1. Introduction The study of language contact has been significantly shaped by the work of Victor Friedman. His contributions are so profound that it is impossible to do research within the field of contact without being influenced by, and in dialogue with, his work, either explicitly or implicitly. The present article, inspired by his foundational work on the Balkans, addresses the issue of contact between East Slavic varieties.1 The Balkan Sprachbund is characterized by a shared set of grammatical and lexical features that cannot be explained by inheritance from a common source but rather as the result of language contact (Friedman 2006). Cutting across the Balkans is the Balkan Slavic continuum of Bulgarian and Macedonian (Friedman 1994), analogous to the East Slavic language/dialect continuum that extends from Poland in the west into Russia in the east (Section 2). Here I argue for the existence of an East Slavic contact zone that in some ways similar to the Balkan zone and in some very different (Section 3). As in South Slavic, Eastern Slavic dialects reflect their diachronic development in synchronic spatial terms (Friedman 1994). Despite the number of non-Slavic languages spoken today, and spoken historically within the geographic territories of the East Slavic zone, the primary contact is between closely-related languages: East Slavic varieties in contact (Russian and Belarusian, Russian and Ukrainian) or West and East Slavic varieties (Polish and Ukrainian, Polish and Belarusian). The present study examines the contact between East Slavic.
Russian, Ukrainian and Belarusian are sister languages belonging to the East Slavic branch of the Slavic family; contact between them provides information of a different sort than found in many studies of language contact, which focus on the interactions between more distantly-related or unrelated languages. In the East Slavic contact zone, the situation is further complicated by migration patterns and language policies. In both Belarus and Ukraine this contact has resulted in mixed languages (Trasjanka from Belarusian and Russian, Surzhyk from Ukrainian and Russian) that differ from one another and from the dialects of Russian, Belarusian, Ukrainian and Polish (which is in contact with these languages in the western border regions). The labels of these two varieties are themselves derogatory: surzhyk refers to a mixture of wheat and rye, and trasjanka to a mixture of wheat and straw. As these names suggest, both varieties are low prestige; speakers tend to think of them as “bad” Ukrainian or Belarusian. Despite their derogatory connotations, I use the labels Surzhyk and Trasjanka here as they are commonly used by other linguists and non-specialists. Although an argument could be made for replacing them with more neutral terms such as Russian-Ukrainian (or Belarusian-Russian) mixed speech, I continue to use these labels precisely because they invoke specific claims about the linguistic status of these two varieties, including the question of whether it is appropriate to consider them two unified varieties, or whether there is more variation in their structures than such an analysis would imply.
This article thus examines the linguistic structures of Surzhyk and Trasjanka, and examines their linguistic statuses: as mixed languages versus idiosyncratic, one-off combinations or code-mixes; as each representing a single variety or multiple varieties (i.e., a single Surzhyk or multiple Surzhyks; see Flier 2008); and as interlanguages that are part of a transitional stage to Russian monolingualism (Zaprudski 2007). Although many of the features of Russian found in both Surzhyk and Trasjanka are lexical, there are a number of morphosyntactic borrowings, which provide the focus here. Such features include the use of prepositions and numerical constructions, and changes in government patterns (Flier 1998). The study of contact between closely-related varieties provides particular challenges in distinguishing shared innovation, language-internal change and contact-induced change. The facts of Surzhyk and Trasjanka are of interest for theoretical linguistics, in particular for a theory of mixed languages (Section 5).
Here I argue that both Surzhyk and Trasjanka are mixed lects: they contain elements of two languages but are not sufficiently stable to be defined as mixed languages whose systems are conventionalized. They are not cases of code-mixing in the ordinary sense (Muysken 2000) inasmuch as there is considerable amount of predictability as to which items come from which source language. At the same time, there is not an absolute systematicity, but rather strong tendencies with variation. In this sense Surzhyk and Trasjanka are not quite code-mixes and not quite mixed lects. In such cases, where new mixed lects are in the process of being established, there is a strong tendency for a fair amount of variation until at least the fourth generation of speakers of the mixed variety; stabilization requires significant reduction in variation (Trudgill 1986). Auer (1999) proposes a continuum of language alternation phenomena, with a transition from code-switching (CS) on one end toward increasing structural sedimentation and fused lects (FLs) on the other end; fused lects are stabilized, forms are predictable. In this schema, language mixing (LM) is located between CS and FL such that CS LM FL. Language mixes, or mixed lects, are transitional from CS to FL; both Surzhyk and Trasjanka represent LMs. Language mixes such as these exhibit heavy code-mixing, with strong tendencies in terms of which forms are supplied by each of the two source languages. They differ from code-switches in that the source material is not random. In contrast, fused lects are stable and conventionalized with predictable structures.
From a theoretical standpoint, mixed lects such as Surzhyk and Trasjanka are worthy of study as prime examples of emergent languages: their forms and structures are not fully conventionalized but rather are still in flux. They provide a contrastive study in that both involve an East Slavic language in contact with Russian, but differ in the social dynamics of language use and prestige. I return to this point in Section 5.
2. The East Slavic Language-Dialect Continuum The Slavic languages are generally classified into three branches, a classification which takes into account not only geography but also the distribution of a core set of features, such that each branch stands for a higher node on the Indo-European tree, representing a period of unity after the other two groups had separated. These three branches are: West (Polish, Czech, Slovak, Upper and Lower Sorbian, Kashubian); South (Bulgarian, Macedonian, Serbian, Bosnian, Croatian and Slovenian); and East (Belarusian, Russian and Ukrainian). Another classification system divides the North Slavic languages (East and West Slavic) from South Slavic. In both systems, classification is diachronic and synchronic, based on a historical period of unity, separate from the other branch(es) and on modern structural similarities.
Schematically, Belarussian is situated to the northwest of the continuum, with Polish varieties on the western edges and Russian on the eastern edges; Ukrainian stretches to the south of Russian and Belarusian, flanked on the west by Polish and on the east by Russian. The political boundaries that divide the modern nations of Russia and Ukraine do not map neatly onto language boundaries, and varieties spoken in border regions are closer to one another than to the standard varieties. The structural differences between Russian and Belarusian are arguably more dialect-like than language-like (Hentschel 2014). The Western Russian dialect zone is of particular relevance: it covers the territory of the initial East Slavic expansion and is transitional in the East Slavic zone. It is divided into Northwestern and Southwestern subgroups. A number of linguistic features found in the Western Russian dialect zone are shared with Belarusian and Ukrainian, something that complicates the analysis of contact effects in East Slavic.
The territories of modern Belarus and Ukraine have both been areas of contact for centuries, in particular with Polish and Russian, along with other, non-Slavic languages. Both have historically been border regions, and Russian and Polish were socially and politically dominant. The development of written standard languages is closely tied to the history of the region; the linguistic and socio-political histories of the region are deeply intertwined. Old Church Slavic, a liturgical language based on South Slavic varieties, initially functioned as the written standard for Christianized Slavs from the 10th century. In the 14th century Ukrainian territory was incorporated into the Grand Duchy of Lithuania and a new written language emerged for Belarusian and Ukrainian. This written language included OCS, Ukrainian, Belarusian and Polish elements and was used in some form until the late 18th century when non-Ukrainian items were removed and a new standard Ukrainian emerged (Shevelov 1993:948). The histories of both the Belarusian and Ukrainian languages have been instrumental in shaping the social dynamics of their use vis-à-vis Russian today.
During the Soviet period, Belarusian and Ukrainian served as titular languages for each of these Soviet Republics, while Russian functioned as the single language of wider communication. By 1989, both regions had sizeable populations of ethnic Russians (who spoke Russian as a first language). This includes some 11.3 million in Ukrainian SSR and, These Russians made up large percentages of the local populations, in particular in urban centers and the capitals (Heleniak 2004:101-02). By the time of the break of the USSR, not only had Russian been well-established as the national lingua franca by the Soviet government, but it had also become fixed as the first language of significant percentage of the residents of both Belarus and Ukraine.
In the post-Soviet era, much of the discussion of national identity and nation-state building has been shaped by invoking language ideologies, but these discussions have played out very differently in Belarus and Ukraine. The modern-day arguments cannot be understood without understanding the linguistic realities in both countries and their relationship to Russia, as well as the historical and social settings in which the East Slavic languages are used. The sociolinguistic context is mapped out for Belarus in Section 3 and for Ukraine in Section 4.
2.1. Demographics Belarusian and Ukrainian are titular languages in their territories and have relatively large speaker bases. The language ecologies differ considerably in the two countries: in Belarus there are 2,200,000 speakers of Belarusian and 6,670,000 of Russian (2009 census), thus Belarusian is a minority language. In Ukraine there are 32,000,000 speakers of Ukrainian and 8,330,000 speakers of Russian (2001 census).
Language Usage in Belarus, Russia and Ukraine2
The language issue is so politicized in both regions that it is difficult to find hard, reliable data on language knowledge and use. For example, one survey conducted in 2012 showed that 49.9% of respondents considered Ukrainian to be their native language, but only 44.9% claimed to speak it regularly; moreover, the study showed that 39.3% of the respondents speak Russian regularly (Rating Group 2012). Native language is not necessarily understood as L1 or as a primary language; in many cases respondents may be interpreting the term as referring to their ancestral language. Other data show that speakers claiming Ukrainian to be their native language are much more likely to speak it with their parents (69.7%) than with colleagues (12.4%) (Vakhtin et al. 2003). Regardless, survey data indicate that fewer people speak Ukrainian than claim it as native. Moreover, these numbers change with the political climate. In reality, knowledge and use of the languages vary in urban versus rural settings with rural more likely to have higher levels of Ukrainian, and geographically with eastern parts of Ukraine having a much higher use of Russian and western areas a much lower. In Belarus, the situation is quite different, in that Russian is dominant and primary for 70% of the population (2009 census).
2.2. Structural Comparison In order to understand the complexities of the two mixed lects, Surzhyk and Trasjanka, it is important to understand how they draw on features of the languages which form their basis. With two closely-related languages undergoing changes due to contact with one and the same language, we are in a position to compare the emergence of mixed languages by looking at their structures. The standard languages serve as the basis of comparison here, with reference to dialect differences; basic information is drawn from Mayo (1993) for Belarusian; Shevelov (1993) for Ukrainian; and Timberlake (2004) for Russian, unless otherwise noted. Although the description here is anchored in the standard languages, dialect variation and the East Slavic language/dialect continuum are central to the ultimate analysis. Rather than give a full account of the different systems, I highlight the similarities and salient differences here that are relevant for understanding the structures of Surzhyk and Trasjanka. Tables 1 and 2 present an overview of these differences. Critically, the linguistic systems of Belarusian, Russian and Ukrainian are sufficiently alike to facilitate code-switching and mixing. In certain mixes, it is impossible to determine the source language for a particular item, because it could belong to either system. There are a number of phonetic and phonological features shared across the continuum (Hentschel 2014:99-100), so that phonetics cannot be used as a diagnostic for determining language an individual token belongs to because it could be either. In other cases, however, the differences make it possible to determine a source code for particular items and to speak of the source of the phonology of the mixed lects. Two examples serve to illustrate these issues: (1) with Ukrainian and Russian elements, and (2) with Belarusian and Russian. Here and throughout, Russian is italicized, Belarusian/Ukrainian is boldfaced, and items that could belong to either (or both) codes are in plain text. Where the source material provides relevant information about the phonetics of a given item, it is underlined:3
In (1) Kent notes that prosto conforms to Ukrainian phonology in that the unstressed /o/ in the second syllable does not lower as expected in Russian (Section 2.2.1). The NP xarošij tsvetok ‘good flower’ is unambiguously Russian, and vse ‘everything’ unambiguously Ukrainian, but otherwise the words could be either Russian or Ukrainian and one cannot define a matrix or primary code. A similar situation holds in (2), where a number of items are unambiguously Russian (such as nadə kuda-tə ‘have.to somewhere’ or štop ‘in order to’) and some Belarusian (e.g., the preposition u ‘to’) but a large number of elements could be either language, or simultaneously could be both.
2.3. Phonemic Systems Given their shared history, it is not surprising that the phonemic systems of all three languages are quite similar, although there are some differences that are highly salient and significant for both production and perception. These differences involve: (1) vowel quality; (2) palatalization of consonants; and (3) distribution of a voiced velar fricative or stop. Belarusian and Russian have five vowel phonemes: /i, u, e, o, a/ and an allophonic variant of /i/, [y], (orthographically ы) which appears after unpalatalized consonants in stressed position. The Ukrainian vowel system is more distinct. There are six vowel phonemes. One that is particularly distinct from Belarusian/Russian is orthographic ы, which is a central back high vowel (allophone) in these two languages and is a central mid-vowel phoneme in Ukrainian.
Belarusian and Russian both have vowel reduction in unstressed syllables. However, the vowel in pretonic position following a palatalized consonant is realized as [a] in Belarusian and as [i] in (standard) Russian, a core salient difference in terms of both production and perception (Hentschel 2014:99-100). Some Russian dialects exhibit the Belarusian reduction pattern, but they are viewed as substandard and are stigmatized by speakers of standard Russian. Vowel reduction is written in Belarusian but not in Russian, and so there is a core visual difference between the two languages. Ukrainian vowel reduction differs. The phoneme /o/ is realized as an open [ɔ] in stressed position but as [o] in pretonic position before /u/ or /i/
The consonantal phonemic inventories are similarly comparable among the three languages and similarly have some key differences. Belarusian (including the standard language and dialects), Ukrainian and Southern Russian dialects use a velar fricative /Ɣ/ where Russian (standard, northern and central dialects) uses a stop /g/ (Hentschel 2014:99-100). The difference between the stop and the fricative is highly salient and a strong marker that identifies a speaker of standard Russian or not. Other salient consonantal differences include two different sets of palatalized consonants. Palatalized /rj/ is lost in Belarusian but maintained in Russian and Ukrainian. East Slavic /tj/ and /dj/ are maintained in Russian but are affricated in Belarusian, as /tcj/ and /dzj/. Prothetic consonants are found in Belarusian, including the standard language, where Russian has none, as in BR hètamu, R ètomu ‘this.masc/neut.dat.sg’ and BR ën, R on ‘he.’ Prothetic v- is found in BR before stressed /u, o/, as in BR vulica, R ulica.
These phonological differences are sufficiently salient to invoke one or the other language. In Trasjanka, the phonology is largely Belarusian, and in Surzhyk, it is primarily Ukrainian, with Russian lexemes adapted to the other language.
2.4 . Morphosyntactic Features There are a number of core morphosyntactic features that distinguish the three standard languages from one another, and use of one or the other of these features can invoke one or the other code. In some cases, certain linguistic elements are found in Belarusian or Ukrainian dialects but not in the corresponding standard language, but do correspond to Russian. (See Ramza 2008 for Belarusian.) Thus the source for some items in Surzhyk or Trasjanka may be the dialect, not the standard language and not Russian. Therefore the East Slavic dialect continuum confounds classification of the mixed varieties themselves, as well as identification of the source codes. That said, a few salient differences between Belarusian and Ukrainian on the one hand, and Russian on the other, are worth noting here. This section provides a selective illustration of just two salient differences; more complete accounts can be found in the descriptions cited here.
2.4.1. Prepositions Prepositions are one area of notable correspondence and difference between the three languages. A set of prepositions can signal one code or the other, depending on how they are used. These are summarized in Table 2:
u + acc
u + acc
v + acc
u + loc
u + loc
v + loc
‘from, out of’
z + gen
z + gen
iz + gen
do + gen
do + gen
k + dat
za + acc
ab, pra + acc
o + loc
Some Correspondences between Prepositions in East Slavic
Table 2 simplifies the use and distribution of prepositions in each of the languages, highlighting only major differences. The preposition na is used in both Belarusian and Russian, for example, but is much more restricted in (standard) Russian.
2.4.2. Verbal Morphology There are two noteworthy areas of verbal morphology in both Surzhyk and Trasjanka. The 3rd person non-past form is notable in both. The Russian 3rd person (singular and plural) non-past verb forms end in a non-palatalized /t/, as in 3sg non-past čitaet ‘reads,’ 3pl non-past: čitajut ‘read.’ In Belarusian, the etymological palatalized /t’/ of the 3rd person is retained only in the second conjugation, with the expected phonological change to /c’/. (It is also found in reflexive first conjugation verbs.) Similarly in Ukrainian, the final consonant has been lost in the 3rd person singular (Ukr čytaje ‘reads’). In the past tense, the historical l-participle is maintained in Russian. Due to regular phonological changes, -l has been replaced by -v in Ukrainian and -u in Belarusian.Surzhyk takes Ukrainian morphology and Belarusian Belarusian morphology in the 3rd person non-past. In the past tense forms, again Belarusian and Ukrainian morphology are used for the masculine singular. These differences are summarized in Table 3:
3rd sg non-past ‘reads’
3rd sg past ‘read’
Some Differences in Verbal Morphology
These two sets of differences (prepositions and 3rd person verb forms) illustrate how a very simple change in a single form, in some cases a single phoneme, can invoke one language over the other while at the same time be completely comprehensible, even if the interlocutor is not fully fluent in this alternate code.
3. Belarusian, Russian and Trasjanka In Belarus, there are two competing standard languages, Belarusian and Russian, both of which have official status. Russian is dominant in social domains and official domains (administrative, governmental, education). This dominance is longstanding and can be traced at least to the 19th century, when Belarusian territory was part of the Russian Empire and the language was viewed as a peasant language. It was never established as an urban language; cities were instead dominated by Russian and Yiddish speakers. Despite the creation of a standard language in the early Soviet period, the language failed to establish a strong foothold due to a confluence of factors, including Stalinist purges of the Belarusian elite and immigration of Russian speakers to Belarus. The net result is that Russian has continued to maintain prestige since the breakup of the Soviet Union, in contrast to the linguistic situation in Ukraine (Giger and Sloboda 2008:317).
3.1. Trasjanka Trasjanka has three basic sets of linguistic components: (1) those that clearly come from Belarusian; (2) those that clearly come from Russian; and (3) elements that are shared by both languages, and thus could be interpreted as coming from one or the other language, or as coming from both at the same time. There is no clear distribution of languages along a lexical-grammatical split as argued for mixed languages (Matras and Bakker 2003). Rather, the structure of Trasjanka supports the hypothesis that mixed codes are indicators of the speakers having access to multiple languages simultaneously; the structure of the mixed language is the starting point (see Auer and Muhamedova 2005). Trasjanka is not an alternation of two distinct monolingual systems but rather the result of linguistic congruence (Section 5).
Hentschel (2014:117) proposes a hierarchy of elements, from those most likely to be in Belarusian to those most likely to be in Russian, given here in Figure 1:
Belarusian ↔ Russian
phonic elements < inflectional endings (with pronouns < with lexical words)
The hierarchy is graded, not absolute, and Belarusian and Russian elements from any of the categories can be found in Trasjanka. Instead, it is more likely that items on one end of the scale will be Russian, and on the other Belarusian. (Hentschel argues that a distinct split between the lexicon and grammar does not occur because the two languages are structurally and genetically so close.) This hierarchy is similar to borrowing hierarchies (see Matras 2007 for an overview and discussion), a point I return to in Section 5.
3.2. Trasjanka Like Surzhyk, Trasjanka is low-prestige but widely used, primarily alongside standard Russian, the language of education (Hentschel and Zeller 2012). Belarusian has long existed under the shadow of Russian; in western and northwestern Belarus, Belarusians themselves have considered Russian to be a standardized variety of their own language, not a separate language (Gustavsson 1997:1922). Trasjankathus stems from a combination of the intense Russification policies, the prestige of Russian and the linguistic similarity of Belarusian and Russian. Hentschel and Kittel (2011) report that 38% of respondents [n=1,200] named Trasjanka as their native language, 50% claimed it to be their first language, and 41% stated it was the language they primarily used. Comparing these responses about Trasjanka to Russian (30% native, 42% first and 55% primarily used) and Belarusian (49% native, 18% first and only 4% primarily used), we see that a very significant percentage of the population in Belarus claims quite a bit of knowledge of and, significantly, only a small percentage (4%) claim to use the titular language. Just which language (Russian, Belarusian or Trasjanka) is used depends on a combination of factors such as, significantly, with younger speakers the amount of Trasjanka decreases in favor of Russian, suggesting an ongoing shift to Russian. Thus the situation is Belarus can be characterized as replacive bilingualism (Zaprudski 2007).
Trasjanka typically adheres to Belarusian phonetics and intonation, a mixed morphology and mixed lexicon. For example, it is possible to find instances where the preposition is Belarusian and the nominal morphology Russian, as in ab čeloveke ‘about the person.’ This mixture is illustr ated in example (3), where the Russian elements are underlined and the Belarusian is in bold face:
ščas pagljažu iakie sapožki pradajuc’
now look.1sg what boots sell.3pl
‘I’ll take a look to see which boots are for sale.’ (Gustavsson 1997).
Phonetically there are Belarusian elements, such as the pronunciation of pagljažu (versus a more R paglji žu) with lexical elements from both. The verb pradajuc’ could be found in either language, but the final consonant (part of the 3rd person plural suffix) -c’ is BR, instead of the expected R -t. Although Trasjanka is typically defined as a jargon or being limited to casual speech, it is widespread in spoken speech of all levels; Gustavsson (1997) argues that there are few people who can speak without some admixture of Russian. It also occurs in written form, in particular in texts purporting to report actual speech. In current politicized, nationalist discourse in Belarus, the term Trasjanka is often used to refer to any kind of speech that deviates from both standardized Belarusian and Russian. (For linguists the term refers specifically to a mixed language that combines elements of both languages.) As with Surzhyk, there is some disagreement as to how regular the combinations are and to what extent they are idiosyncratic, and the amount of Russianisms may vary from speaker to speaker. These kinds of debates (for both languages) suggest that in each system the changes are ongoing, not yet complete.
As Hentschel (2014:95) points out, the linguistic system that is at least partially mixed dialects has a strong tendency to exhibit considerable variation as long as the two (or more) original lects continue to be used in different social settings. This is in line with the claim made by Auer (1999) that code switching, code mixing and fused lects (using his terminology) can co-occur in speech.
In his analysis of a large corpus of approximately 212,000 word forms in about 38,000 utterances of spoken conversations between friends, family members and neighbors in Belarus, Hentschel (2008) classifies utterances as (1) hybrid (with both Russian and Belarusian elements, not including those common to both); (2) Belarusian (with only Belarusian and common elements); (3) Russian (Russian and common elements); and (4) those utterances containing only elements shared by both languages and structurally possible in both. The results of the corpus analysis are given in Table 3:
Analysis of Belarusian Corpus (Hentschel 2014)
Following these figures, only one fourth of the corpus contains utterances in only Belarusian or Russian; mixes or cases of complete congruence constitute the remainder. Thus the overwhelming majority of utterances cannot be attributed to a single source language; in this sense Trasjanka is a mixed code (but see Section 5).
That said, the use of Trasjanka is in flux and appears to be being replaced by Russian. Current data show that speakers have an unbalanced knowledge of Russian and Belarusian, with younger generations using mixed speech only in informal settings, and most frequently with members of the older generation. Younger speakers have good or very good command of Russian, and so Trasjanka cannot be analyzed as an interlanguage in its current use (Hentschel and Zeller 2013; Hentschel 2014:102). Specifically, a study of the corpus shows a overall tendency for speakers to use primarily Russian, or a combination of Russian and mixed speech:
The majority of speakers interviewed could produce long blocks of speech in Russian only, or in hybrid utterances.
Only a few were able to produce large blocks of Belarusian speech, and here, too, their speech patterns showed a large number of mixed utterances. More specifically, only two speakers produced large blocks of Belarusian, Russian and mixed speech.
Only a few people use blocks of mixed speech without any blocks of Russian or Belarusian blocks.
This distribution of speakers and language allows Hentschel and Zeller to conclude that there has been an overall change in the status of Trasjanka. In the Soviet Union in the 1970s and 1980s, Trasjanka (or what they prefer to call mixed Belarusian mixed speech), was arguably an interlanguage, as proposed by Liskovec (2006), a variety typically used by L2 users who have not yet mastered the full linguistic system of L2 (or Russian, in this case). This is not, however, true for the children of these speakers, who have good or very good knowledge of Russian and use the mixed variety primarily in informal, family context, primarily (but not exclusively) in communications with the older generation (Hentschel and Zeller 2012:144; Hentschel 2014:102).
3.3. Summary In conclusion, there is no evidence that there is a stable, conventionalized mixed variety of Belarusian and Russian that serves as a single, homogenous lect spoken throughout Belarus, nor is there evidence for several such lects. Rather, there are patterns and tendencies as to what elements are taken from each of the source languages, as outlined in Figure 1. The corpus analysis presented in Hentschel and Zeller (2012) indicates that Trasjanka may perhaps be best analyzed as a transitional variety that is currently being replaced by Russian. Their study suggests an overall shrinkage in the domains in which Trasjanka is regularly used, which itself is indicative of its loss. While it arguably may have emerged due to imperfect learning of Russian, younger speakers can and do use Russian. This increase in Russian language proficiency is most probably fostered by measures of the government to lift prohibitions on the use of Russian, or more explicitly, to facilitate “freedom from Belarusian” (Giger and Sloboda 2008:323).
4. Ukrainian, Russian and Surzhyk The relationship between Russian and Ukrainian is somewhat different from that of Russian and Belarusian, linguistically and socio-politically. In Ukraine today, the Ukrainian language is a marker of a national, Ukrainian identity, one that is distinct from and decidedly not Russian, and for some people it has also become symbolic of a separate, independent Ukrainian state (Søvik 2010:7). It is the sole official national language, as stated in Article 12 of Law N 5029-VI of 3 July 2012.4 At the same time, there are large numbers of Russian speakers in Ukraine (Table 1), and use of one or the other language can invoke not only an ethnic identity, but also a political stance. Based on extensive survey data, Ukrainian speakers in Ukraine want to see the Ukrainian language used as the primary language in all domains, but have, in general, become reconciled to the widespread usage of Russian. In contrast, ethnic Russian-speaking citizens of Ukraine want the official bilingualism policy, at least in principle (Kulyk 2013). This speaks to an underlying tension as to whether the Russian language should have any place in Ukraine. The Russian-speaking population lives primarily in the western part of Ukraine, and so there is a long-standing regional polarization in terms of language use and attitudes (Arel and Khmelko 1996; Kulyk 2013). The situation has become even more charged since the Russian military intervention began in February 2014.
4.1. Standard Ukrainian and Standard Russian
Although Russian and Ukrainian are very similar, there are some core differences, and the two languages are not mutually intelligible. An estimated 38% of lexicon differs between the two languages (Tyshchenko 2000, cited in Kent 2012:9-10). Another 44% of the lexicon is identical in each language, and an additional 11% very similar. This latter category includes lexical items that differ only in one phoneme (such as R /kon’/ Ukr /kin’/ ‘horse’) due to regular sound changes. Taken together, the shared lexical items and the relatively large set of cognates that differ minimally phonologically facilitate comprehension and congruence.
4.2 . Surzhyk
In Ukraine today, Surzhyk is generally considered to be a “contaminated” form of Ukrainian, unsystematic and corrupt, showing random effects or intrusions from Russian. It is in fact regular and rule-governed, and there are structural hierarchies that affect how Russian and Ukrainian are combined in regular, predictable ways (Flier 1998). Language attitudes and ideologies are frequently intertwined with linguistic analyses of the phenomenon. As Flier (2008:40) points out, “discussions of Surzhyk are often filtered through the perspective of language ideology, in which case matters of prestige, pride, patriotism, class and intelligence inter alia are all significant.”
One ongoing challenge is the classification of Surzhyk since it is not a single, uniform linguistic variety but rather a set of varieties, with differences depending on several parameters: rural versus urban users; level of education of speakers; and time period involved (pre-Soviet, Soviet, post-Soviet). These have been analyzed as five different categories (Bilaniuk 2004, 2005) but are probably more accurately seen as different sociolects that have changed over time but can be united in one larger category of Surzhyk (Flier 2008).
Many of the features of Russian found in Surzhyk (and Trasjanka) are lexical, including a large number of lexical borrowings and calques of Russian words not normally found in Ukrainian or Belarusian. In addition, there are a number of morphosyntactic borrowings. (For a more complete inventory, see Vakhtin et al. 2003.) In Surzhyk these include the following:5
(1) Syntactic calques are common, especially in prepositional and numeral phrases, as well as head phrases, where government patterns differ between Ukrainian and Russian, as in R soveščanie po problemam, Sur narada po problemam ‘a conference on issues,’ using the preposition po ‘on,’ ‘concerning’ with the dative case, versus Ukr narada z problem, using z plus the genitive. (Use of the preposition po does not otherwise govern the dative in Ukrainian.) Time expressions follow the Russian model of the preposition v ‘at’ plus the accusative of a cardinal numeral which itself then governs the genitive (R v desjat’ časov, Sur v desjat’ godyn ‘at ten o’clock’) versus the Ukrainian pattern using o plus the locative case of a cardinal numeral and noun (Ukr o desjatij godynĭ) (Flier 1998:119).
As a broad category, prepositional phrases appear to be particularly susceptible to mixes. Kent (2012:103) argues that there is no predictable pattern here, with the exception of verbs of motion. They come from Ukrainian but select for Russian prepositions.
What is predictable and striking, however, is the use of Russian prefixes with both Ukrainian and Russian verbal stems. There is a strong preference for prefixes or Russian or common origin; those prefixes that could belong to both codes conform to the rules of Russian phonology (Vakhtin et al. 2003).
(2) Pronouns and adjectives take Ukrainian morphology. In nouns, the vocative follows the Russian pattern of being identical to the nominative, and the genitive singular of masculine nouns is generally -a, not -u, unlike Ukrainian, where the reverse is true. The dative singular of masculine animate nouns also typically follows the Russian -u, as opposed to the Ukrainian form in -ovi.
(3) Relatively regular use of the Ukrainian prepositional case -i with words of Russian origin: Sur na Zapad-i (Ukr na Zaxod-i) ‘in the west’ and -u as in Sur na ruskom jazyku (versus R jazyk-e) ‘in the Russian language,’ where zapad and jazyk are Russian, not Ukrainian, stems.
(4) Verbal paradigms largely follow Ukrainian forms. Notably, the 3rd person non-past ending in -t’ is not replaced by Russian non-palatalized -t; the past tense forms use the Ukrainian ending -v and not the Russian ending -l. This pattern is highly regular. At the same time, there is a strong tendency to form a 1st person plural imperative using the non-past 1st person form (on the Russian model).
These patterns are systematic but not consistent, across speakers and for individual speakers. In (5), the speaker alternates between the Ukrainian past-tense morpheme -v and the Russian morpheme -l:
Surzhyk, Kent (2012:87)
Ja jak by skaza-t’, rozvaln-uva-v-sia6
1sg how would say-inf worry-impf-2sg.m.pst-refl
brosi-l eta selo, rasshita-v-sia
leave-3sg.m.pst this village quit-2sg.m.pst-refl
i perejeha-l v rajon.
and move- to urban.center
‘How to put it, I got worried, left this village, quit, and moved to the urban center.’
Kent (2012:94-8) shows a split in the distribution of forms of the verb be according to tense. In the present tense, the source is Russian, but in the past, pluperfect and future it is Ukrainian. The unmarked present-tense form in Russian is zero. Kent’s examples of an overt present-tense form of ‘be’ in Russian are all in the 3rd person singular (est’), an overt form used in the standard language in certain contexts in existential and possessive sentences (Timberlake 2004:313-6). This distribution is, however, further complicated by the fact that the extended form of Ukr je is jest’, corresponding to the Russian form. The Ukrainian extended form is less frequent than je, but the existing correspondent in Russian could certainly support its use.
(4) Discourse markers, adverbs, particles and complementizers come from Russian (Kent 2012:6, Vakhtin et al. 2003). This category includes a relatively large number of items, including conjunctions such as esli ‘if,’ ili ‘or’ and no ‘but,’ and potom ‘then,’ as well as adverbs such as skučno ‘boring’ and particles such as vot ‘here.’ Vakhtin et al. (2003) provides a relatively exhaustive inventory, asserting that these particular items regularly come from Russian.
5. Conclusion: Surzhyk, Trasjanka and the Mixed Language Debate The present article provides only sketches of some of the key patterns found in Surzhyk and Trasjanka. They provide the basis for preliminary comparisons and conclusions.
5.1. Comparison of Surzhyk and Trasjanka The social situations in which Surzhyk and Trasjanka are embedded are similar with some striking differences. Russian enjoys official status in Belarus where it is a majority language but not in Ukraine, where Ukrainian is the sole official language and has majority status. In general Russian has higher status in Belarus than in Ukraine, although global statements like that mask the regional differences: where there are concentrations of Russian speakers, the overall prestige of Russian is understandably higher; in Ukraine as a whole at least one-third of the population reports daily use of Russian (see Section 2.1). Since social factors are often considered to be important in determining the outcome of language contact, it would be reasonable to expect different outcomes.
In comparing the structures of Surzhyk and Trasjanka, we can define the two possible source languages as being either non-Russian (i.e., Ukrainian or Belarusian in the case of Surzhyk or Trasjanka, respectively) or Russian. In both cases, the phonological system is primarily not Russian. With regard to verbal morphology, 3rd person non-past forms and masculine singular forms are non-Russian. In terms of prepositional phrases, elements from both source languages do occur, with mismatches between the source preposition and the source case morphology.
5.2 . Classification Issues An examination of the structures of Surzhyk and Trasjanka cannot help but invoke the question of their classification: are they instances of heavy code-switching, mixed languages or something else? Mixed languages are defined as prototypically involving a source language that provides morphosyntax (the grammaticizer language) and a different source language that provides the lexicon (the lexifier language) with variation in whether the function words come from the grammaticizer or the lexifier language (see, for example, Bakker and Muysken 1995; for discussion, see Matras 2003).
Matras (2003) refines this categorization to account for a number of problems with the prototype, one of the most noteworthy for my purposes here being that a number of mixed languages do not conform to it (Matras 2003:151-54). (In addition to the structural characteristics, there are sociolinguistic aspects to the prototype, namely, that mixed languages not only serve as all-purpose, full-fledged languages within the community languages, but more specifically that they are stable and systematic, and not spontaneous code-mixes (Ibid:154). As Matras points out, there are cases where the grammaticizer language is used in everyday communications, and so it follows that the mixed language might best be seen as a mixed register (of the grammaticizer language).
Although they are often referred to as mixed languages, Surzhyk and Trasjanka differ from the prototypical definition of a mixed language in that there is no dichotomy between the grammaticizer and the lexifier language. Rather, both languages contribute to the mixed variety along a number of different parameters. This is not surprising, inasmuch as the structures of the languages involved in the mix are so closely related genealogically and so similar typologically. In most instances, one would be hard pressed to identify a grammaticizer language, as the structures were congruent to begin with.
In both Surzhyk and Trasjanka, there are a number of elements that are identical in both source codes, and so could be analyzed as Russian or Ukrainian/Belarusian, respectively. Alternatively, they could be seen as simultaneously being parts of both systems. (See, for example, the discussion of examples (1) and (2) in Section 2.2.) After an examination of the issues surrounding language mixing, Auer and Muhamedova (2005:52) conclude that “bilingual talk cannot be analyzed as a mixing of two monolingual codes,” invoking Sebba’s (1998) model of congruence, which argues that bilingual speakers create congruence to facilitate codeswitching. A central part of this theory is the implicit assertion that bilinguals simultaneously access both codes in the creation and use of such mixed lects.
Notes 1. The present article is part of a larger project on the impact of contact with Russian in Eurasian languages. Research was supported by generous funding from the American Council of Learned Societies, from the American Councils for International Education ACTR/ACCELS Title VIII Research Scholar Award, and the Humanities Division of the University of Chicago. My thinking has been deeply informed and shaped by working closely with Victor over the last few years on the study of contact and in particular by the opportunities I have had to team-teach contact linguistics with him. I have benefited immeasurably from our many discussions. All opinions and errors here are my own.
2. Data for Table 1 come from the following sources: Belarus: Population Census 2009 [belstat.gov.by] (accessed 15 February 2013); Ukraine: population estimate for 2014, in “People and Society: Ukraine;” CIAWorldFactbook [accessed 18 July 2014]; Ethnic groups in Ukraine [ukrcensus.gov.ua]; UkrainianOfficeofStatistics (archived from the original on 23 March 2008 and accessed 17 April 2010; dead link); Russian population: Ob itogax Vserossijskoj perepisi naselenija 2010 goda [http://www.perepis-2010.ru/].
3. Transliteration in (1) has been adapted from the original source to conform to the guidelines of Balkanistica. In (2), Woolhiser presents the data in phonetic transcription, which I have maintained here. Abbreviations used here are: BR = Belarusian; Sur = Surzhyk; R = Russian; Ukr = Ukrainian. For glosses I follow Leipzig Glossing rules: ACC = accusative; DAT= dative; IMPF = imperfective; INF = infinitive; M= masculine; PL = plural; PST = past; REFL = reflexive; SG = singular.
4. Law N 5029-VI of 03 July 2012, revised 06 September 2014, available at [http://zakon2.rada.gov.ua/laws/show/5029-17] (accessed 6 March 2015).
5. Data here are taken primarily from Flier (1998), Vakhtin et al. (2003) and Kent (2012), with the source noted. Flier relies on written texts and Kent and Vakhtin et al. on spoken corpora, which accounts for some of the differences.
6. I have maintained Kent’s original transcription in this excerpt. She does not discuss the phonetics of the forms in this excerpt but does assert that all five verbs are from Russian. Her transcription of rozvalnuva- is unexpected for the Russian orthographic razvolnova- and rasshita- for orthographic rasčita-. Kent provides partial phonetic transcription, which suggests that the prefix roz- is either a typographical error (for the expected Russian raz-) or is in fact Ukrainian.