most persistent and stable ones not only in the BSCZ, but also in the areas neighboring toward the Northeast. Th e only structural diff erence between Slavic varieties and Baltic ones is that the latter use participles infl ected for case (nominative), number and gender, whereas in the Slavic varieties they are indeclinable. Th e areal pattern of the predicative use of anteriority participles was analyzed in considerable detail by Wiemer/Giger (2005). Against a background of the BSCZ and the region to the Northeast of it, the occurrence of this type of participles is nothing extraordinary. Th ere are three striking pieces of evidence. First, predicative participles in {vši} only occur in Belarusian spoken in the BSCZ (see ex. 2) while being absent in the southern dialects; this is in line with their absence in Ukrainian and in the southern variant of PolKres (which developed on Ukrainian soil). Second, they are attested from verbs of diff erent diathesis types (transitive and intransi- tive ones), but they can only be subject-oriented. Th is conforms to the pattern in Baltic (Wiemer/Giger 2005: 54f.). Together with this, and thirdly, there is no confusion with n/t -participles, which become more pervasive (in all diathe- sis types and for both subject- and object-oriented resultatives) in the area north(east) from Pskov. All these facts taken together yield a strong argument in favor of regarding the BSCZ as the south(west)ern “prolongation” of a con- tinuum which starts at Lake Ladoga and Lake Onega but with an inverse distribution of suffi xes ({vši} vs. {n/t}) over diathesis types (Wiemer/Giger 2005: 33). (xvii) Epistemic particle As recently pointed out by Nau (forthcoming), Slavic and Baltic varieties that have come into contact with each other since the time of the Grand Duchy of Lithuania show considerable convergence in their system of expressions for modal meanings. Below we reproduce one of her tables. As we can see, almost all varieties have lexicalized expressions of epistemic necessity and possibility. To these varieties we should, following Nau, add the Latgalian particle (or sentence adverb) muszeń , which denoted epistemic B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 207 Table 2 Lexicalizations of ‘can be’ and ‘must be’ as epistemic particles in the BSCZ (= N. Nau (forthc.): Table 2 ) derived from LITH PolKres BEL UKR RUS meaning range ‘can (be)’ gal, galbūt może moža, mo, moža być možet, možet byt’ ‘maybe’, modal particle ‘must (be)’ ( musėnt etc., only dialectal) musi, musić musi być musi, music’ musi byc’ musit’ buti ( dolžno byt’, weakly lexicalized) ‘certainly’, ‘probably’, ‘maybe’ ‘has to be’ turbūt mabyc’ mabut’ necessity 100 years ago but can still be encountered in contemporary speech (Nau, p.c.). Nau convincingly argued for the assumption that language variet- ies of the former Grand Duchy of Lithuania have suffi ciently infl uenced each other so that eventually a convergent subsystem of epistemic particles emerged. Th is subsystem is distinct from what we fi nd in surrounding varieties (includ- ing standard Polish, Russian, and Latvian). We may thus argue for contact as the decisive factor in the appearance of this subsystem. While Latgalian muszeń is most obviously a borrowing either directly from Polish (PolKres?) or via some northeastern Lithuanian dialects, we can only speculate about the manner in which the equivalent units found their way into the other contacting varieties and established themselves there. Cf. Nau (forthcoming, p. 30): Regional Polish musi ‘certainly’ may be a calque (a “replica lexicalization”?) from Belarusian, but also from Lithuanian – or Lithuanian turbūt may be a calque from Polish, or Belarusian, and so on. Whatever happened, happened mostly in oral uses of the languages in previous centuries, and we are left without documents. 2.4 Broader areal features not attested in rural Belarusian So far we have discussed some representative and salient morphosyntactic fea- tures found in rural Belarusian; we have also dwelt on a broader areal back- ground against which features of rural Belarusian in the BSCZ have to be judged. We have argued that, on the one hand, there is virtually nothing in rural Belarusian that cannot be expected given the properties of its immedi- ately contacting varieties. We have furthermore detailed facts known from dialect geography also embracing the area of NW Russian dialects (around 208 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 Pskov, Novgorod and farther to the north) with which the BSCZ shows many parallels. Th is demonstrates that it would be better if the BSCZ were not con- sidered from an areally isolated point of view. On the other hand, we have indicated that there are features of rural Belarusian for which language contact need not be made “responsible” as the primary factor, in particular for which Lithuanian (or Baltic in general) should not be identifi ed as the main source of infl uence. Now we want to corroborate our argument that rural Belarusian is a faith- ful refl ection of features converging in its Slavic and/or Baltic contact variet- ies by turning the perspective, as it were, upside down. Indeed, we want to show that rural Belarusian does not have properties which are encountered only in contiguous areas, in particular the Pskov – Novgorod area, but not in the BSCZ. For we think that negative evidence can also be telling provided that we clearly oppose the smaller region which rural Belarusian is part of against broader areas beginning just beyond the BSCZ proper. Th e following fi ve features are characteristic for at least certain East Slavic dialects northeast from the BSCZ: 1) nom.pl.m ending in stressed {a}: Th is feature originated in north Russian dialects; from there it entered into standard Russian and has since been spreading among noun stems. In rural Belarusian it does occur, but extremely rarely. In the records rehearsed sys- tematically we have found only one token (see 7a), and it occurred with two other “variants” (see 7b) in the speech of the same informant: (7a) paspart-á ‘passports’; (7b) páspart-y , pášpart-y (Lid.). Th e realization paspartá can most probably be explained as a “loan” from standard Russian. Th e following four features are altogether unattested in our material of rural Belarusian: 2) Case homonymy (“syncretism”) of dat.pl and ins.pl with a unifi ed ending {im}, {am}, typical of the Russian dialects of the Pskov region; e.g. svo-im glaz-am videl ‘(I) saw (it) with my eyes’ (cf. Bromlej/Bulatova 1972: 99; Bukrinskaja et al. 2008 : 123). 3) Nominatival object; e.g. nado korov-a . nom doit’ . inf ‘(It’s) necessary to milk the cow’ (cf. Kuz’mina/Nemčenko 1964 for Russian dialects and Ambrazas 2001 on its genesis and spread in the whole area, in particular in Baltic and Finnic). 4) Non-agreeing participial passive with acc -NP; e.g. from North Russian dialects: B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 209 (8) vsju kartošku s”edeno all . acc.sg.f potato . acc.sg.f eat : pfv.ptcp ( indecl ) ‘(People) have eaten up all the potatoes.’ (cited from Sobolev 1998 ) Th is construction happens to be extended by the PP u +gen denoting the agent (cf. Trubinskij 1984 : 139-149; Sobolev 1998 with further references). Th is construction is not only well known from northern Russian dialects and has also been attested in some small south Russian “pockets”, but, above all, it is a productive construction in standard Polish. However, neither does its occurrence in standard Polish and the north Russian dialects rest on any shared dialect continua in the past; the evolu- tion of this construction arose on independent grounds (cf. Wiemer, forth- coming2: ch. 2, 3.2; Wiemer/Hansen, forthcoming: ch. 2.4, for a critical assessment) — nor is this construction known to Slavic dialects in the BSCZ, i.e. in a sense, it is absent exactly in the middle between two areas where it is well attested (Polish and north Russian dialects). 5) Postposed emphatic (demonstrative) “particles”; compare two examples from North Russian dialects (cited after Kasatkina 2008 : 311f.; see for further discussion): (9) Ogurcy ran’še-to ne sažali. cucumber. acc.pl earlier- ptc neg plant: ipfv.past.3pl ‘Earlier- ptc nobody planted cucumbers (here).’ (10) Škole-to otopl’enie nado, school. dat.f-ptc heating. nom/acc needed. aux(indecl) nu ja i vzjalasja . ptc I. nom ptc take_initiative: pfv.past.3sg.f.refl ‘Th e school- ptc needed heating, so I took the initiative.’ In no variety of the BSCZ have any similar tendencies been observed. 3. Assessment of fi ndings Let us now recollect the fi ndings, which are disparate not only relating to the kind of structural phenomena, but also lending themselves to diff erent answers on account of the questions (a-d) formulated in 1.3. Certainly, for the mo ment, the data we possess are too selective and the picture obtained on their basis too fragmentary for any more far-reaching conclusions. Nonetheless, some assump- tions of a methodological nature can be formulated, which could then also lead to some cautious preliminary conclusions concerning the motifs behind convergent features of the BSCZ that become manifest in rural Belarusian. 210 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 First of all, as far as we can judge by now, the high degree of variation of structural properties in the Belarusian rural vernacular has been due to con- stant, largely unhampered innovations in everyday oral communication; mostly these innovations have been propagated only within very restricted local communities of speakers, and rarely do they become fi rmly entrenched in some kind of internalized (in no case codifi ed) norm. In dialectological terms, rural Belarusian has to be characterized as a mixed variety containing combinations of features that cannot be reduced to traditional dialect conti- nua but which are situated somewhere, i.e. not quite predictably, on the con- tinuum of isoglosses between the Russian and Polish speaking territories. Th is can be seen particularly well with animacy marking (see 2.2.2) or the redistri- bution of allomorphs marking the dat.sg of masculine nouns (see 2.1). Th e variation of rural Belarusian, even on an idiolectal level, is considerable, and this has largely defi ed its description as a variety with a clear and “tidy” system. Its Belarusian morphology has been infl uenced by local Russian and Polish varieties, in part quite heavily, as is demonstrated, for instance, by the fact that the verbal suffi x {va} used for the derivation of imperfective stems has largely been replaced by {iva / ɨ va}. We have to, however, acknowledge that this allo- morph was not altogether unknown to Belarusian dialects even prior to heavy Russian infl uence since it was a minor choice in complementary distribution with {iva / ɨ va} depending on the stem fi nal consonant (see 2.2.1). As concerns question (c), i.e. the relative impact of contact-induced change and internally motivated evolution, our survey should have demonstrated that, for many features, one cannot sharply distinguish between results of con- tact with other Slavic dialects of the region or with adjacent dialects of Lithuanian (or Latgalian, which has not further been considered here). Th e reason for this diffi culty consists in the fact that all these dialects have been in close contact for centuries. Th e diffi culty is to some degree amplifi ed by the close resemblance of substantial parts of Slavic and Baltic morphology and phonology forming part of their common “heritage”; but for many details, it is diffi cult to decide whether the changes they have undergone have not occurred due to contact and, thus, mutual infl uence only quite recently. Th is problem manifests itself, for instance, in the weakening or even loss of the neuter gender, less so in the unifi cation of the gen.pl -ending in nouns (see 2.2.1, 2.1). Th e latter shows many parallels in other diatopic and diastratic varieties of Slavic, among which most do not show any closer relationship to the BSCZ; it is thus easier to explain this unifi cation on internal, purely cogni- tive grounds, e.g. by a tendency to paradigmatic economy and saliency. However, at least as an additional factor, contact cannot be dismissed even in such cases. B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 211 Generally, in many cases the results of change converge, at least superfi cially, even though they are most probably outcomes of diff erent factors. Cases in point are phonological and word prosodic features like akan’e, which occurs not only in entire northern Belarusian (to which the rural vernacular focused on here belongs), but also in some Aukštaitian dialects of Lithuanian, as well as aff rication of [d, t] and epenthetic [l‘] (see features (iii, iv, ix) in List 1 ). Note that when such features are investigated, the researcher – even if s/he can be suffi ciently sure about their unrelated origin in contacting East Slavic and Lithuanian varieties – has to account for the perspective of the dialect speaker for whom the genesis of what s/he hears in oral communication is not acces- sible. Instead, multidialectal speakers adapt to what they perceive in immedi- ate daily communication, in which case phonological and prosodic features salient to him/her are likely to be unifi ed in a sort of dialectal diasystem regardless of the motifs of their origin. Of course, the same applies, mutatis mutandis , for features in morpho(no)logy or syntax. As elaborated on in 1.1, the decision about whether, and to what extent, contact is the decisive factor in change hinges on the (implicit or explicit) null hypothesis assumed by the researcher provided s/he looks at a larger area showing clines for some specifi c structural features. If we look at the BSCZ from the broader perspective of the Eastern part of the CBA, rural Belarusian turns out to virtually in no way display any remarkable peculiarities since it fi ts perfectly into both the larger cline and the specifi c features of the BSCZ. Th is fi t holds for both positive and negative evidence: on the one hand, rural Belarusian shows features ubiquitous in most regions of the Eastern part of the CBA like, for instance, the case marking pattern of indefi nite quantity or the use of predicative resultative participles and the use of a particle marking epis- temic necessity ( mus’i ‘(most) probably’) common to virtually all varieties spoken on the territory of the former GDL (see 2.3). On the other hand, in rural Belarusian, we apparently do not encounter any kind of homonymy between declensional forms or of argument-coding patterns that are absent from the BSCZ but occur in the Russian dialect area neighboring to it in the northeast, (south)west and/or south (see 2.4). Moreover, this positive – negative balance of fi ts applies even if we examine the functional distribution of forms (or constructions) such as the resultative participles in {vši} (see 2.3) more closely. Th e restriction of their diathesis orientation is in perfect har- mony with the grammatical pattern in Baltic and diff ers (together with the Baltic perfect paradigm) from the functional distribution of resultative parti- ciples in NW-Russian dialects (where we in turn fi nd coincidences with Finnic). At this point, again, we have to add that it would be premature to conclude that Baltic infl uence is the trigger for this type and functional range 212 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 of resultative predicates to occur. For these participles have a very long, con- tinuous history in this part of Slavic (cf. Wiemer/Giger 2005: 40f.). Th e pro- ductive use of anteriority participles in the Baltic languages (inter alia as the core of their perfect paradigms) can, however, plausibly be regarded as a factor that has helped preserve the forms and functions of {vši}-participles in the Slavic varieties of the former GDL and their modern counterparts (including rural Belarusian). Finally we come to the question of whether genealogical closeness has an impact on structural features and which ones are more likely to be aff ected by contact. More precisely, is Lithuanian infl uence in rural Belarusian more restricted because it is less closely related than the surrounding Slavic varieties? Here we should admit that some features of List 1 have either remained entirely unaff ected by Lithuanian, or that such an infl uence is at least doubtful or can, to the best of our knowledge, be considered as marginal. Lithuanian infl uence is tangible in phonology, prosody and syntax (let alone in the lexi- con, which has not been considered here), but in morphology and morpho- nology its impact seems to be rather negligible. We have already commented on the unifi cation of the gen.pl -ending and the loss of neuter gender. Apart from these features, the following appear to be restricted to an “internal Slavic business”: (a) animacy distinctions; (b) the choice of derivational suffi xes of verbs; (the patterns of (East) Slavic have quite heavily infl uenced some Lithuanian dialects, but not the other way around) (cf. Wiemer 2009: 359-363) (c) declensional endings (with the rare exception of {ai} marking the dat.sg of the a -declension; see 2.2.1); (d) distribution of conjugation classes (not treated in this article); (e) morphonological alternations in the present stem (/d → dž/ for traditional Belarusian vs. /d --> ž/ for Russian). In fi nishing our pilot exploration of the specifi cs of rural Belarusian as the core representative of the BSCZ, we want to stress that we did not intend any exhaustive examination of its features. It has been our endeavor to show that rural Belarusian is the most typical representative of the BSCZ since its struc- tural properties, together with their enormous variation, demonstrate what one has to reckon with in unembellished rural speech in which typical features of the BSCZ converge most spontaneously. Th e outcome of diff erent layers on the intersection between East Slavic dialect continua, long-lasting contact with Baltic dialects and the interference of standards from beyond the BSCZ (Russian, Polish) and their local varieties provides an ideal playground for B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 213 multifarious imminent changes refl ecting the convergence of factors from these layers, to which we have to add tendencies toward change which are probably conditioned on cognitive grounds. From an areal linguistic point of view, rural Belarusian is sort of crucible of the BSCZ, it highlights, as it were, features of the latter, which can, in turn, be seen as part of a larger continuum. For the BSCZ can either concentrically be “inserted” into a larger area, or it can be viewed as a region where diff erent continua intersect. Th e whole region can thus be regarded as a buff er zone or something like a Zentrum einer Konvergenzlandschaft , as Stolz ( 1991 ) baptized the contact zone with Latvian, Livonian and Estonian as its core. References Ambrazas , Vytautas . 2001 . On the development of the nominative object in East Baltic . In Dahl Östen and Maria Koptjevskaja-Tamm (eds.), Th e Circum-Baltic Languages (Typology and Contact) , vol. 2: Grammer and Typology , 391 – 412 . Amsterdam-Philadelphia . Anan’eva , Natal’ja E. 2008 . Periferijnye pol’skie dialekty kak rezul’tat vzaimodejstvija s vostočnoslavjanskimi i baltijskimi jazykami. In Kalnyn‘ , Ljudmila Ė. (ed.), Issledovanija po slavjanskoj dialektologii 13: Slavjanskie dialekty v situacii jazykovogo kontakta (v prošlom i nastojaščem) , 72 – 79 . Moskva . Avanesa ŭ , Ruben , Kondrat Atraxovič , and Juzefa Mackevič . 1968 . Linhvistyčnaja heahrafi ja i hrupo ŭ ka belaruskix havorak . Minsk . Avanesov , Ruben I. , Orlova , Raisa G . 1965 2 . Russkaja dialektologija . Moskva : Nauka . Blinava , Ėvelina and Ėudakija Mjacel’skaja. 1969 . Belaruskaja dyjalektalohija . Minsk . Blinava , Ė. and Ėudakija Mjacel’skaja . 1991 . Belaruskaja dyjalektalohija : Praktykum . Minsk . Breza , Edward and Jerzy Treder . 1981 . Gramatyka kaszubska (Zarys popularny) . Gdańsk . Bromlej , Sof ’ja V. , Bulatova , Lidija N. 1972 . Očerki morfologii russkix govorov . Moskva . Bukrinskaja , Irina A. , Karmakova , Ol’ga E. , Ter-Avanesova , Aleksandra V. 2008 . Govory belorussko-russkogo pogranič’ja . In Kalnyn‘ , Ljudmila Ė. (ed.), Issledovanija po slavjanskoj dialektologii 13: Slavjanskie dialekty v situacii jazykovogo kontakta (v prošlom i nastojaščem) , 118 – 179 . Moskva . Čagiševa , Vera I . 1968 . K voprosu o kategorii srednego roda v mestnyx govorax . Pskovskie govory II : 146 – 159 . Pskov . DABM Avanesa ŭ , Ruben I. , Krapiva , Kandrat K. , and Juzefa F. Mackevič . (eds.). 1963 . Dyjalektalagičny atlas belaruskaj movy (Ustupnyja artykuly, davedačnyja matėryjaly i kamentaryi da kart) . Minsk . Erker , Aksana . 2009 . Faktor jazykovogo kontakta v issledovanii osobennostej fl ektivnoj morfologii ‘Movy prostaj’ (na primere okončanij roditel’nogo padeža suščestvitel’nyx) . In Scholze , Lenka and Björn Wiemer (eds.), Von Zuständen, Dynamik und Veränderung bei Pygmäen und Giganten (Festschrift für Walter Breu zu seinem 60. Geburtstag) , 293 – 303 . Bochum . Erker, Aksana. (forthcoming): Specifi ka imennyx paradigm v belorusskom smešannom govore na belorussko-litovskom pograničje. Balto-slavjanskie issledovanija 18. Moskva. Grek-Pabisowa , Iryda . 1992 . Język polski na Kresach północno-wschodnich. Legendy a rzeczywistość . Warszawa . 214 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 Grek-Pabisowa , Iryda and Irena Maryniakowa . 1999 . Współczesne gwary polskie na dawnych kresach północno-wschodnich . Warszawa . Grinaveckene , Ėlena (= Grinaveckienė) and Juzefa Mackevič . 1979 . Nekotorye neobyčnye mikrosistemy v dialektnoj strukture belorusskogo govora okrestnostej Opsa. Baltistica XV (1) : 52 – 65 . Hentschel , Gerd . 2008 . On the development of infl ectional paradigms in Belarussian trasjanka: the example of demonstrative pronouns . In Hentschel , Gerd and Sjarhej Zaprudski (eds.), Belarusian Trasjanka and Ukrainian Suržyk (Structural and social aspects of their description and categorization) , 99 – 133 . Oldenburg . Jankowiak , Mirosław . 2009 . Gwary białoruskie na Łotwie w rejonie krasławskim . Warszawa . Kardelis , Vytautas and Björn Wiemer . 2003 . Kritische Bemerkungen zur Praxis der Erstellung litauischer Wörterbücher, insbesondere von Mundarten – am Beispiel des slavischen Lehnguts und des ‘veikslas’ . In Ostrowski , Norbert and Ona Vaičiulytė-Romančuk (eds.), Prace bałtystyczne. Język, literatura, kultura , 45 – 72 . Warszawa . Kasatkina , Rozalija F. 2008 . Artikli, predartikli i psevdoartikli v slavjanskix jazykax . In Moldovan, Aleksandr M. , Petr M. Arkad’ev et al . (eds.), Slavjanskoe jazykoznanie (XIV Meždunarodnyj s"ezd slavistov, Oxrid, 10-16 sentjabrja 2008 g.) . Doklady rossijskoj delegacii , 305 – 321 . Moskva . Kittel , Bernhard , Diana Lindner , Sviatlana Tesch , and Gerd Hentschel . 2010 . Mixed language usage in Belarus: the sociostructural background of language choice . International Journal of the Sociology of Language 206 : 47 – 71 . Koptjevskaja-Tamm , Maria . 2001 . “A piece of the cake” and “a cup of tea”: Partitive and pseudo- partitive nominal constructions on the Circum-Baltic languages . In Dahl , Östen and Maria Koptjevskaja-Tamm (eds.), Th e Circum-Baltic Languages (Typology and Contact), vol. 2: Grammer and Typology , 523 – 568 . Amsterdam-Philadelphia . Koptjevskaja-Tamm , Maria and Bernhard Wälchli 2001 . Th e Circum-Baltic Languages: An Areal-Typological Aproach . In Dahl , Östen and Maria Koptjevskaja-Tamm (eds.), Th e Circum-Baltic Languages (Typology and Contact) , vol. 2: Grammer and Typology , 615 – 750 . Amsterdam-Philadelphia . Kuraszkiewicz , Władysław . 1963 2 . Zarys dialektologii wschodniosłowiańskiej (z wyborem tekstów gwarowych) . Warszawa . Kuznecov , Petr S. 1954 . Russkaja dialektologija (Učebnik dlja učitel'skich institutov) . Moskva . Kuznecov , Petr S. (red.) . 1973 . Russkaja dialektologija (Učebnoe posobie dlja studentov ped. in-tov po spec. № 2101 „Rus. jaz. i literatura“) . Moskva . Kuz’mina , Irina B. and Elena V. Nemčenko . 1964 . K voprosu o konstrukcijax s formoj imenitel’nogo padeža imeni pri perexodnyx glagolax i pri predikativnyx narečijax v russkix govorax . In Avanesov , Ruben I. (ed.), Voprosy dialektologii vostočno-slavjanskix jazykov , 151176 . Moskva . Lipovskaja , N.A. 1983 . Forma roditel’nogo padeža množestvennogo čisla suščestvitel’nyx ženskogo roda s osnovoj na parnyj tverdyj soglasnyj . In Avanesov , Ruben I. (ed.), Russkie narodnye govory (lingvogeografi českie issledovanija) , 93 – 99 . Moskva . Mackevič , Juzefa F. 1959 . Marfalohija dzejaslova ǔ belaruskaj move . Minsk . Morita , Koji . 2006 . Przemiany socjolingwistyczne w polskich społecznościach na Litwie (region trocki) i Białorusi (region iwieniecki) . Warszawa . Nau, Nicole. (forthcoming): Modality in an areal context: the case of a Latgalian dialect. In Wiemer, Björn, Bernhard Wälchli, and Björn Hansen (eds.), Grammatical replication and borrowability in language contact . Accepted by Mouton de Gruyter (Trends in Linguistics). Seržant , Ilja A . 2005 . Otnositel’naja xronologija osnovnyx fonetičeskix izmenenij v istorii verxnelatyšskogo dialekta . Acta Linguistica Lithuanica 55 : 39 – 89 . Seržant, Ilja A. 2010. Phonologische Isoglossen des Hochlettischen, Nord-Ost-Litauischen, Nord-West-Russischen und Weissrussischen. Baltic Linguistics 1, 193-214. B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 215 Skudrzykowa , Adolna , Jolanta Tambor , Krystyna Urban , and Olga Wolińska . 2001 . Gwara śląska – świadectwo kultury, narzędzie komunikacji . Katowice . Smułkowa , Elżbieta . 2010 . Problem zapożyczeń w warunkach rozpadającej się społecznej wielojęzyczności. Rozważania wstępne . Studia Białorutenistyczne 2010-4 : 283 – 295 . Sobolev , Andrej N . 1998 . O predikativnom upotreblenii pričastij v russkix dialektax . Voprosy jazykoznanija 47-5 : 74 – 89 . SRA: Faßke , Helmut (ed.). 1975 . Serbski rěčny atlas / Sorbischer Sprachatlas , vol. 11: Morphologie. Die grammatischen Kategorien – Die Paradigmatik des Substantivs . Bautzen . Stolz , Th omas . 1991 . Sprachbund im Baltikum? Estnisch und Lettisch im Zentrum einer sprachlichen Konvergenzlandschaft . Bochum . Sudnik , Tamara M. 1975 . Dialekty litovskogo-slavjanskogo pogranič’ja (očerki fonologičeskix sistem) . Moskva . Trubinskij , Valentin I . 1984 . Očerki russkogo dialektnogo sintaksisa . Leningrad . Turska , Halina . 1995 . O powstaniu polskich obszarów językowych na Wileńszczyźnie . Vilnius . [photomechanic reprint of: Studia nad polszczyzną kresową I . Ed. by Janusz Rieger and Wiaczesław Werenicz. Wrocław etc., 1982 , 19 – 122 .] Urbańczyk , Stanisław . 1984 7 . Zarys dialektologii polskiej . Warszawa . Wälchli, Bernhard. (forthcoming): Grammaticalization clines in space: zooming in on synchronic traces of diff usion processes. In Wiemer, Björn, Bernhard Wälchli, and Björn Hansen, N. (eds.), Grammatical replication and borrowability in language contact . Accepted by Mouton de Gruyter (Trends in Linguistics). Wiemer , Björn . 2003a . „ Mowa prosta“ - Präliminaria zu einer strukturellen Beschreibung . In Blankenhorn , Renate , Joanna Blaszczak , and Robert Marzari (eds.), Beiträge der Europäischen Slavistischen Linguistik (POLYSLAV) 6 , 227 – 237 . München . Wiemer , Björn . 2003b . Dialect and language contacts on the territory of the Grand Duchy of Lithuania from the 15 th century until 1939 . In Braunmüller , Kurt and Gisela Ferraresi (eds.), Aspects of Multilingualism in European Language History , 105 – 143 . Amsterdam, Philadelphia . Wiemer , Björn . 2003c . Zur Verbindung dialektologischer, soziolinguistischer und typologischer Methoden in der Sprachkontaktforschung (am Beispiel slavischer und litauischer Varietäten in Nordostpolen, Litauen und Weißrußland) . Zeitschrift für Slawistik 48-2 : 212 – 229 . Wiemer , Björn . 2004 . Population linguistics on a micro-scale. Lessons to be learnt from Baltic and Slavic dialects in contact . In Kortmann , Bernd (ed.), Dialectology Meets Typology (Dialect Grammar from a Cross-Linguistic Perspective) , 497 – 526 . Berlin , New York . Wiemer , Björn . 2006 . Osuščestvlenie soglasnyx v « move prostoj » kak pokazatel’ izmenenij v uslovijax kontaktov . Acta Baltico-Slavica 30 : 439 – 452 . Wiemer, Björn. 2009. Zu entlehnten Präfi xen und anderen morphosyntaktischen Slavismen in litauischen Insel- und Grenzmundarten. In: Scholze, Lenka and Björn Wiemer (eds.): Von Zuständen, Dynamik und Veränderung bei Pygmäen und Giganten (Festschrift für Walter Breu zu seinem 60. Geburtstag) , 347-390. Bochum . Wiemer, Björn. (forthcoming1): Metodičeskie zametki k issledovaniju „Movy prostoj“. Balto- slavjanskie issledovanija 18, Moskva. Wiemer, Björn. (forthcoming2): On the rise, establishment and continued development of subject impersonals in Polish, East Slavic and Baltic. In Kittilä, Seppo and Leonid Kulikov (eds.), Diachronic typology of voice and valency-changing categories. Amsterdam, Philadelphia. Wiemer, Björn. (in prep.): Die Mowa prosta: ein Mischidiom als Schmelztiegel im Sprachkontakt (Forschungsüberblick). Wiemer , Björn and Markus Giger . 2005 . Resultativa in den nordslavischen und baltischen Sprachen (Bestandsaufnahme unter arealen und grammatikalisierungstheoretischen Gesichtspunkten) . München . 216 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 Wiemer, Björn and Björn Hansen. (forthcoming): Assessing the range of contact-induced grammaticalization in Slavonic. In Wiemer, Björn, Bernhard Wälchli, and Björn Hansen. (eds.), Grammatical replication and borrowability in language contact . Accepted by Mouton de Gruyter (Trends in Linguistics). Zaprudski , Sjarhej . 2008 . Ma ŭ lenčaja akamadacyja i peraključėnne koda ŭ u pracėse mižkul’turnaj kamunikacyi: vypadak Belarusi . In Hentschel , Gerd and Sjarhej Zaprudski (eds.), Belarusian Transjanka and Ukrainian Suržyk (Structural and social aspects of their description and categorization) , 57 – 97 . Oldenburg . Zemskaja , Elena . 2000 . O jazyke russkogo zarubež’ja . In Zybatow , Lew N . (ed.), Sprachwandel in der Slavia. Die slavischen Sprachen an der Schwelle zum 21. Jahrhundert (Ein internationales Handbuch) , Teil 2 , 767 – 785 . Frankfurt/M. etc . Zemskaja , Elena A. and Margarita V. Kitajgorodskaja . 1984 . Nabljudenija nad prostorečnoj morfologiej . In Zemskaja , Elena A. and Dmitrij N. Šmelev (eds.), Gorodskoe prostorečie (Problemy izučenija) , 66 – 102 . Moskva . Zemskaja , Elena A. and Dmitrij N. Šmelev . (eds.). 1984 . Gorodskoe prostorečie (Problemy izučenija) . Moskva . http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Language Contact Brill

Manifestations of areal convergence in rural Belarusian spoken in the Baltic-Slavic contact zone

Journal of Language Contact , Volume 4 (2): 184 – Jan 1, 2011

Loading next page...
 
/lp/brill/manifestations-of-areal-convergence-in-rural-belarusian-spoken-in-the-WmtF05t4h1

References (53)

Publisher
Brill
Copyright
© 2011 Koninklijke Brill NV, Leiden, The Netherlands
ISSN
1877-4091
eISSN
1955-2629
DOI
10.1163/187740911X589280
Publisher site
See Article on Publisher Site

Abstract

Journal of Language Contact 4 (2011) 184–216 © Koninklijke Brill NV, Leiden, 2011 DOI 10.1163/187740911X589280 brill.nl/jlc Manifestations of areal convergence in rural Belarusian spoken in the Baltic-Slavic contact zone* Björn Wiemer a & Aksana Erker b a) Professor für slavische Sprachwissenschaft, Johannes-Gutenberg-Universität Institut für Slavistik, Jakob-Welder-Weg 18, D- 55099 Mainz, Germany wiemerb@uni-mainz.de b) Johannes-Gutenberg-Universität Institut für Slavistik erker001@uni-mainz.de Abstract Th is article combines methods and insights from dialect geography, areal and contact linguistics. It focuses on a specifi c, yet heavily understudied mixed dialect of Belarusian spoken in the entire Slavic-Baltic contact region. It turns out that on practically all levels (from phonetics to syntax), the variation of features encountered in this dialect is, from all varieties of that region, the most representative for convergence phenomena characteristic of East Slavic, Polish and/or Baltic vari- eties in contact with each other. Th is convergence is so far-reaching that, for instance, it seems impossible to distinguish the rural Belarusian vernacular from regional varieties of Polish on the basis of structural properties alone, despite the fact that these varieties are clearly perceived as diff erent by both native speakers and fi eld linguists. Simultaneously, features of the Belarusian dialect – and, thus, of the whole contact region which it most accurately refl ects – should be judged under the perspective of larger areal clines (in particular, within the eastern part of the Circum Baltic Area); this view is pursued on the basis of Wiemer (2004) and more recent insights into the areal distribution of structural features crossing family boundaries. Such areal continua, in turn, intersect with inner-Slavic dialect continua and phenomena occurring in vari- ous locally restricted “pockets” scattered around in Slavic. On the background of this, we approach answers to the problem of determining the infl uence of contact (with Baltic and/or Finnic) over “genetic heritage” and the question of which features are more “immune” against infl uence from genealogically less close contact varieties. Keywords Baltic-Slavic contact zone ; Circum-Baltic Area ; Belarusian ; areal convergence ; mixed dialects * We owe our gratitude to an anonymous reviewer who scrupulously pointed out some short- comings of a previous version of this article. Even though we might not have followed his/her advices in every detail, we hope that the article has increased in accessibility for readers not familiar with facts of Slavic and Baltic. Of course, the usual disclaimers apply. B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 185 Introduction: basics about a particular kind of rural Belarusian mixed dialects Mixed Belarusian dialects are spoken all over the countryside along the over- lap region of (East) Slavic and Baltic dialects stretching from Northeast Poland in a half moon-like way up to Latgalia (i.e. Southeast Latvia). Th ese dialects are commonplace on the Belarusian side of the contemporary border between Belarus and Lithuania, but they are also still encountered in many spots on the Lithuanian side of this borderline (see Map 1 ). Th is region will henceforth be called the Baltic-Slavic contact zone (BSCZ). It belongs to the Southeastern part of the Circum Baltic Area (= CBA; cf. Dahl/Koptjevskaja-Tamm 2001 ). Th e geographical expansion of mixed Belarusian rural dialects more or less coincides with the region in which local Polish, more precisely the north- ern variety of polszczyzna kresowa (henceforth PolKres), has evolved (see section 1.2). It is shown on Map 1 . Hereafter we will refer to rural mixed varieties of Belarusian in the BSCZ simply as ‘rural Belarusian’ (or ‘vernacular’) if not indicated otherwise. Th e mixed Belarusian rural dialects are often referred to by their speakers as Mowa Prosta . Th is self-denomination cannot be regarded as a term (or lingu- onym) proper. It has arisen (and is normally used) in contrast to PolKres. It occurs most consistently in the southwestern part of the BSCZ where, until 1939, the density of representatives of the Polish gentry (Pol. szlachta ) was higher than in most of the remaining parts of the BSCZ. Polish was usually chosen as a means to underline cultural superiority 1 so that Belarusian found itself in sociolinguistic opposition to Polish 2 (see 1.2). Simultaneously and during all periods, Belarusian has probably been used ubiquitously in public life, and there is grounds to assume that only few speakers of PolKres had no or only a poor command of Belarusian, whereas speakers of Belarusian were often not fl uent in Polish (either PolKres or the standard variety), at least after World War II. Furthermore, during the interwar period, Polish was the language of instruction in schools even in the BSCZ with its predominantly 1 Th is tradition developped in the times of the Grand Duchy of Lithuania (15th-18th c.); cf. Wiemer ( 2003b : 109f.). 2 Mowa Prosta literally means ‘straightforward, easy, unsophisticated speech’. An alterna- tive expression – also in Polish – is język tutejszy literally meaning ‘our language’. Informants themselves often say that they speak pa prostu or po tutejšemu (with various phonetic variants between Polish and Belarusian). Th is seems to be a very common way of self-denomination serving also as a sign of intra-group solidarity in linguistic and/or ethnic minorities. In even other localities, inhabitants would rather call their speech “mixed” or “local language” (cf., e.g., Jankowiak 2009 : 70). 186 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 Map 1: Dispersion of Belarusian rural dialects and PolKres (from Grek- Pabisowa 1992 : 56) B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 187 East-Slavic speaking population, while Belarusian was being ousted from education. From the sociolinguistic viewpoint, speakers of contemporary rural Belarusian in the BSCZ certainly do not constitute a homogeneous group. We know neither how many of them are or have been bi- or multilingual (or multi-dialectal), nor how many are or have been illiterate. Diastratically, rural Belarusian is “fl at” and represents a decidedly “low” variety. From the diatopic viewpoint, it is very dispersed, but we do not even know whether (and to what extent) its subvarieties in diff erent parts of the BSCZ share the same features and/or amount of structural variation in detail, or in other words: to what extent their variation has been subject to infl uence from a partially diff erent set of contact varieties (see Table 1 ). In rural Belarusian within the BSCZ, we observe an extraordinary amount of idiolectal variation, especially in phonet- ics and morphology. As a consequence, it turns out to be particularly diffi cult to describe it in structural terms including all kinds of phonic and morphic variants, for which even the status as allophones or allomorphs, respectively, is often unclear. Eff orts to comprehensively capture some structural variation in the rural Belarusian of the BSCZ have so far been made by Wiemer ( 2003a ; 2006 ) and Erker ( 2009 ; forthcoming) as well as by Jankowiak ( 2009 ). Recently Smułkowa ( 2010 ) has dealt with dialect levelling and mixing on the back- ground of an advanced process toward an East Slavic (Belarusian + Russian) – Polish merger for which it is often diffi cult to establish a matrix code. Finally, we ought to mention Sudnik ( 1975 ) who, in her impressive work on phonetic variation and a phonological diasystem of multilingual speakers, took account of local Belarusian varieties. Th ere have also been a number of articles in which structural features of the dialects we are interested in here have been investi- gated de facto, albeit often under other names (cf. the surveys in Wiemer, forthcoming1; in prep.). Table 1 subsumes the partially divergent arrays of Slavic and Baltic contact varieties which one has to reckon with when doing research into rural Belarusian in the BSCZ. Th e columns from left to right are ordered iconically according to the geographical direction from southwest (near to the Polish border) to northeast. By ‘regional Russian’ we mean varieties of Russian that are more or less close to the Russian standard language but are themselves usually infl uenced by Belarusian traits to a considerable degree. To what extent Russian played a role in the Lithuanian part of the border region (and to what extent it continues doing so now) is not clear. Th e weak Belarusian standard does not exert any tangible infl uence on rural Belarusian speakers. Th e same obviously holds for other standards, i.e. Polish (West Slavic), Lithuanian and (in the northeastern 188 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 3 Cf. Hentschel ( 2008 ), Zaprudski ( 2008 ), Kittel et al. ( 2010 ). 4 Cf. the volume edited by Zemskaja/Šmelev (1984). 5 Th e much smaller group of Polesian dialects in the southwest, bordering with Ukrainian, is beyond our present concern. edge) Latvian or Latgalian. Whereas Belarusian provides the dialectal basis everywhere, Lithuanian has ceased to play a role beyond the contemporary state boundary with Belarus; it seems to have retreated at least at some spots on the Lithuanian side, too (cf. Wiemer 2006 ). Th e diatopic and diastratic characteristics of rural BSCZ-Belarusian mark- edly diff er from urban mixed Belarusian (or Belarusian-Russian) varieties denoted with the quasi-term Trasjanka , 3 although the results of the “mix” rep- resented in rural Belarusian become manifest in structural properties and variation reminding us partially of phenomena observed in Trasjanka or Russian Prostorečie . 4 It is anything but clear whether and to what extent dialects distinguished in Belarusian (or East Slavic) dialectology are still alive in the BSCZ. It is thus very diffi cult to say anything specifi c about the possible impact of former dia- lect continua on the formation of the mixed rural dialects, which are the topic of our contribution. On the basis of isogloss bundles, traditional Belaruthenist dialectology distinguishes a northern and a southern group, which more properly should be called north-eastern and south-western, respectively. A third, transitional group is situated between them (Avanesa ŭ et al. 1968 ). 5 From Map 1 we can easily see that the BSCZ, where mixed rural Belarusian is spoken, crosses these traditional Belarusian dialect zones almost vertically, although to a large extent it belongs to a special agglomeration of dialects traditionally called the ‘Northwestern dialect zone’ (Bel. pa ŭ nočna-zaxodnjaja dyjalektnaja zona ). Table 1 Varieties in contact with rural Belarusian in the BSCZ Hrodna – Lida Dieveniškės, Šalčininkai, Eišiškės Pabradė, Švenčionys, Adutiškis Pastavy – Brasla ŭ Latgalia regional Russian, PolKres regional Russian, PolKres, Lithuanian regional Russian, PolKres regional Russian, PolKres, Latgalian, Latvian (?) B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 189 6 Most of these features have been included in List 1 , but we will consider some others, too. Table 1 helps us understand why rural Belarusian plays a central role in the Baltic-Slavic border region: it occurs in virtually all areas regardless of what other languages are used. Regional varieties of these languages have – to diff er- ent degrees and depending on the specifi c region within the BSCZ – had an impact on the formation of Belarusian mixed dialects. It is, therefore, any- thing but astonishing that, in a sense, Belarusian dialects of the BSCZ have incorporated features and highlight tendencies that are common in most or even all contact varieties in this region. A great deal of its rise and seemingly chaotic variation can most probably be explained by the dialect mixing that was enhanced by migration (caused, e.g., by the abandonment of serfdom in Tsarist Russia 1861, by wars, and by inter-war Polish as well as post-war Soviet policy) and also by its low-prestige, uncodifi ed and only oral basis. It is exactly its “unembellished”, low-prestige status that makes rural Belarusian a typical representative of the BSCZ and an outstanding candidate for the study of contact-induced variation. By the same token, the features and the varia- tion we encounter in these dialects are nothing special in comparison to what we can observe in the contact varieties used in the same area. Th ese points will become central in the further course of this paper. In what follows, we will fi rst point out some methodological intrica- cies caused by the endeavor to evaluate the contact phenomena found in rural Belarusian from a broader perspective of the Eastern part of the CBA (section 1.1). Th en we will briefl y discuss features that give evidence for a tight interconnection of Belarusian rural dialects with PolKres (1.2) and spell out the reasons why this is so (1.3). In section 2, we will elaborate on the question of whether and how one can distinguish between the reasons and triggers for the structural phenomena (and their variation) encountered in rural Belarusian: whether they are to be sought after in internal development or belong to fea- tures of inner-Slavic dialect continua, or whether they are contact-induced and, if the latter is the case, whether we can distinguish inner-Slavic from Baltic infl uence. In particular, we want to discuss a couple of features in greater detail, which have had attention paid to them in the dialectological literature of the Slavic varieties spoken in the BSCZ. 6 Our guiding questions are: (a) Which features of rural Belarusian are atypical in terms of traditional Belarusian dialectology? Or otherwise: which features of rural Belarusian of the BSCZ testify to its mixed character violating boundaries between previously established dialect groups? 190 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 (b) Which features of rural Belarusian are manifestations of change, and which are conservative Belarusian features of the given localities? (c) To what extent can we distinguish contact-induced change from the out- comes of internally motivated evolution? It is plausible to assume that, in the BSCZ, both factors, more often than not, go hand in hand. We may thus ask more particularly: can we tell triggers and subsequent processes of propagation apart? (d) Does the degree of genealogical closeness matter and, if so, to what extent? More precisely: are there features more resistant to infl uence from Lithu- anian than others for which Lithuanian infl uence is very likely or even obvious? Section 3 summarizes our fi ndings and contains an attempt at answering whether the degree of genetic closeness between varieties in the Baltic-Slavic region is of any signifi cance for contact-induced results in the structure of rural Belarusian (and partially also other varieties). 1. Local signifi cance and areal linguistic background 1.1 Local variation as parts of areal clines Th e genealogical affi liation of the East Slavic (Russian, Belarusian) and West Slavic (Polish) varieties in the BSCZ is very close. Th e East Slavic varieties belong to a single dialect continuum, and some morphosyntactic features of Belarusian are, in turn, transitional between Russian in the (North)East and Polish in the (South)West; this is partially refl ected in dialect groupings accepted in traditional Belarusian dialectology (see above). As for Baltic, Lith- uanian and Latgalian show many features shared with local varieties of East Slavic and Polish although they are genealogically “farther” apart. Th is refl ects the traditional groupings into families within IE., among which, how- ever, Slavic and Baltic have always fi gured as particularly intimately related. Whether and to what extent this is due to common ancestry or caused by long-lasting contacts (after a split within IE. had occurred) has largely remained an unsolved question. Nonetheless, it is indisputable that, in loca- tions where speakers of Baltic and Slavic dialects have come into intimate and long-lasting contact, heavy traces of (mutual or one-sided) infl uence have become apparent. Th e basic problem consists of determining which conver- gent features have really been triggered and/or expanded by contact with varieties of the respective other group, and which features have become com- mon on the basis of the formation of dialect continua (transitional zones). B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 191 Wälchli’s comments (forthcoming: ch. 1) are applicable here: ‘Contact- induced change is (…) more likely to occur between closely related languages than between unrelated languages’. Among closely related languages, contact- induced change ‘is diffi cult to distinguish from inherited common innova- tions’. Th e problem is aggravated if we apply this question to features inside the Slavic family, or even only its Eastern branch, to which Russian and Belarusian belong. Th is basic problem has been detailed in recent work on areal typology and population linguistics, cf. again Wälchli (forthcoming: ch. 1): Even though neighboring languages with a history of language contact will always exhibit many partial structural similarities, language contact is only one of many factors leading to language change and so it is often impossible to prove diff usion in individual cases. (…) quirks are only a small proportion of all contact phenom- ena (only the tip of the iceberg). Contact-induced change is more likely to occur in cross-linguistically frequent features where it will always be claimed that the areal explanation is not necessary because such features develop easily without external causes in very diff erent areas. In other words, the less spectacular observed features (or changes) are, the more diffi cult it becomes to diagnose the reason for areal convergence. Further- more, Wälchli points out that conclusions drawn from the emergence of con- vergent features in contact languages seem to depend on the formulation (or hidden assumption) of what we may call the ‘null hypothesis’. Whereas areal linguists explicitly or tacitly assume null hypothesis (A), typologists tradition- ally have relied on null hypothesis (B): (A) Lack of contact phenomena even between neighboring languages. (areal linguistics) (B) Contact phenomena are omnipresent even across macro-areas. (typology) In either case the motif is easy to disclose: typologists are primarily interested in the regularities hidden behind structural variation and are eager to deter- mine the limits of variation which natural languages are capable of; to this end they try to achieve insights into the widest possible range of phenomena and have been keen to avoid any “disturbance” from external factors. Th is reason- ing yields the ground for sampling procedures in the Greenbergian tradition. By contrast, linguists interested in the geographical distribution of structural patterns are eager to discover areal biases and, moreover, to explain areal clines (continua or clusters across languages and language families) by contact. In either case, the respective null hypothesis should be easy to disprove but also serve as the basis for a kind of alpha-error (to maintain the analogy with 192 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 7 On this caveat concerning approaches from population linguistics cf. Wiemer ( 2003c : 216f.; 2004: 498-501). 8 Cf. Wiemer/Hansen (forthcoming: ch. 3) for a justifi cation of this discrimination with respect to contact-induced grammaticalization. statistics), i.e. to track down (and then exclude) cases that are able to run counter to the desired overall picture. Moreover, the reasoning typical for areal linguistics resembles dialect geo- graphy in principle whose representatives, following the founders of wave theories, have assumed that the spread of features (and isogloss bundles setting their approximate boundaries) comes about as a result of stepwise accommo- dation between speaker groups in restricted local areas: the accumulation of local areas chained one after the other sums up to more global eff ects. However, the analogy with assumptions in dialect geography is not perfect since “the question arises as to what extent micro- and macro-areal clines are really comparable” (Wälchli, forthcoming: ch. 1). For this reason, one has to be very careful in transferring methods applied by (and results achieved in) areal lin- guistics to more locally restricted areas such as the BSCZ. 7 With this caveat in mind, we will nonetheless take advantage of recent insights of areal linguistics insofar as our main concern will be to show that the assessment of what has been going on in the Baltic-Slavic region in general and with rural Belarusian in particular becomes more complicated and at the same time easier to judge if we account for its features from both a micro-areal and a more global perspective on (at least) the Eastern part of the CBA. Th is approach should be complemented by a further consideration. Namely, we have to distinguish between three phases (or steps) in the consolidation of structural features: (a) initiation of innovations, (b) propagation, (c) entrench- ment. If we do not (try to) do this, a discriminate treatment of system-internal factors vs. external (i.e. contact-induced) triggers proves impossible not only for the moment of innovation (initiation), but also for the spread of innovations. 8 1.2 Comparison with ‘borderland Polish’ (‘północna polszczyzna kresowa’, PolKres) As was remarked already in Wiemer ( 2003a : 232-235), it is almost impossible to distinguish between rural Belarusian and PolKres on the basis of structural features alone, simply because most features mentioned as typical for PolKres are of a clearly East Slavic provenance; often they are shared by practically all Slavic and Baltic varieties of the BSCZ or even farther beyond. Th is causes B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 193 9 Information on Lithuanian dialects by courtesy of V. Kardelis (Vilnius). 10 Varies with “Polish manner” (cp. [xodźit ɕ , dźeń] etc.) and “Russian manner” (cp. [xad’it’, d’en’, bat’ka]) to the extent that three degrees of original /d’, t’/ occur mixed up: Russ. [d’, t’], Bel. [dz", c"], Pol. [dź, t ɕ ]. diffi culties in tracing the roots of the feature in question. Th e fact that Polish scholars have mentioned such features as “peculiarities” of PolKres – although as such these features do not discern PolKres from Belarusian of the same region – partially arises from the specifi c focus of attention paid to PolKres, namely: these features have been highlighted in comparison to standard Polish. However, standard Polish (as well as standard Russian) is in many respects atypical; it is the standard varieties exterior to the BSCZ that deviate from the varieties within the latter, not the other way around! Th erefore, using these standards as a sort of gauge against which local varieties of the BSCZ are described distorts the picture (cf. Wiemer 2004 ). For instance, Grek-Pabisowa/Maryniakowa (1999: 19-41), in their intro- duction to a volume of representative texts of PolKres in this borderland region, pointed out features of PolKres, among which at least the 17 shown in List 1 coincide with what we encounter in local Belarusian. Virtually all of them are characteristic of other contact varieties in this region, too. List 1 : Features shared by PolKres and rural Belarusian (not exhaustive) 9 I. Segmental and suprasegmental phonetics, phonotactics (i) Word stress on last syllable; e.g., PolKres pa ˡ t ͜ ʃ ai ‘look!’. All varieties of the BSCZ, except PolKres (penultimate stress) and Latgalian (ini- tial stress), have movable word stress, which also allows fi nal syllables to be stressed. (ii) No penultimate stress in PPs if preposition and noun are monosyllabic; e.g., PolKres do i ̯ éi ̯ ‘to her’, n’e mám ‘I don’t have’. Th is phenomenon is just a consequence of the violation of clitic rules (together with penulti- mate stress) obtaining in standard Polish. In no variety of the BSCZ do clitic rules work. (iii) So-called C’ekann’e/Dz’ekann’e : the etymological dental plosives [d, t], when palatalized, become aff ricates [dz", c"]; e.g. PolKres xadz”ic” ‘to walk’, dz”en’ ‘day’, bac”ka ‘daddy’. 10 Th is is a very salient Belarusian fea- ture, but it also occurs in Lithuanian insular dialects and in Southern Aukštaitija (Dzūkian dialects), although probably for a diff erent reason. 194 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 (iv) Palatalized consonants can also occur before back and middle vowels; e.g., PolKres pouc”akal’i ‘(they) ran away’, zav’azu ‘I’ll tie up’, z”aml’u ‘land. acc’ , l’on ‘linen’. Th is feature is common to northern Belarusian and Russian as well; it is also attested for Aukštaitian dialects, although the reason is diff erent (disappearance of * j ). (v) Obviously free variation of [n, s, z, t s] vs. [n’, s’/ś, z’/ź, t s’/t ś]; analogous to feature (iii). (vi) Mostly voiced /v/ in the groups /kv, sv, t sv, xv, ʃ v, t ʃ v/; e.g., PolKres [kv’at, śv’at] ‘fl ower, world’, instead of standard Polish [kf ’at, śf ’at]. Lack of assimilation of /v/ after these voiceless consonants is common to all varieties of the BSCZ; the odd variety out is standard Polish. (vii) Obviously free variation of [x] and [ γ ]. Th e same can be observed not only in rural Belarusian, but also in some Lithuanian dialects of the BSCZ. (viii) Pro- and epenthetic (semi-)consonants; e.g. PolKres vušy ‘ears’, jon / jana ‘he / she’. Th is feature can be encountered in many non-standard varieties practically all over Slavic as well as in Lithuanian dialects. (ix) Akan’e. Th is term refers to the reduction of etymological /o/ in unstressed syllables. It exists in diff erent variants but is characteristic for the entire middle part of the East Slavic dialect continuum; this includes middle Russian dialects (and standard Russian, whose basis are the middle Russian dialects of the Moscow – Suzdal’ – Vladimir region) and the northern half of Belarus relevant for the BSCZ. In PolKres, akan’e occurs with no observable consistency, but examples like adná adnú ‘one the other’ (etymologically /o/), vajná tája pačalá s'e ‘that war began’ (cp. standard Polish: wojna ‘war’, początek ‘beginning’) are well known. A phenomenon similar to akan'e is also encountered in the Latgalian dialect (Seržant 2010). Although its rise in Latgalian has to be explained on a diff erent basis (cf. Seržant 2005 ; 2010), it is striking that, despite probably independent (internal) motivations for the appearance of this phenomenon in both East Slavic and Latgalian, similar lines of development converged at some time. As a result, at later stages, bi-/ multilingual speakers at a synchronic level are likely to identify akan’e in these varieties. (x) Apparently free variation of [v – ŭ – u]; e.g. PolKres [ f s’o p’eraval’i ŭ ] ‘(he) turned all over’ vs. [ ŭ s’o, ŭ s’ayo ], [ ŭ že ] vs. [ uže ] ‘already’, [ dz”ec’i u plat ʃ ] (vs. [ v pla t ʃ ]) ‘the children began to cry’. Th is, again, is also typical for rural Belarusian and Lithuanian dialects of the BSCZ. (xi) Palatalized pronunciation of /x/, i.e. [x’i, x’e] (instead of [x ɨ , xe]); e.g. PolKres [mux’i, mux’e] ‘fl y. gen/dat ’. Th is pronunciation is the norm in northern Belarusian and Russian and can as well, although much more rarely, be encountered in Lithuanian. B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 195 11 A list of abbreviations is given at the end. If not indicated otherwise, data are taken from tape-recordings made during fi eld work in 2000-2004. II. Morphology (xii) Ending {u} instead of {ov’i} to mark dat.sg of the most productive mas- culine declension. Th is conforms to {u} being the exclusive ending in Russian. It seems to preponderate in rural Belarusian, too. (xiii) Ending {ai ̯ } instead of {o ̰ } for ins.sg of the most productive feminine declension. Th is corresponds to the East Slavic ending {oi } (under con- ditions of akan’e, see feature (ix)). (xiv) Unifi ed ending {ov} (phonetically [of, af ] or [ou ̯ , au ̯ ]) for gen.pl in all declensions (cf. Erker, 2009 ; forthcoming). Th e unifi cation of one end- ing for the gen.pl regardless of gender and declensional class belongs among the most persistent features throughout the BSCZ (compare, e.g. Lith. {ų}). A tendency toward unifying {ov} as the single gen.pl - marker can furthermore be observed in many other varieties of north Slavic beyond the BSCZ (see 2.1). III. Syntax (xv) Use of the genitive to mark indefi nite quantity for subjects and objects (see ex. 1a-b). Th is is another of the most pervasive features of all varie- ties in the BSCZ (see 3.2). (xvi) Predicative use of indeclinable anteriority participles (ending in {vši}) leading to resultatives (perfects); see ex. (2). Th is is likewise a ubiquitous feature of all varieties of the BSCZ. (1a) xl’ éba b ýla / kart ófl ’e b ýlo // (Str.) 11 bread. gen be. pst.3sg.n potato. nom.pl (?) be. pst.3sg.n ‘there was bread / there were potatoes’ (1b) i an’ í tak’ íx sm’ešn ýx anegd ót’ikaf rask á zyval’i // (Ad.) and they. nom such funny. gen.pl joke. gen.pl tell: ipfv.pst.3pl ‘and they told (us) very funny stories ’ (2) dz’ádz’ka byu ̯ pryi ̯ éxau ̯ šy , nu i pai ̯ éxal’i (Br.) uncle. nom be. pst.3sg.m arrive. ptcp ptc ptc go: pfv.pst.3pl ‘(our) uncle had come [lit. ‘was arrived’], so that we went off ’; ma i ̯ é rúk ’ i paščapáu ̯ šys’a (Br.) have.prs.3sg hand. acc.pl become_rugged. ptcp:refl ‘(s/he) has (his/her) hands (which have become) rugged’ ; 196 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 navučý u ̯ šy nas bylá, ni klása ni kónčy u ̯ šy (Lid.) teach: pfv.ptcp we. acc be. pst .3 sg.f neg class.gen neg end: pfv.ptcp ‘(she) had taught us, (although she) hadn’t (lit. didn’t) fi nish a class’; nas byl ’ í spal ’ í u ̯ šy (Br.) we. acc be. pst .3 pl burn_down. ptcp ‘(they) had burnt us’. IV. Discourse (propositional modifi cation) (xvii) Use of epistemic particle mus’i (< 3. sg.prs of mus’ec’ ‘must’); see ex. (3). Another commonplace feature in Slavic and Baltic varieties of the BSCZ. See some examples from rural Belarusian: (3) vot / γ etyx γ rošau ̯ u m’en’a było / ptc this . gen.pl money . gen.pl at me . gen be . pst .3 g.n mus”i i ̯ ak’ix dv’enaccac” tys”ač must_be indf. gen.pl twelve thousand . gen.pl ‘well / I did have some money / probably some 12 thousand (Latgalia; cited from Jankowiak 2009 : 155) Th ese 17 features are by no means an exhaustive list of micro-areal coinci- dences, but they provide a good idea of what comes to a dialectologist‘s eye (or rather: ear) if one looks at the data from outside the BSCZ and restricts one- self to merely enumerating what has been encountered. If, however, we restricted ourselves in this way, we would arrive at the strange conclusion that Belarusian and PolKres do not diff er in the BSCZ, or only scarcely. Th is is in fl agrant contrast to what speakers of these varieties as well as fi eld workers know: there is a distinct, well-perceived diff erence between both varieties not only in sociolinguistic, but also in structural terms. Everybody who has been in the fi eld and has listened to recordings of speech can usually tell whether a particular stretch of discourse is in Belarusian or PolKres. While we are not going to dwell on sociolinguistics in more detail, the lesson to be drawn from this weird superfi cial conclusion is that, if we want to render a proper linguis- tic description of these two varieties, we must integrate frequency patterns, more precisely: proportions of occurrence of variants (“allo-realizations”) both in the paradigms (types) and on discourse level (tokens; cf. Wiemer 2003a : 236; forthcoming1). And we should probably distinguish salient features from less salient ones, on both the acoustic and the morphosyntactic level. 1.3. Reasons that have led to coincident features of PolKres and rural Belarusian After all, what is the reason for this overwhelming convergence of struc- tural features in Belarusian and PolKres? Th ough the historical and social B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 197 12 According to Morita (2006: 127), most specialists on PolKres even assume that the Polish- speaking inhabitants of the villages use Belarusian ( Mowa Prosta ) more often than PolKres in their everyday life. 13 Th is concerns at least its rural variety in the border region. Th e urban variety of Vilnius, which is closer to standard Polish, is beyond the scope of this paper. 14 Turska ( 1995 ) considered that, after the abolition of serfdom, many Lithuanians began to switch directly to Polish without the mediation of Belarusian. However, she herself also noted a faster expansion of Polish particularly among those Lithuanians who previously had already changed to Belarusian (1995: 52). No less important is the consideration that the Lithuanian peasants were most probably able to communicate in Belarusian, so that the latter has to be considered as an adstrate anyway. For a more detailed discussion cf. Wiemer ( 2003b : 124-127). background is very complicated, part of the explanation presents itself quite readily if we realize that most (all?) rural speakers of PolKres also use Belarusian, at least in the localities where we are adducing data. 12 Moreover, Belarusian and PolKres 13 evolved in the same region most probably because the Belarusian and Lithuanian speaking population was considerably infl u- enced by Polish or even switched to Polish during the course of a few genera- tions. According to Turska ( 1995 [1982]), this process started after serfdom had been abolished in 1861 and became rapid at the latest during the interwar period when, in 1920, the whole region found itself under Polish rule and schooling was conducted exclusively in Polish. In this period, many Belarusian or Lithuanian speakers did not accomplish a transition toward Polish but con- tinued using Belarusian vernaculars 14 as Belarusian had already most probably been serving the role of a mediator variety for several centuries, i.e. from the times of the Grand Duchy of Lithuania. For the same reason, Belarusian ver- naculars must have constituted an adstrate or substrate for all speakers at least in the countryside of the Vilnius region where Lithuanian and Belarusian dia- lect zones overlap in intricate ways. Th e contemporary Belarusian – Lithuanian border runs right through the middle of the half moon-like shaped stretch of land referred to in the intro- duction in our comment on Map 1 . Th e spots where PolKres occurs are also well-established spots of Belarusian usage. Th e single diff erence is that, in comparison to PolKres, Belarusian certainly occupies a more compact terri- tory with fewer “holes”. Given the assumptions about the genesis of rural PolKres in the BSCZ (see above), one can say that, to a large extent, rural mixed varieties of Belarusian developed at least partially as an off spring of the at times quite rapid switch from regional East Slavic and Baltic vernaculars toward Polish. 198 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 15 For East Slavic cf., among others, Kuznecov ( 1954 : 75; 1973: 250-260), Lipovskaja ( 1983 ). 16 Cf. Urbańczyk (1984 7 : 45f.) for Polish, Skudrzykowa et al. ( 2001 : 192) for Silesian Polish, Breza/Treder (1981: 112) for Kashubian, SRA (1975/xi: 218-223) for Sorbian dialects. 2. Th e Baltic-Slavic contact zone as part of larger areas Th e 17 features included in List 1 diff er in terms of their relevance for areal linguistics. Some of the features appear to be really restricted more or less to the BSCZ or even to its Slavic varieties alone. Other features can reasonably be compared to a distribution within Slavic as a whole as they belong to dia- lect continua or are “scattered” without a connection to the BSCZ (see 2.2.1- 2.2.2). Further features are also well known in non-Slavic languages (varieties) of neighboring regions, above all Finnic ones (see 2.3). When speaking about convergence in that relatively restricted area which forms part of larger areas (as in a model with concentric circles), we have to distinguish between these layers of diff erent range. Even though we still have quite insuffi cient know- ledge about Belarusian in the BSCZ, we see that it is in these mixed varieties that these layers become manifest with their probably densest concentration. From among the 17 features given in List 1 , let us fi rst choose features related to nominal infl ection in order to demonstrate the intricacy of the ques- tion formulated in 1.3 above, in particular for question (c). 2.1 On telling Slavic-internal developments and Lithuanian infl uence apart Let us examine feature (xiv), the unifi cation of the gen.pl -ending. Here we are dealing with the reduction of (allo)morphic variation for which an ana- logical process can be observed in Lithuanian (and Latvian); nonetheless no convincing answer can be given to the question of whether and in which way Baltic varieties have participated in this process. Th e ending {ų} for all declen- sion classes is the norm in Lithuanian, but there are a few circumstances that diminish the probability that Lithuanian impact has been a decisive factor. Moreover, it turns out to be very diffi cult to even prove that Lithuanian had a strengthening eff ect in addition to other, major factors. Th ese circumstances are as follows: (xiv.a) Th e simplifi cation of the gen.pl -ending can be encoun- tered at many places in North Slavic; it has been noticed in dialectological textbooks and special investigations not only in the BSCZ or its immediate neighborhood, 15 but also in West Slavic down through its most extreme vari- eties in the West. 16 (xiv.b) Th e generalization of {ov} as marker of gen.pl is a B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 199 17 Cf. Zemskaja/Kitajgorodskaja (1984: 75). Zemskaja ( 2000 : 779) noted that this general- ization is also known from fi rst-language acquisition and the speech of Russian emigrants in Western Europe and Northern America. Th ese observations speak in favor of a common under- lying cognitive propensity responsible irrespective of (or in addition to) language contact. commonplace feature in studies on so-called (gorodskoe) prostorečie (‘substan- dard (urban) speech’) in Russian. 17 Th is shows that this phenomenon also has a widely known diastratic dimension. Both (xiv.a) and (xiv.b) make it unlikely for Baltic infl uence to have been a trigger for the unifi cation process; it tends to occur in central dialects as well as on the edges of Slavic (at least in the west) where no phonetic or functional parallels with nominal paradigms exist in non-Slavic contact dialects. Lastly, for (xiv.c), it is tempting to explain the spread of {ov} (at least partially) in the framework of Natural Morphology. Consider that, in North Slavic, {ov} as an ending is used in no other slot of case-number paradigms in Northern Slavic; it is salient and therefore an excel- lent candidate for being generalized in this function. Only here is a point where Lithuanian infl uence could have occurred: Lith. {ų} as marker of gen. pl is almost equally isolated in the case-number paradigms of nouns and adjectives; only the u -declension (exclusively masculine) also has it in the acc. sg (e.g., žmog-us ‘man, human being’ → žmog-ų . acc.sg , al-us ‘beer’ → al-ų . acc.sg , gudr-us ‘clever. m ’ → gudr-ų . acc.sg.m ). In a diasystem of multilin- gual dialect speakers, this pattern might have been strengthened since it favors paradigmatic economy. But can we discern whether the Lithuanian pattern served as a trigger or only strengthened the propagation of a tendency that was obviously already inherent to the Slavic varieties themselves? Note, moreover, that even if Natural Morphology could be used as an explanatory tool in this particular case, it cannot be generalized for other cases. Consider feature (xii), which will be further treated immediately below. It runs counter to what Natural Morphology would predict: {ov’i} as marker of dat.sg (with masculine nouns) has been ousted by {u} in most varieties of East Slavic (or become the marked choice), although it is no less salient and exclusive for dat.sg .m than is {ov} for gen.pl . Obviously, considerations based on type and token frequency have to be taken into account, too. In addition, the spread of the ending {u} at the expense of {ov’i} seems to have started much earlier than the unifi cation of {ov} and in several Slavic lan- guages has more or less come to an end. Th ere are other “peculiarities” of rural Belarusian beside those included in List 1 , which are well attested either elsewhere in Slavic (2.2) or in a broader area encompassing not only Baltic, but also Finnic languages (2.3). Let us browse through some of them. 200 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 18 An alternative explanation might be that the ending of the instrumental of the same inher- ited paradigm has been transferred to the dative. However, to our knowledge, this kind of case syncretism is totally unattested in East Slavic leading us to readily dismiss this alternative. 2.2 Further features attested elsewhere in Slavic We can further subdivide this subsection into features that are “scattered around” in Slavic, at least in North Slavic, or, more narrowly, in East Slavic (2.2.1), and features for which dialectologists of East Slavic have established continua and isoglosses for quite a long time (2.2.2). 2.2.1 Features attested in various Slavic regions We may assume that morphological features like (xii-xiii) are clearly restricted to inner-Slavic variation. What is at stake here is a choice among allomorphs that belong to a specifi c slot in the case (+ number) paradigm of Slavic nouns of diff erent genders. Since what is relevant for the variation are the forms of the allomorphs (not their abstract representations), it does not make much sense to compare them to anything beyond the particular Slavic varieties involved in our borderline area. Baltic (Lithuanian) could have only infl u- enced the functional range of particular cases and the inventory of grams. On the contrary, there is hardly any evidence that Lithuanian had an impact on the phonological shape of declensional markers. Th e only case known to us is the allomorph {ai} as marker of dat.sg of the a -declension (mostly feminine nouns). Th is can be inferred from examples like (4a-b), in which {aj} and {oj} occur as allomorphs of the dat.sg of this declension class (instead of expected {e}); {aj} conforms to Lith. {ai} phonetically, used in a stressless syllable, while {oj} has to be interpreted as another realization under stress. Accidentally, this makes s’astroj a grammatical homonym of the instrumental form (which, how- ever, would not render any sensible interpretation), but actually (4a) could be understood as demonstrating a further step after the assumed borrowing from Lithuanian happened. Namely: {ai} must have become entrenched in the par- adigm to a degree that it cannot only replace the inherited ending {e}, but also even show phonological alternation according to the usual rules of Akan’e (for which see comment to feature (ix) in List 1 ): 18 (4a) a na što ž nam byló spláčyvac’ s’astr ój (Br.) and on what ptc we. dat be. past .3. n pay_off : ipfv.inf sister. dat.sg ‘and why did we have to pay off [the debt] to my sister?’ (4b) mal’čyk byu ̯ n’ev’al’íki, boy. nom.sg be. past .3. m small. nom.sg.m B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 201 19 According to Anan’eva ( 2008 : 77) they have also been attested, albeit rarely, in PolKres. 20 In Russian and Polish themselves {iva} and { ɨ va} are allomorphs. 21 A Russian and/or Polish infl uence seems to also be obvious in Lithuanian dialects of the BSCZ. In many of these dialects, especially on the territory of Belarus‘, proliferous use of mátk aj za rukú učap ’íu ̯ šys’ a išou ̯ (Lid.) mother. dat.sg at hand. acc.sg seize: pfv.ptcp:refl and go: ipfv.past .3 sg.m ‘the boy was rather small, he seized his mother’s hand and went (with her)’ (lit. ‘seized for (his) mother at (her) hand’; the dative codes an external possessor) Such instances only occur exceptionally in our records (cf. Erker, forthcom- ing: 2.3). 19 However, according to Grinaveckene/Mackevič (1979: 52-56), such forms have been attested in Belarusian dialects of the ‘Northwestern dia- lect zone’ (see section 1) where Lithuanian infl uence until the most recent past is plausible. Regardless of such rare instances, as such, features like (xii-xiii) are quite widespread within Slavic, also outside the BSCZ. Th eir variation arose from the interference with case-marking patterns whose allomorphs derive from diff erent declensional classes; these patterns have undergone redistribution for many centuries (probably since the late Common Slavic period, i.e. from the 6 th -7 th c. AD). Even in closely related varieties (like the East Slavic ones), these patterns could have “merged” in the course of usual dialect mixing, and no external, non-Slavic infl uence is required to explain this unstable variation in rural Belarusian. A similar remark concerns the realization of animacy patterns (see 2.2.2) as well as the allomorphs of the verbal suffi x {va} vs. {iva / ɨ va}. 20 Th is suffi x is a productive device for the derivation of imperfective stems. Th e allomorph {va} is characteristic of Belarusian and absent as a productive device in both Russian and Polish. According to Mackevič ( 1959 : 90), in traditional Belarusian dia- lects, the forms {va} and {iva} are distributed by a simple morphonologi- cal rule: the latter occurs only after consonant clusters ending in a sonorant (e.g. vykarm l ̱ -iva-c’ ‘to rear’, peradraž n ̱ -iva-c’ ‘ to tease (by imitating)’), in all other cases {va} applies (e.g. raskaz-va-c’ ‘ to tell (a story)’). However, in our Belarusian data {va} occurs only sporadically with {iva, ɨ va} being used instead; e.g. raskaz-yva-la ‘she told’ (Br.), raz γ avar-yva-jim ‘we talk (with each other)’ (Br.), abmán-yva-je deceives’ (Lid.), raskuláč-yva-li ‘(they) dispossessed the kulaks’ (Lid.). Th is can be explained by the infl uence of Russian or Polish where {iva} is very productive; on the other hand, this inner-Slavic infl uence certainly builds on the availability of {iva} as an originally morphonologically conditioned allomorph (beside {va}). 21 202 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 the suffi x {(d)inė} has been attested. Note that this suffi x is not borrowed but demonstrates a productive calque (or selective copy) of a derivational technique (with an unclear distribution of functions; for a critical overview cf. Kardelis/Wiemer 2003: 59-66). Furthermore, it is an open question as to whether this calquing process of a Slavic pattern was not sustained predominantly by Belarusian dialects, which previously might themselves have adapted to the Russian and Polish allomorph. Unfortunately, at the moment, such considerations have to remain specula- tions since no investigations of such issues have been produced. 22 Since the retreat of neuter nouns has been observed predominantly in dialects with akan’e, the “phonetic” explanation was often considered as decisive (cf. Kuznecov 1954 : 76f. and others). However, this explanation cannot apply in South Russian dialects (with Okan’e, e.g. in the group around Kursk-Orlov) and in other dialects, in which nouns ending in stressed [o] control agreement with the masculine or (more frequently) the feminine (e.g., molokò vytek : pf.past.sg.m ‘milk has fl own out’; kakaja . nom.sg.f molokò ‘which milk’, bol’šaja . nom.sg.f selò ‘large village’); cf. Avanesov/Orlova (1965 2 : 100), Kuznecov ( 1973 : 253) and others. 23 Th is loss also becomes obvious with other categories, e.g. with resultative participles (cf. Wiemer/Giger 2005: 29-33). Besides the gen.pl -ending, loss of neuter gender is another case attested in diff erent regions of East Slavic. In rural Belarusian (as well as in PolKres) this process has been noticed to variable degrees. It is plausible to consider Lithuanian (which has a two-gender system of the French-Italian type) as a supporting factor, but it is debatable whether it has been a proper trigger of this process (cf. Wiemer 2004 : 512). Th e point is that we fi nd too many spots in other regions quite far away from the Baltic-Slavic zone in which the same or similar processes have been observed. Th erefore, reduction or elimination of the neuter need not be explained by contact with Baltic. 22 On the other hand, a phonetic explanation which rests on akan’e (see comment to feature (ix) in List 1 ) cannot give an account of the whole story even for Belarusian, because neuter nouns take on the ending {i / ɨ } in the plural, not {a}. Th is looks like unifi cation of noun endings in the plural. In our data from Brasla ŭ and Lida, the ending {i / ɨ } of historically neuter nouns preponderates as well; e.g. sv’at-y ‘feasts’, vojsk’-i ‘troops’, saapščenn’-i ‘messages’. Th e original ending {a} occurs more rarely, e.g. sv’at-a ‘feasts’, čysl-a ‘numbers’, and adjectives in the singular can agree with these nouns in the masculine or feminine gender. Th erefore, it is plausible to assume that, at some stage of this process, analogy played a crucial role. Reduction or loss of neuter gender is also amply attested for in North Russian dialects (characterized by okan’e); cf., for instance, Čagiševa ( 1968 ). Yet another reason seems even more plausible for these dia- lects, namely contact with Finnic varieties. Finnic languages do not know gender at all; as a substrate or adstrate they have caused loss of agreement marking in Slavic dialects around Pskov and north to this region. 23 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 203 2.2.2 Features belonging to East Slavic dialect continua We have to bear in mind that, as a whole, Belarusian can, in structural and geographical respects, be regarded as an intermediary between Russian and Ukrainian (cf., among others, Kuraszkiewicz 1963 2 : 79-86), and also between Russian and Polish. As pointed out in section 1, the Belarusian dialects we are investigating are spoken on a stretch of land that, as it were, crosscuts the traditional division of Belarusian dialect groups. In line with this division, the patterns by which animacy is marked with case+number endings constitute a continuum on an East-West axis. Northeastern Belarusian is closer to (stan- dard) Russian whereas western Belarusian dialects “show patterns which are either totally consistent with standard Polish or are intermediate between Russian and the standard Polish pattern” (Wiemer 2004 : 510; consult for the areal and typological background). On the East-West axis the patterns shown by standard Russian and by standard Polish demonstrate a sort of extreme. Th ese can be exemplifi ed as follows: List 2 : Case marking (plural) for referential distinctions 1) In the nominative plural, standard Russian does not exhibit any animacy- related distinctions, i.e. animate and inanimate nouns are infl ected the same way; e.g. stol-y ‘tables’, student-y ‘students’, studentk-i ‘female stu- dents’, vnuk-i ‘grandchildren’, kot-y ‘tom-cats’. Standard Polish, on the contrary, distinguishes virile (= male adult) from the rest of the nouns; e.g. stoł-y ‘tables’, studentk-i ‘female students’, wnuk-i ‘grandchildren’, kot-y ‘cats’ vs. studenc-i ‘students’, pan-owie ‘gentlemen’. 2) In the accusative plural, standard Russian treats all animate nouns alike by using an ending which is identical with the ending of the gen.pl ( acc=gen ) , e.g. student-ov ‘students’, student(o)k-Ø ‘female students’, vnuk-ov ‘grand- children’, kot-ov ‘tom-cats’. Inanimate nouns have endings homonymous with the nom.pl (e.g. stol-y ‘tables’). Standard Polish, again, treats only virile nouns diff erently; the acc.pl is only identifi ed with the gen.pl with such nouns, all other nouns show acc=nom -homonymy. Compare, e.g., virile student-ów ‘students’, pan-ów ‘gentlemen’ vs. all other like stoł-y ‘tables’, studentk-i ‘female students’, wnuk-i ‘grandchildren’, kot-y ‘ cats’. Intermediate patterns in some Belarusian dialects demonstrate, for instance, acc.pl = gen.pl in nouns denoting women but not animals, and they lack a declensional distinction in nom.pl (cf. DABM 1963: 753-756 and the maps belonging to these comments). Th e acc.pl = nom.pl pattern for animals can be seen in the following instances: 204 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 24 u corresponds to v (as in Polish or Russian), its appearance is conditioned by phonotactic rules (see feature (x) in List 1). (5) us’é kúr-y ad nas pazab’irál’i (Lid.) all. acc hen. acc.pl from us.gen take_away: pfv .3 pl ‘they took all hens from us’; tréba byló karóv-y pás’ic’ (Lid.) be_necessary be. pst .3 sg.n cow. acc.pl shephard: ipfv.inf ‘it was necessary to shepherd cows’; kón’-i m’él’i (Br.) horse. acc.pl have. pst .3 pl ‘(they) had horses’. Against this background, it has, so far, been impossible to fi gure out whether speakers of rural Belarusian follow any clear pattern. Furthermore, in the records of the two locations, which we have systemati- cally studied, we can sporadically encounter inanimate nouns in the singular marked acc = gen : (6) máju star-ó γ a páspart-a (Lid.) have. prs .1 sg old. acc.msg = gen.sg.m passport. acc.msg=gen.sg.m ‘I have an old passport’; xto a γ urc-á máje, somebody. nom cucumber. acc.msg=gen.sg.m have .prs .3 sg xto pam’idór-a (Lid.) somebody.nom tomato. acc.msg=gen.sg.m ‘some have cucumbers, others tomatoes’ (lit. ‘who has (a) cucumber, who (a) tomato’); sad’is’ u 24 vóz’ik-a (Br.) sit: ipfv.imp.sg in(to) cart. acc.msg=gen.sg.m ‘sit into the cart’. One is tempted to explain such examples as calques (or even patent borrow- ings) from Polish, especially in Lida where PolKres is still more widespread than in the surroundings of Brasla ŭ . But we simply lack a fair enough number of similar attestations (including negative data) in order to be able to make any conclusions. Apart from this, the fi rst example in (6) is troublesome to explain as a straightforward polonism because the Pol. noun paszport does not allow for acc=gen . We thus would have to speculate that the speaker who produced this utterance must have further extended a pattern, which s/he might have taken over from some variety of Polish. B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 205 A further remark is in order here. We are unaware of any piece of evidence that suggests that the marking of animacy would have started (or been immi- nent) to develop in any Lithuanian dialect. As for Latgalian, Nau (p.c.) has noticed a few instances of masculine object nouns with animate referents whose case marking coincided with the genitive; however, whether this testi- fi es to an initial stage of animacy marking still remains to be investigated. Th us, as it seems so far, Lithuanian (and maybe Baltic in general) has been practically “immune” against this categorial distinction, even if all Slavic vari- eties that have ever been in intimate contact with Baltic dialects have displayed one or other ubiquitous and salient pattern of animacy marking. Looking at the phenomenon from this angle, we can therefore say that animacy (and, thus, diff erential object marking) has always been a purely inner-Slavic matter in the BSCZ. 2.3 Features attested in an area reaching both beyond Slavic and the BSCZ Th e last three features listed in List 1 are representative of this. We repeat them here for the sake of convenience: List 1 Features (xv-xvii), repeated syntax (xv) genitive of indefi nite quantity (‘pseudo-partitive’), for subjects and objects (see ex. 1a-b) commonplace (xvi) indeclinable anteriority participles (ending in {v ʃ i}) in predicative use ( → resultative or perfect); see ex. (2) commonplace (xvii) discourse (propositional modifi cation) commonplace epistemic particle mus’i (< prs .3 sg of mus’ec’ ‘must’); see ex. (3) Each of these features will have to be commented on separately. (xv) Genitive of indefi nite quantity Th is usage of the genitive as marking the Undergoer of either transitive ( → object) or intransitive ( → subject) predicates has been coined ‘pseudo- partitive nominal constructions’ by Koptjevskaja-Tamm ( 2001 ). Th is usage is widespread in the entire Eastern part of the CBA, and it increases from south to north (cf. also Koptjevskaja-Tamm/Wälchli 2001: 650-660). We can thus 206 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 25 Its non-occurrence in (standard) Latvian is due to a general decrease of genitival use. To this extent, Latvian presents an anomaly to this areal cline (to be explained on other grounds). only expect this pattern to also occur in Belarusian spoken in the BSCZ. We actually fi nd it the usual choice in Lithuanian. 25 (xvi) Predicative anteriority participles Th is feature is one of the diachronically most persistent and stable ones not only in the BSCZ, but also in the areas neighboring toward the Northeast. Th e only structural diff erence between Slavic varieties and Baltic ones is that the latter use participles infl ected for case (nominative), number and gender, whereas in the Slavic varieties they are indeclinable. Th e areal pattern of the predicative use of anteriority participles was analyzed in considerable detail by Wiemer/Giger (2005). Against a background of the BSCZ and the region to the Northeast of it, the occurrence of this type of participles is nothing extraordinary. Th ere are three striking pieces of evidence. First, predicative participles in {vši} only occur in Belarusian spoken in the BSCZ (see ex. 2) while being absent in the southern dialects; this is in line with their absence in Ukrainian and in the southern variant of PolKres (which developed on Ukrainian soil). Second, they are attested from verbs of diff erent diathesis types (transitive and intransi- tive ones), but they can only be subject-oriented. Th is conforms to the pattern in Baltic (Wiemer/Giger 2005: 54f.). Together with this, and thirdly, there is no confusion with n/t -participles, which become more pervasive (in all diathe- sis types and for both subject- and object-oriented resultatives) in the area north(east) from Pskov. All these facts taken together yield a strong argument in favor of regarding the BSCZ as the south(west)ern “prolongation” of a con- tinuum which starts at Lake Ladoga and Lake Onega but with an inverse distribution of suffi xes ({vši} vs. {n/t}) over diathesis types (Wiemer/Giger 2005: 33). (xvii) Epistemic particle As recently pointed out by Nau (forthcoming), Slavic and Baltic varieties that have come into contact with each other since the time of the Grand Duchy of Lithuania show considerable convergence in their system of expressions for modal meanings. Below we reproduce one of her tables. As we can see, almost all varieties have lexicalized expressions of epistemic necessity and possibility. To these varieties we should, following Nau, add the Latgalian particle (or sentence adverb) muszeń , which denoted epistemic B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 207 Table 2 Lexicalizations of ‘can be’ and ‘must be’ as epistemic particles in the BSCZ (= N. Nau (forthc.): Table 2 ) derived from LITH PolKres BEL UKR RUS meaning range ‘can (be)’ gal, galbūt może moža, mo, moža być možet, možet byt’ ‘maybe’, modal particle ‘must (be)’ ( musėnt etc., only dialectal) musi, musić musi być musi, music’ musi byc’ musit’ buti ( dolžno byt’, weakly lexicalized) ‘certainly’, ‘probably’, ‘maybe’ ‘has to be’ turbūt mabyc’ mabut’ necessity 100 years ago but can still be encountered in contemporary speech (Nau, p.c.). Nau convincingly argued for the assumption that language variet- ies of the former Grand Duchy of Lithuania have suffi ciently infl uenced each other so that eventually a convergent subsystem of epistemic particles emerged. Th is subsystem is distinct from what we fi nd in surrounding varieties (includ- ing standard Polish, Russian, and Latvian). We may thus argue for contact as the decisive factor in the appearance of this subsystem. While Latgalian muszeń is most obviously a borrowing either directly from Polish (PolKres?) or via some northeastern Lithuanian dialects, we can only speculate about the manner in which the equivalent units found their way into the other contacting varieties and established themselves there. Cf. Nau (forthcoming, p. 30): Regional Polish musi ‘certainly’ may be a calque (a “replica lexicalization”?) from Belarusian, but also from Lithuanian – or Lithuanian turbūt may be a calque from Polish, or Belarusian, and so on. Whatever happened, happened mostly in oral uses of the languages in previous centuries, and we are left without documents. 2.4 Broader areal features not attested in rural Belarusian So far we have discussed some representative and salient morphosyntactic fea- tures found in rural Belarusian; we have also dwelt on a broader areal back- ground against which features of rural Belarusian in the BSCZ have to be judged. We have argued that, on the one hand, there is virtually nothing in rural Belarusian that cannot be expected given the properties of its immedi- ately contacting varieties. We have furthermore detailed facts known from dialect geography also embracing the area of NW Russian dialects (around 208 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 Pskov, Novgorod and farther to the north) with which the BSCZ shows many parallels. Th is demonstrates that it would be better if the BSCZ were not con- sidered from an areally isolated point of view. On the other hand, we have indicated that there are features of rural Belarusian for which language contact need not be made “responsible” as the primary factor, in particular for which Lithuanian (or Baltic in general) should not be identifi ed as the main source of infl uence. Now we want to corroborate our argument that rural Belarusian is a faith- ful refl ection of features converging in its Slavic and/or Baltic contact variet- ies by turning the perspective, as it were, upside down. Indeed, we want to show that rural Belarusian does not have properties which are encountered only in contiguous areas, in particular the Pskov – Novgorod area, but not in the BSCZ. For we think that negative evidence can also be telling provided that we clearly oppose the smaller region which rural Belarusian is part of against broader areas beginning just beyond the BSCZ proper. Th e following fi ve features are characteristic for at least certain East Slavic dialects northeast from the BSCZ: 1) nom.pl.m ending in stressed {a}: Th is feature originated in north Russian dialects; from there it entered into standard Russian and has since been spreading among noun stems. In rural Belarusian it does occur, but extremely rarely. In the records rehearsed sys- tematically we have found only one token (see 7a), and it occurred with two other “variants” (see 7b) in the speech of the same informant: (7a) paspart-á ‘passports’; (7b) páspart-y , pášpart-y (Lid.). Th e realization paspartá can most probably be explained as a “loan” from standard Russian. Th e following four features are altogether unattested in our material of rural Belarusian: 2) Case homonymy (“syncretism”) of dat.pl and ins.pl with a unifi ed ending {im}, {am}, typical of the Russian dialects of the Pskov region; e.g. svo-im glaz-am videl ‘(I) saw (it) with my eyes’ (cf. Bromlej/Bulatova 1972: 99; Bukrinskaja et al. 2008 : 123). 3) Nominatival object; e.g. nado korov-a . nom doit’ . inf ‘(It’s) necessary to milk the cow’ (cf. Kuz’mina/Nemčenko 1964 for Russian dialects and Ambrazas 2001 on its genesis and spread in the whole area, in particular in Baltic and Finnic). 4) Non-agreeing participial passive with acc -NP; e.g. from North Russian dialects: B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 209 (8) vsju kartošku s”edeno all . acc.sg.f potato . acc.sg.f eat : pfv.ptcp ( indecl ) ‘(People) have eaten up all the potatoes.’ (cited from Sobolev 1998 ) Th is construction happens to be extended by the PP u +gen denoting the agent (cf. Trubinskij 1984 : 139-149; Sobolev 1998 with further references). Th is construction is not only well known from northern Russian dialects and has also been attested in some small south Russian “pockets”, but, above all, it is a productive construction in standard Polish. However, neither does its occurrence in standard Polish and the north Russian dialects rest on any shared dialect continua in the past; the evolu- tion of this construction arose on independent grounds (cf. Wiemer, forth- coming2: ch. 2, 3.2; Wiemer/Hansen, forthcoming: ch. 2.4, for a critical assessment) — nor is this construction known to Slavic dialects in the BSCZ, i.e. in a sense, it is absent exactly in the middle between two areas where it is well attested (Polish and north Russian dialects). 5) Postposed emphatic (demonstrative) “particles”; compare two examples from North Russian dialects (cited after Kasatkina 2008 : 311f.; see for further discussion): (9) Ogurcy ran’še-to ne sažali. cucumber. acc.pl earlier- ptc neg plant: ipfv.past.3pl ‘Earlier- ptc nobody planted cucumbers (here).’ (10) Škole-to otopl’enie nado, school. dat.f-ptc heating. nom/acc needed. aux(indecl) nu ja i vzjalasja . ptc I. nom ptc take_initiative: pfv.past.3sg.f.refl ‘Th e school- ptc needed heating, so I took the initiative.’ In no variety of the BSCZ have any similar tendencies been observed. 3. Assessment of fi ndings Let us now recollect the fi ndings, which are disparate not only relating to the kind of structural phenomena, but also lending themselves to diff erent answers on account of the questions (a-d) formulated in 1.3. Certainly, for the mo ment, the data we possess are too selective and the picture obtained on their basis too fragmentary for any more far-reaching conclusions. Nonetheless, some assump- tions of a methodological nature can be formulated, which could then also lead to some cautious preliminary conclusions concerning the motifs behind convergent features of the BSCZ that become manifest in rural Belarusian. 210 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 First of all, as far as we can judge by now, the high degree of variation of structural properties in the Belarusian rural vernacular has been due to con- stant, largely unhampered innovations in everyday oral communication; mostly these innovations have been propagated only within very restricted local communities of speakers, and rarely do they become fi rmly entrenched in some kind of internalized (in no case codifi ed) norm. In dialectological terms, rural Belarusian has to be characterized as a mixed variety containing combinations of features that cannot be reduced to traditional dialect conti- nua but which are situated somewhere, i.e. not quite predictably, on the con- tinuum of isoglosses between the Russian and Polish speaking territories. Th is can be seen particularly well with animacy marking (see 2.2.2) or the redistri- bution of allomorphs marking the dat.sg of masculine nouns (see 2.1). Th e variation of rural Belarusian, even on an idiolectal level, is considerable, and this has largely defi ed its description as a variety with a clear and “tidy” system. Its Belarusian morphology has been infl uenced by local Russian and Polish varieties, in part quite heavily, as is demonstrated, for instance, by the fact that the verbal suffi x {va} used for the derivation of imperfective stems has largely been replaced by {iva / ɨ va}. We have to, however, acknowledge that this allo- morph was not altogether unknown to Belarusian dialects even prior to heavy Russian infl uence since it was a minor choice in complementary distribution with {iva / ɨ va} depending on the stem fi nal consonant (see 2.2.1). As concerns question (c), i.e. the relative impact of contact-induced change and internally motivated evolution, our survey should have demonstrated that, for many features, one cannot sharply distinguish between results of con- tact with other Slavic dialects of the region or with adjacent dialects of Lithuanian (or Latgalian, which has not further been considered here). Th e reason for this diffi culty consists in the fact that all these dialects have been in close contact for centuries. Th e diffi culty is to some degree amplifi ed by the close resemblance of substantial parts of Slavic and Baltic morphology and phonology forming part of their common “heritage”; but for many details, it is diffi cult to decide whether the changes they have undergone have not occurred due to contact and, thus, mutual infl uence only quite recently. Th is problem manifests itself, for instance, in the weakening or even loss of the neuter gender, less so in the unifi cation of the gen.pl -ending in nouns (see 2.2.1, 2.1). Th e latter shows many parallels in other diatopic and diastratic varieties of Slavic, among which most do not show any closer relationship to the BSCZ; it is thus easier to explain this unifi cation on internal, purely cogni- tive grounds, e.g. by a tendency to paradigmatic economy and saliency. However, at least as an additional factor, contact cannot be dismissed even in such cases. B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 211 Generally, in many cases the results of change converge, at least superfi cially, even though they are most probably outcomes of diff erent factors. Cases in point are phonological and word prosodic features like akan’e, which occurs not only in entire northern Belarusian (to which the rural vernacular focused on here belongs), but also in some Aukštaitian dialects of Lithuanian, as well as aff rication of [d, t] and epenthetic [l‘] (see features (iii, iv, ix) in List 1 ). Note that when such features are investigated, the researcher – even if s/he can be suffi ciently sure about their unrelated origin in contacting East Slavic and Lithuanian varieties – has to account for the perspective of the dialect speaker for whom the genesis of what s/he hears in oral communication is not acces- sible. Instead, multidialectal speakers adapt to what they perceive in immedi- ate daily communication, in which case phonological and prosodic features salient to him/her are likely to be unifi ed in a sort of dialectal diasystem regardless of the motifs of their origin. Of course, the same applies, mutatis mutandis , for features in morpho(no)logy or syntax. As elaborated on in 1.1, the decision about whether, and to what extent, contact is the decisive factor in change hinges on the (implicit or explicit) null hypothesis assumed by the researcher provided s/he looks at a larger area showing clines for some specifi c structural features. If we look at the BSCZ from the broader perspective of the Eastern part of the CBA, rural Belarusian turns out to virtually in no way display any remarkable peculiarities since it fi ts perfectly into both the larger cline and the specifi c features of the BSCZ. Th is fi t holds for both positive and negative evidence: on the one hand, rural Belarusian shows features ubiquitous in most regions of the Eastern part of the CBA like, for instance, the case marking pattern of indefi nite quantity or the use of predicative resultative participles and the use of a particle marking epis- temic necessity ( mus’i ‘(most) probably’) common to virtually all varieties spoken on the territory of the former GDL (see 2.3). On the other hand, in rural Belarusian, we apparently do not encounter any kind of homonymy between declensional forms or of argument-coding patterns that are absent from the BSCZ but occur in the Russian dialect area neighboring to it in the northeast, (south)west and/or south (see 2.4). Moreover, this positive – negative balance of fi ts applies even if we examine the functional distribution of forms (or constructions) such as the resultative participles in {vši} (see 2.3) more closely. Th e restriction of their diathesis orientation is in perfect har- mony with the grammatical pattern in Baltic and diff ers (together with the Baltic perfect paradigm) from the functional distribution of resultative parti- ciples in NW-Russian dialects (where we in turn fi nd coincidences with Finnic). At this point, again, we have to add that it would be premature to conclude that Baltic infl uence is the trigger for this type and functional range 212 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 of resultative predicates to occur. For these participles have a very long, con- tinuous history in this part of Slavic (cf. Wiemer/Giger 2005: 40f.). Th e pro- ductive use of anteriority participles in the Baltic languages (inter alia as the core of their perfect paradigms) can, however, plausibly be regarded as a factor that has helped preserve the forms and functions of {vši}-participles in the Slavic varieties of the former GDL and their modern counterparts (including rural Belarusian). Finally we come to the question of whether genealogical closeness has an impact on structural features and which ones are more likely to be aff ected by contact. More precisely, is Lithuanian infl uence in rural Belarusian more restricted because it is less closely related than the surrounding Slavic varieties? Here we should admit that some features of List 1 have either remained entirely unaff ected by Lithuanian, or that such an infl uence is at least doubtful or can, to the best of our knowledge, be considered as marginal. Lithuanian infl uence is tangible in phonology, prosody and syntax (let alone in the lexi- con, which has not been considered here), but in morphology and morpho- nology its impact seems to be rather negligible. We have already commented on the unifi cation of the gen.pl -ending and the loss of neuter gender. Apart from these features, the following appear to be restricted to an “internal Slavic business”: (a) animacy distinctions; (b) the choice of derivational suffi xes of verbs; (the patterns of (East) Slavic have quite heavily infl uenced some Lithuanian dialects, but not the other way around) (cf. Wiemer 2009: 359-363) (c) declensional endings (with the rare exception of {ai} marking the dat.sg of the a -declension; see 2.2.1); (d) distribution of conjugation classes (not treated in this article); (e) morphonological alternations in the present stem (/d → dž/ for traditional Belarusian vs. /d --> ž/ for Russian). In fi nishing our pilot exploration of the specifi cs of rural Belarusian as the core representative of the BSCZ, we want to stress that we did not intend any exhaustive examination of its features. It has been our endeavor to show that rural Belarusian is the most typical representative of the BSCZ since its struc- tural properties, together with their enormous variation, demonstrate what one has to reckon with in unembellished rural speech in which typical features of the BSCZ converge most spontaneously. Th e outcome of diff erent layers on the intersection between East Slavic dialect continua, long-lasting contact with Baltic dialects and the interference of standards from beyond the BSCZ (Russian, Polish) and their local varieties provides an ideal playground for B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 213 multifarious imminent changes refl ecting the convergence of factors from these layers, to which we have to add tendencies toward change which are probably conditioned on cognitive grounds. From an areal linguistic point of view, rural Belarusian is sort of crucible of the BSCZ, it highlights, as it were, features of the latter, which can, in turn, be seen as part of a larger continuum. For the BSCZ can either concentrically be “inserted” into a larger area, or it can be viewed as a region where diff erent continua intersect. Th e whole region can thus be regarded as a buff er zone or something like a Zentrum einer Konvergenzlandschaft , as Stolz ( 1991 ) baptized the contact zone with Latvian, Livonian and Estonian as its core. References Ambrazas , Vytautas . 2001 . On the development of the nominative object in East Baltic . In Dahl Östen and Maria Koptjevskaja-Tamm (eds.), Th e Circum-Baltic Languages (Typology and Contact) , vol. 2: Grammer and Typology , 391 – 412 . Amsterdam-Philadelphia . Anan’eva , Natal’ja E. 2008 . Periferijnye pol’skie dialekty kak rezul’tat vzaimodejstvija s vostočnoslavjanskimi i baltijskimi jazykami. In Kalnyn‘ , Ljudmila Ė. (ed.), Issledovanija po slavjanskoj dialektologii 13: Slavjanskie dialekty v situacii jazykovogo kontakta (v prošlom i nastojaščem) , 72 – 79 . Moskva . Avanesa ŭ , Ruben , Kondrat Atraxovič , and Juzefa Mackevič . 1968 . Linhvistyčnaja heahrafi ja i hrupo ŭ ka belaruskix havorak . Minsk . Avanesov , Ruben I. , Orlova , Raisa G . 1965 2 . Russkaja dialektologija . Moskva : Nauka . Blinava , Ėvelina and Ėudakija Mjacel’skaja. 1969 . Belaruskaja dyjalektalohija . Minsk . Blinava , Ė. and Ėudakija Mjacel’skaja . 1991 . Belaruskaja dyjalektalohija : Praktykum . Minsk . Breza , Edward and Jerzy Treder . 1981 . Gramatyka kaszubska (Zarys popularny) . Gdańsk . Bromlej , Sof ’ja V. , Bulatova , Lidija N. 1972 . Očerki morfologii russkix govorov . Moskva . Bukrinskaja , Irina A. , Karmakova , Ol’ga E. , Ter-Avanesova , Aleksandra V. 2008 . Govory belorussko-russkogo pogranič’ja . In Kalnyn‘ , Ljudmila Ė. (ed.), Issledovanija po slavjanskoj dialektologii 13: Slavjanskie dialekty v situacii jazykovogo kontakta (v prošlom i nastojaščem) , 118 – 179 . Moskva . Čagiševa , Vera I . 1968 . K voprosu o kategorii srednego roda v mestnyx govorax . Pskovskie govory II : 146 – 159 . Pskov . DABM Avanesa ŭ , Ruben I. , Krapiva , Kandrat K. , and Juzefa F. Mackevič . (eds.). 1963 . Dyjalektalagičny atlas belaruskaj movy (Ustupnyja artykuly, davedačnyja matėryjaly i kamentaryi da kart) . Minsk . Erker , Aksana . 2009 . Faktor jazykovogo kontakta v issledovanii osobennostej fl ektivnoj morfologii ‘Movy prostaj’ (na primere okončanij roditel’nogo padeža suščestvitel’nyx) . In Scholze , Lenka and Björn Wiemer (eds.), Von Zuständen, Dynamik und Veränderung bei Pygmäen und Giganten (Festschrift für Walter Breu zu seinem 60. Geburtstag) , 293 – 303 . Bochum . Erker, Aksana. (forthcoming): Specifi ka imennyx paradigm v belorusskom smešannom govore na belorussko-litovskom pograničje. Balto-slavjanskie issledovanija 18. Moskva. Grek-Pabisowa , Iryda . 1992 . Język polski na Kresach północno-wschodnich. Legendy a rzeczywistość . Warszawa . 214 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 Grek-Pabisowa , Iryda and Irena Maryniakowa . 1999 . Współczesne gwary polskie na dawnych kresach północno-wschodnich . Warszawa . Grinaveckene , Ėlena (= Grinaveckienė) and Juzefa Mackevič . 1979 . Nekotorye neobyčnye mikrosistemy v dialektnoj strukture belorusskogo govora okrestnostej Opsa. Baltistica XV (1) : 52 – 65 . Hentschel , Gerd . 2008 . On the development of infl ectional paradigms in Belarussian trasjanka: the example of demonstrative pronouns . In Hentschel , Gerd and Sjarhej Zaprudski (eds.), Belarusian Trasjanka and Ukrainian Suržyk (Structural and social aspects of their description and categorization) , 99 – 133 . Oldenburg . Jankowiak , Mirosław . 2009 . Gwary białoruskie na Łotwie w rejonie krasławskim . Warszawa . Kardelis , Vytautas and Björn Wiemer . 2003 . Kritische Bemerkungen zur Praxis der Erstellung litauischer Wörterbücher, insbesondere von Mundarten – am Beispiel des slavischen Lehnguts und des ‘veikslas’ . In Ostrowski , Norbert and Ona Vaičiulytė-Romančuk (eds.), Prace bałtystyczne. Język, literatura, kultura , 45 – 72 . Warszawa . Kasatkina , Rozalija F. 2008 . Artikli, predartikli i psevdoartikli v slavjanskix jazykax . In Moldovan, Aleksandr M. , Petr M. Arkad’ev et al . (eds.), Slavjanskoe jazykoznanie (XIV Meždunarodnyj s"ezd slavistov, Oxrid, 10-16 sentjabrja 2008 g.) . Doklady rossijskoj delegacii , 305 – 321 . Moskva . Kittel , Bernhard , Diana Lindner , Sviatlana Tesch , and Gerd Hentschel . 2010 . Mixed language usage in Belarus: the sociostructural background of language choice . International Journal of the Sociology of Language 206 : 47 – 71 . Koptjevskaja-Tamm , Maria . 2001 . “A piece of the cake” and “a cup of tea”: Partitive and pseudo- partitive nominal constructions on the Circum-Baltic languages . In Dahl , Östen and Maria Koptjevskaja-Tamm (eds.), Th e Circum-Baltic Languages (Typology and Contact), vol. 2: Grammer and Typology , 523 – 568 . Amsterdam-Philadelphia . Koptjevskaja-Tamm , Maria and Bernhard Wälchli 2001 . Th e Circum-Baltic Languages: An Areal-Typological Aproach . In Dahl , Östen and Maria Koptjevskaja-Tamm (eds.), Th e Circum-Baltic Languages (Typology and Contact) , vol. 2: Grammer and Typology , 615 – 750 . Amsterdam-Philadelphia . Kuraszkiewicz , Władysław . 1963 2 . Zarys dialektologii wschodniosłowiańskiej (z wyborem tekstów gwarowych) . Warszawa . Kuznecov , Petr S. 1954 . Russkaja dialektologija (Učebnik dlja učitel'skich institutov) . Moskva . Kuznecov , Petr S. (red.) . 1973 . Russkaja dialektologija (Učebnoe posobie dlja studentov ped. in-tov po spec. № 2101 „Rus. jaz. i literatura“) . Moskva . Kuz’mina , Irina B. and Elena V. Nemčenko . 1964 . K voprosu o konstrukcijax s formoj imenitel’nogo padeža imeni pri perexodnyx glagolax i pri predikativnyx narečijax v russkix govorax . In Avanesov , Ruben I. (ed.), Voprosy dialektologii vostočno-slavjanskix jazykov , 151176 . Moskva . Lipovskaja , N.A. 1983 . Forma roditel’nogo padeža množestvennogo čisla suščestvitel’nyx ženskogo roda s osnovoj na parnyj tverdyj soglasnyj . In Avanesov , Ruben I. (ed.), Russkie narodnye govory (lingvogeografi českie issledovanija) , 93 – 99 . Moskva . Mackevič , Juzefa F. 1959 . Marfalohija dzejaslova ǔ belaruskaj move . Minsk . Morita , Koji . 2006 . Przemiany socjolingwistyczne w polskich społecznościach na Litwie (region trocki) i Białorusi (region iwieniecki) . Warszawa . Nau, Nicole. (forthcoming): Modality in an areal context: the case of a Latgalian dialect. In Wiemer, Björn, Bernhard Wälchli, and Björn Hansen (eds.), Grammatical replication and borrowability in language contact . Accepted by Mouton de Gruyter (Trends in Linguistics). Seržant , Ilja A . 2005 . Otnositel’naja xronologija osnovnyx fonetičeskix izmenenij v istorii verxnelatyšskogo dialekta . Acta Linguistica Lithuanica 55 : 39 – 89 . Seržant, Ilja A. 2010. Phonologische Isoglossen des Hochlettischen, Nord-Ost-Litauischen, Nord-West-Russischen und Weissrussischen. Baltic Linguistics 1, 193-214. B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 215 Skudrzykowa , Adolna , Jolanta Tambor , Krystyna Urban , and Olga Wolińska . 2001 . Gwara śląska – świadectwo kultury, narzędzie komunikacji . Katowice . Smułkowa , Elżbieta . 2010 . Problem zapożyczeń w warunkach rozpadającej się społecznej wielojęzyczności. Rozważania wstępne . Studia Białorutenistyczne 2010-4 : 283 – 295 . Sobolev , Andrej N . 1998 . O predikativnom upotreblenii pričastij v russkix dialektax . Voprosy jazykoznanija 47-5 : 74 – 89 . SRA: Faßke , Helmut (ed.). 1975 . Serbski rěčny atlas / Sorbischer Sprachatlas , vol. 11: Morphologie. Die grammatischen Kategorien – Die Paradigmatik des Substantivs . Bautzen . Stolz , Th omas . 1991 . Sprachbund im Baltikum? Estnisch und Lettisch im Zentrum einer sprachlichen Konvergenzlandschaft . Bochum . Sudnik , Tamara M. 1975 . Dialekty litovskogo-slavjanskogo pogranič’ja (očerki fonologičeskix sistem) . Moskva . Trubinskij , Valentin I . 1984 . Očerki russkogo dialektnogo sintaksisa . Leningrad . Turska , Halina . 1995 . O powstaniu polskich obszarów językowych na Wileńszczyźnie . Vilnius . [photomechanic reprint of: Studia nad polszczyzną kresową I . Ed. by Janusz Rieger and Wiaczesław Werenicz. Wrocław etc., 1982 , 19 – 122 .] Urbańczyk , Stanisław . 1984 7 . Zarys dialektologii polskiej . Warszawa . Wälchli, Bernhard. (forthcoming): Grammaticalization clines in space: zooming in on synchronic traces of diff usion processes. In Wiemer, Björn, Bernhard Wälchli, and Björn Hansen, N. (eds.), Grammatical replication and borrowability in language contact . Accepted by Mouton de Gruyter (Trends in Linguistics). Wiemer , Björn . 2003a . „ Mowa prosta“ - Präliminaria zu einer strukturellen Beschreibung . In Blankenhorn , Renate , Joanna Blaszczak , and Robert Marzari (eds.), Beiträge der Europäischen Slavistischen Linguistik (POLYSLAV) 6 , 227 – 237 . München . Wiemer , Björn . 2003b . Dialect and language contacts on the territory of the Grand Duchy of Lithuania from the 15 th century until 1939 . In Braunmüller , Kurt and Gisela Ferraresi (eds.), Aspects of Multilingualism in European Language History , 105 – 143 . Amsterdam, Philadelphia . Wiemer , Björn . 2003c . Zur Verbindung dialektologischer, soziolinguistischer und typologischer Methoden in der Sprachkontaktforschung (am Beispiel slavischer und litauischer Varietäten in Nordostpolen, Litauen und Weißrußland) . Zeitschrift für Slawistik 48-2 : 212 – 229 . Wiemer , Björn . 2004 . Population linguistics on a micro-scale. Lessons to be learnt from Baltic and Slavic dialects in contact . In Kortmann , Bernd (ed.), Dialectology Meets Typology (Dialect Grammar from a Cross-Linguistic Perspective) , 497 – 526 . Berlin , New York . Wiemer , Björn . 2006 . Osuščestvlenie soglasnyx v « move prostoj » kak pokazatel’ izmenenij v uslovijax kontaktov . Acta Baltico-Slavica 30 : 439 – 452 . Wiemer, Björn. 2009. Zu entlehnten Präfi xen und anderen morphosyntaktischen Slavismen in litauischen Insel- und Grenzmundarten. In: Scholze, Lenka and Björn Wiemer (eds.): Von Zuständen, Dynamik und Veränderung bei Pygmäen und Giganten (Festschrift für Walter Breu zu seinem 60. Geburtstag) , 347-390. Bochum . Wiemer, Björn. (forthcoming1): Metodičeskie zametki k issledovaniju „Movy prostoj“. Balto- slavjanskie issledovanija 18, Moskva. Wiemer, Björn. (forthcoming2): On the rise, establishment and continued development of subject impersonals in Polish, East Slavic and Baltic. In Kittilä, Seppo and Leonid Kulikov (eds.), Diachronic typology of voice and valency-changing categories. Amsterdam, Philadelphia. Wiemer, Björn. (in prep.): Die Mowa prosta: ein Mischidiom als Schmelztiegel im Sprachkontakt (Forschungsüberblick). Wiemer , Björn and Markus Giger . 2005 . Resultativa in den nordslavischen und baltischen Sprachen (Bestandsaufnahme unter arealen und grammatikalisierungstheoretischen Gesichtspunkten) . München . 216 B. Wiemer & A. Erker / Journal of Language Contact 4 (2011) 184–216 Wiemer, Björn and Björn Hansen. (forthcoming): Assessing the range of contact-induced grammaticalization in Slavonic. In Wiemer, Björn, Bernhard Wälchli, and Björn Hansen. (eds.), Grammatical replication and borrowability in language contact . Accepted by Mouton de Gruyter (Trends in Linguistics). Zaprudski , Sjarhej . 2008 . Ma ŭ lenčaja akamadacyja i peraključėnne koda ŭ u pracėse mižkul’turnaj kamunikacyi: vypadak Belarusi . In Hentschel , Gerd and Sjarhej Zaprudski (eds.), Belarusian Transjanka and Ukrainian Suržyk (Structural and social aspects of their description and categorization) , 57 – 97 . Oldenburg . Zemskaja , Elena . 2000 . O jazyke russkogo zarubež’ja . In Zybatow , Lew N . (ed.), Sprachwandel in der Slavia. Die slavischen Sprachen an der Schwelle zum 21. Jahrhundert (Ein internationales Handbuch) , Teil 2 , 767 – 785 . Frankfurt/M. etc . Zemskaja , Elena A. and Margarita V. Kitajgorodskaja . 1984 . Nabljudenija nad prostorečnoj morfologiej . In Zemskaja , Elena A. and Dmitrij N. Šmelev (eds.), Gorodskoe prostorečie (Problemy izučenija) , 66 – 102 . Moskva . Zemskaja , Elena A. and Dmitrij N. Šmelev . (eds.). 1984 . Gorodskoe prostorečie (Problemy izučenija) . Moskva .

Journal

Journal of Language ContactBrill

Published: Jan 1, 2011

Keywords: BELARUSIAN; MIXED DIALECTS; AREAL CONVERGENCE; BALTIC-SLAVIC CONTACT ZONE; CIRCUM-BALTIC AREA

There are no references for this article.