Practising receptive multilingualism: Estonian–Finnish communication in Tallinn

Abstract

The present article deals with receptive multilingualism (RM) in communication between Estonian salespersons and Finnish customers in Tallinn. A historical background and general description of this communication is provided. It is argued that RM is an established practice in this type of communication and that a nation-state is not necessarily marginalizing communication that deviates from monolingual norms. The notion of mutual intelligibility is critically analysed, and the data confirm that understanding is not exclusively dependent on material similarities between closely related varieties. Accommodation and negotiation are present in both sides, yet strategies employed by Finnish and Estonian interlocutors differ. The distinction between inherent and acquired RM is relevant in Russian–Finnish communication which takes place in the same environment. Finally, further research questions are formulated.

Keywords

Estonian Finnish language contacts receptive multilingualism

1 Introduction

During the past few decades, research in contact linguistics, pragmatics, language acquisition and bilingualism has undergone a significant focus shift: monolingual speech communities and individuals are no more considered as an unmarked case and as a departure point for studies on multilingual speech, and multilingualism is not treated as a peripheral, marginal or marked phenomenon (Franceschini, 2008). Receptive multilingualism (henceforth RM) is a case of multilingualism where the interlocutors use their respective first languages while speaking to each other (Zeevaert & ten Thije, 2007, p. 1). To this definition I would like to add also the possibility of compromise and conscious imitation of the language of interlocutor, as well as creative use of linguistic resources. This addition is necessary because otherwise the understanding of RM would be two narrow (i.e. limited to the use of respective L1s only). Research on RM is a currently developing new field which intercedes with other disciplines, such as contact linguistics, intercultural pragmatics, bilingualism, language acquisition, etc.

The aim of the current study is twofold: (1) a general characterization of Estonian–Finnish communication in Tallinn between Finnish tourists and Estonian salespersons (henceforth EFC); and (2) an outline of relevant research perspectives on the mentioned bilingual communication in the context of the expanding field of RM. My main interests lie in the area of contact linguistics and contact-induced language change and I am going to concentrate on the instances of non-monolingual speech (i.e. forms and items that do not belong to either standard monolingual variety) in EFC. As Riionheimo (2007, p. 155) stresses, so far contact linguistics has mostly dealt with contacts between unrelated or distantly related varieties, while studies on contacts between related varieties are scarce.

Of course, the topic of EFC can also be considered from the viewpoint of SLA, interlanguage, learner varieties, and so on; however, my perspective here is different, that of contact linguistics. One can ask what the nature of utterances produced by Estonians in EFC is and whether they are more Finnish, more Estonian, or anything reminiscent of learner Finnish. I will dedicate some space to the discussion in Section 4. From the contact linguistic perspective, labelling varieties is not a primary concern because the borders are extremely fluid and, strictly speaking, monolingual or monolingual-like Finnish is not a target for the speakers who imitate Finnish, creatively using resources of Estonian and some ideas of what Finnish sounds like.

Estonian and Finnish have a long history of contacts; however, it has to be stressed that, in fact, there are several distinct cases of language contacts as far as sociolinguistic situations, the direction of influence and the results are concerned. At least the following cases should be distinguished: (1) the impact of Estonian on the speech of bilingual Ingrian Finns (Riionheimo, 2007 and references therein); (2) the sporadic/unfocused acquisition of Finnish by Estonians via Finnish TV starting from the late 1970s during the Soviet era; (3) the present EFC situation where Estonians exhibit various degrees of proficiency in Finnish and have different acquisition histories; (4) the bilingual speech of Finnish students studying in Estonian universities (Kataja & Klaas, 2003); and (5) Estonian–Finnish contacts in the immigrant setting (i.e. in Finland or in Estonia). It is possible that (2) is linked to (3) because some speakers working in the service industries may have acquired an ability to understand or have a working command of Finnish first and foremost via TV during the Soviet occupation when contacts with foreigners were limited and controlled by the authorities. Cases other than (2) and (3) remain outside the scope of the present article. Strictly speaking, (2) has prepared the ground for RM and (3) is RM, the main concern of the present article.¹

Mostly, the locus of EFC is interaction over the counter in the Old Town and centre of Tallinn, as well as the port area (for a general description of the sociolinguistic situation in Estonia, see Rannut, 2007; Verschik, 2005). Finnish tourists come to Tallinn to do shopping but also to go to the hairdressers, beauty parlours, optometrists, dentists, SPA, museums, opera, etc. According to an official tourism management website (http://www.infoturism.ee), Finnish tourists made up 49 per cent of all foreign tourists who visited Tallinn in January–November 2008 (http://www.infoturism.ee/upload/Tlni-iiikv_2008.pdf, 07.01.2009). A whole service industry oriented towards Finnish tourists has developed. In the census of 2000, 16.6 per cent of Estonian speakers claimed to have a proficiency in Finnish.

Surprisingly, although EFC in the mentioned settings is an everyday phenomenon known to many in Estonia, linguists and laypeople alike, it has not received any scholarly attention. The neglect may be explained by the fact that the research on contacts between these closely related languages is concentrated in the following areas: traditional SLA approaches to the acquisition of Finnish (including Finnish for the speakers of related languages) and Finnish–Estonian contrastive grammar research (Remes, 1995; Sulkala, 1996), code-switching (CS) and other contact phenomena in the speech of Estonians living in Finland (Praakli, 2007, 2008), Estonian–Finnish bilingualism in children raised bilingually in Finland (Hassinen, 2002), and the impact of Estonian on the Finnish variety spoken by Ingrian Finns, an indigenous minority in Estonia (Riionheimo, 2002, 2007; Savijärvi, 1998).² The named areas of inquiry are either linked to traditional SLA research paradigms that seek to establish the nature of ‘transfer’ in the instructed learning of L2 or deal with language contacts in minority (Ingrian Finns)/immigrant situations (bilingual children and adult Estonian–Finnish bilinguals in Finland). Obviously, EFC differs from what was mentioned here: in many instances, it is neither a language acquisition case in the traditional sense (i.e. instruction and supervision in an institutional setting, usually considered in SLA studies) nor anything even remotely resembling a situation of an (immigrant) minority language, most frequently described in the contact linguistics/bilingualism literature.

Yet, EFC should be acknowledged as a separate case of Estonian–Finnish bilingual (or, in some instances, multilingual) interaction whose primary characteristics are as follows: on the one hand, a striving for effective communication without concern about ‘target forms’ and, on the other, creative and flexible use of linguistic resources. The latter refers mostly to Estonian interlocutors because, for reasons to be discussed further, it is they who are likely to modify their speech; yet, as we shall see, Finnish participants occasionally make use of Estonian-language resources available to them, such as quoting Estonian-language labels, etc.

The article is organized as follows. First, some relevant theoretical and methodological issues of RM will be discussed, such as the relationship of RM to other cases of bilingualism, the concept of mutual intelligibility and its critique, etc. Then historical dimensions of RM, i.e. its role in the period prior to the formation of nation-states and the impact of national language standardization, will be presented with regards to EFC. It will be argued that not all nation-states by definition marginalize RM and multilingualism. After that, based on my data, some important EFC characteristics will be outlined and a hypothesis on the type of the varieties of Finnish used in EFC will be formulated. Here, I draw upon the notion of quasi-Italian proposed by Franceschini (2002). Finally, a brief characterization of Russian speakers communicating to Finns in Tallinn will be provided and hypotheses about their interaction strategies will be suggested.

The EFC data collection is in its initial stage. The data used in this article come from the observation of 40 encounters in Tallinn (Viru Keskus shopping centre and Rotermanni Keskus shopping centre, both situated in the centre of Tallinn) in 2007–2008.

2 Theoretical considerations

The current section deals with the position of RM in the context of linguistic research and with the concepts of mutual intelligibility and semicommunication. It will be argued that structural and material commonalities caused by genetic relatedness cannot alone guarantee understanding and smooth interaction between the speakers of related varieties. It is micro- and macro-sociolinguistic factors, as well as individual factors that, combined with material similarities, are at play here.

2.1 RM in the context of bilingualism research

As Braunmüller (2007, pp. 25, 26) points out, RM is usually not classified as a manifestation of bilingualism in its own right and is often considered as a residual form of a variety in language attrition situations. Still, some better-studied examples such as interlingual communication in Scandinavia demonstrate that this is neither a case of attrition nor of ‘poorly’ acquired L2. RM is not a case of language loss because RM does not replace L1. Moreover, over the past few decades, several bilingualism researchers have advocated the idea that contact-induced language change per se is not indicative of attrition and neither can be equal to it (Bullock & Gerfen, 2004; Bullock & Toribio, 2004; Pavlenko & Jarvis, 2002; Schmid, 2004). In the light of pioneering research in multicompetence (Cook, 1995, 2006, 2007; Pavlenko, 2002), the notions of ‘target language’ and ‘incomplete acquisition’, as well as ‘native speaker’, become increasingly controversial and vague, one of the reasons being that ‘complete’ acquisition is not always the objective for the speakers. Neither is it possible to unambiguously define ‘full competence in L1’ (Cook, 1997). In addition, cognitively, knowledge of each language in a multilingual is different from that in separately considered monolingual speakers of those languages (Cook, 1997; Grosjean, 2001; Jessner, 2008). RM is consistent with the concept of multicompetence: the speakers have at their disposal more resources than a monolingual speaker has; they make use of these resources in a creative way, which does not assume approximation to the monolingual standard. The decision to use RM is conscious; those people using RM realize that in some situations RM can fulfil their communication needs, although RM interaction is not always unproblematic (Braunmüller, 2007, p. 26).

Similarly, it would be counterproductive to describe RM as merely passive command (and, therefore, not as ‘valuable’ as active/ productive command) of a variety (Braunmüller, 2007; Werlen, 2007, pp. 140, 141). The examples presented further in the current article demonstrate that RM requires a certain degree of meta-linguistic awareness (for example, in posing a hypothesis about cross-linguistic equivalences, etc.) and creativity (such as the strategies of L1 modification in order to facilitate understanding by the interlocutor).

2.2 Mutual intelligibility

Being an Estonian linguist, I am often asked by laypeople and by linguists from abroad whether ‘Finns and Estonians can understand each other’. The answer to this question is, in my view, as follows: it depends on which Finns and which Estonians one is talking about. Structural and material similarity in closely related languages and a high number of cognates is usually held as a crucial prerequisite for (mutual) understanding. Indeed, this is important but in itself cannot be the sole explanation for understanding and smooth interaction. Calculation of typological distance between varieties does not provide a satisfactory answer to the question.

Mutual intelligibility is a highly controversial matter. For instance, Van Bezooijen and Gooskens (2005) established, after a series of tests on intelligibility of Afrikaans and Frisian by native speakers of Dutch, that Afrikaans speech and written text was more intelligible to Dutch speakers than Frisian. The result was explained by a higher number of cognates between Afrikaans and Dutch. Thus, geographical proximity and sociocultural similarities between the speakers of Dutch and Frisian did not outweigh the differences between the language systems. On the other hand, Zeevaert (2007, p. 107) questions the applicability of many intelligibility tests, claiming that it is not clear whether the context and other factors have influenced intelligibility and that the tests are rather measuring individuals’ performance. As for inter-Scandinavian communication, Doetjes (2007) is sceptical about tests conducted so far and points out that no study has succeeded in providing an unambiguous answer to the question of how well the speakers of Scandinavian languages understand each other.

Karam (2000, p. 123) maintains that if forms deviating from the acceptable norms are recognizable by the community as remaining within the area of acceptable variation, these forms are intelligible. He also claims that, often at first encounter, the speakers do not understand each other but intelligibility gradually increases once familiarity with each other’s varieties has increased. Referring to the literature, Karam (2000, pp. 126, 127) mentions ‘two-day dialects’, ‘one-week dialects’, etc. I concur with the proposition that exposure to and growing familiarity with each other’s varieties facilitates the development of techniques for the establishment of cross-linguistic equivalences, or, to render it differently, for creating the bridge in the speaker’s mind between the related varieties. In other words, increasing familiarity with a close variety leads to the formation of what is called spontaneous grammar in L3 acquisition (or TLA) literature. At the very beginning, this grammar draws more on the previously acquired language system(s) and is continuously revised and developed during the learning process (Jessner, 2008, p. 24; Meissner, 2004 discusses this in the context of TLA where L3 is related to L1 or L2; but if L1 and L2 are related, the concept of spontaneous grammar can be applied here as well). I will return to the role of exposure to and experience with a closely related variety later in the present section.

Further, Karam (2000, p. 127) argues that when mutual intelligibility is lacking, the result would be the choice of ‘the same language code by the communities involved’, i.e. either a lingua franca or one of the monolingual varieties in question (in the latter case, it will be L1 for one and L2 for the other speaker). This statement appears to me too categorical. Several counter-arguments against this position can be brought up here.

First, it is not clear what ‘the same language code’ means. The data on Estonian–Finnish communication provide instances of items that cannot be described as belonging to either monolingual variety (see also Clyne, 2003, 164ff. on compromise forms and bilingual homophones). Riionheimo (2007, p. 155) refers to the notion of compromise forms (Clyne, 1987, p. 760) and discusses numerous examples where verb forms produced by Ingrian Finns in Estonia are neither Finnish nor Estonian from the monolingual point of view (Riionheimo, 2007, pp. 159–161). Somewhat similar examples are available from EFC. Consider example (1a). A young saleswoman in a shopping centre answers a client’s question. The Estonian elements are in italics, the Finnish is in bold, ambivalences (i.e. belonging to both varieties) are underlined and un-decidable/compromise items have no marking.

(1a)

See on siella toise-lla korruse- lla

it be:3 there second-ADES floor-ADES

‘it is there on the second floor’

Let us compare (1a) to monolingual Estonian (1b) and monolingual Finnish (1c):

(1b) Estonian

See on seal teise-l korruse-l

it be:3 there second-ADES floor-ADES

(1c) Finnish

Se on tuolla toise-ssa kerrokse-ssa

it be:3 there second-INES floor-INES

Certain items are ambiguous (on, ‘is’) or almost ambiguous (see/se, ‘it, this’) because they belong to both monolingual varieties. Others exhibit compromise [+ Finnish lexeme], [– vowel harmony], as in siella (‘there’). However, even if siellä exists in Finnish, tuolla (‘there’) would be a more natural choice. Apparently, the compromise siella was modelled on the Estonian seal (‘there’). As for the rest, toisella (‘on the second’) is probably the only item that can be classified as Finnish in form but, as one can observe from (1c), the case semantics are different in this instance and Finnish uses internal local case Inessive with the marker –ssa (literally, ‘in the second floor’) as opposed to the Estonian external local case Adessive. At the same time, the speaker ‘got right’ the Finnish Adessive marker -lla, the cognate of the respective Estonian marker -l. The last item, korrusella (‘floor-ADES’), was produced according to the model: Estonian stem + Finnish inflectional morphology (the reverse strategy, that is, Finnish stem + Estonian inflectional morphology, is possible as well, as we shall see later). The mentioned strategy is productive in EFC communication and, although not 100 per cent unproblematic, often helps to get the meaning across. Thus, two items in (1a) are ambiguous and can be classified as belonging to both languages; one is Finnish but semantically copied from Estonian, and the other a compromise. As one of my Estonian informants had it, ‘we speak Finnish with Estonian words’. The client had no problems understanding this utterance and was thankful for the helpful information.

Second, it is not entirely clear what the ‘same’ code means, since the issue of what constitutes a particular variety and what should be rejected may be highly debated within a speech community; moreover, opinions on the ‘sameness’ of insiders vs. outsiders or of laypeople vs. linguists are not necessarily identical.

Third, it follows from Karam’s (2000, p. 127) claim that varieties are static and rigidly delimited, and that speech communities have clear borders and membership status, which, in the light of what is known from modern contact linguistics and multilingual speech research, is arguable. This brings forward the notion of monolingual bias, i.e. a tacit assumption that a monolingual speaker can serve as a yardstick in the study of multilingualism, which has been dominating linguistic theories and multilingualism research in particular (see critique in Auer, 2007; Backus, 1999; Cook, 1997; Ellis, 2006; Jessner, 2008). The possibility of accommodation between the speakers and negotiation of form and meaning are being ignored by Karam (2000, p. 127); yet, Thomason (1997, 2001) rightly lists negotiation among the mechanisms of contact-induced language change. Negotiation, as used by Thomason, is not an explicit discussion on what language should be used in the interaction but rather modifications based on the speakers’ hypothesis concerning what can and cannot be understood by the interlocutor. The idea that speakers create forms and items non-existent in monolingual standard varieties, and that such modifications are not marginal but rather central in some types of multilingual communication, is consistent with the concept of languages as dynamic systems (Herdina & Jessner, 2002; Jessner, 2008). It is likely that in cases where the choice of variety is not self-evident, the role of negotiation may be especially relevant. I assume that negotiation as a mechanism of contact-induced language change plays a prominent role in RM; this claim has to be substantiated by further data collection and analysis.

Fourth, even speakers of non-related varieties would eventually find a compromise choice that would be different from either monolingual variety, provided they have some idea about each others’ varieties. My study on Russian–Estonian communication in Tallinn has revealed that sometimes speakers opt rather for compromise, in-between forms and various patterns of bilingual speech even if they are proficient in both monolingual varieties (Verschik, 2008, pp. 181–188).

Thus, we should reconsider the notion of mutual intelligibility because it is speakers who understand or do not understand each other and not the language systems (the same fallacy appears in formally oriented approaches to results of and hypothetical constraints on the language contacts, based exclusively on the language systems; see Backus, 2005 on the sociolinguistic factors that are the ultimate causes of contact-induced language change). Even while learning a closely related variety, learners often have difficulties in automatically recognizing correspondences and similarities and need instruction (see the discussion in Hufeisen & Marx, 2007, along with the concept of the Seven Sieves). From a different perspective, that of functional pragmatics, Rehbein (2006) explains understanding as a phenomenon that is linked to discourse rather than to exclusively structural characteristics of the varieties in question (in his model, understanding is viewed as a gradient process; see also Rehbein & Kameyama, 2003). To quote earlier studies, Haugen (1966) concludes that the correlation between linguistic criteria and understanding is often rather low. The idea of understanding and misunderstanding is in many instances a cultural phenomenon that is constructed (Hinnenkamp, 2000). In sum, it is speakers who can or cannot draw parallels between cognates.

As far as EFC is concerned, no tests measuring intelligibility (similar to those by Van Bezooijen & Gooskens, 2005) have been conducted. Still, it is a sort of common knowledge that many Estonian speakers from Tallinn are able to understand Finnish, while those from other regions of the country are not. If one is convinced that the crux of the matter is material and structural similarity between the varieties, then one has to inevitably conclude that the variety of Estonian as spoken in Tallinn is closer to Finnish than other varieties of Estonian. However, this is very far from the truth and the explanation lies elsewhere, namely in the exposure to and experience of Finnish (the habit of watching Finnish TV and ever-growing needs for communication over the counter with Finnish tourists). Thus, sociocultural, sociolinguistics and individual factors are responsible for the fact that some Estonian speakers would (or would not) exploit cognates and structural similarities between the two languages for the purposes of EFC. Of course, rather often, Estonians from the South (where Finnish TV is not available and where Finns are not as frequent visitors as in Tallinn) tend to bluntly deny the assumption that they can understand Finnish (unless they have studied the language), whereas an answer such as ‘I don’t know because I have not tried it’ would be more accurate.

2.3 Semicommunication

Another controversial notion is that of semicommunication, introduced by Haugen (1966, p. 153) and employed when communication between speakers of closely related varieties is discussed. According to Zeevaert (2007, p. 105), RM refers to unrelated or remotely related languages, whereas semicommunication is a special sub-type of RM between closely related languages. Zeevaert (2007, pp. 106, 107) believes that the communicative conditions for RM and semicommunication are different, and, moreover, the very concept of ‘language’ is different. In his view, RM is a case of communication between speakers of Abstand languages and also between speakers of Ausbau languages that exhibit little distance (see Kloss, 1967 for the concepts of Abstand, that is, varieties distinct by virtue, and Ausbau, varieties which are subject to language planning efforts pursuing distinctiveness from a closely related variety). Otherwise, the term semicommunication is reserved for contact between Ausbau languages but not between dialects within a dialect continuum.

Still, the proposed distinction between RM and semicommunication does not seem very clear. This is not to deny possible differences in communication patterns and structural characteristics of multilingual speech in the case of related vs. non-related varieties; nonetheless, these differences do not boil down to the mere fact of genetic relatedness or lack thereof, or to the vague distinction between Ausbau and dialects, as it is implied. Doetjes (2007) indicates that the term ‘semicommunication’ as applied to interlingual communication in Scandinavia is not used in a pejorative sense like ‘semilingualism’ is in some contexts; nevertheless, the term itself appears monolingually biased for the reason that it implies the existence of ‘full’ communication; still, we do not know what ‘full’ communication is and whether such a thing exists (similarly, Martin-Jones and Romaine (1986) argue against the concept of ‘semilingualism’ or ‘double semilingualism’ because it is impossible to unambiguously define ‘complete monolingualism’; see also Braunmüller, 2007, p. 26 on ‘semicommunication’ as a misleading term). The difference between approach to Ausbau languages and closely related dialects, as suggested by Zeevaert (2007), is not entirely clear either. Sometimes two closely related varieties are treated as ‘languages’ (i.e. are codified and standardized) without any Ausbau efforts on the part of language planners. Finnish and Estonian is but one example; moreover, Estonian language planning consciously used Finnish models in the interwar period: as Raag (1999, p. 30) points out, a considerable number of importations, calques and creations were modelled on Finnish. That would be a process opposite to Ausbau.

While it is true that it may be easier to develop understanding techniques and communicative skills in a closely related variety in a shorter period of time, structural similarity and a high number of cognates is not the only necessary prerequisite for successful communication. Also, a speaker exposed to an L3 that is related to his/her L2 can become a skilful communicator in that L3. A situation where such a speaker has to, so to say, discover the links between L2 and L3 (or between an already known and a new, closely related variety) fosters the development of linguistic awareness and creativity (see Franceschini, 2000 on the re-activation of Italian as L3 for German speakers in Switzerland). Instead of the terms ‘semicommunication’, ‘Ausbau’, dialect continuum, etc., I find useful the distinction between inherent RM (i.e. by the virtue of close genetic relatedness between varieties) and acquired RM (i.e. a communicative situation where L3 is closely related to L2 and is perceived by the speakers through the prism of the latter) (Reinelt, 2008). Of course, ‘inherent’ does not mean that all speakers of related varieties are automatically by definition RM speakers (see the above discussion on exposure) but rather that, under appropriate sociolinguistic conditions, they can learn to exploit the similarities between the varieties in a relatively short space of time. I will return to this distinction in the section on Russian–Finnish communication in Tallinn.

2.4 Directionality of understanding

Another issue that cannot be considered here at length but, nevertheless, has to be mentioned, is the directionality of understanding. For instance, Van Bezooijen, Gooskens, Kürschner, and Schüppert (2008) have established through a series of experiments that Swedish-speaking children understand Danish better than vice versa. Although no similar studies have been conducted for Finnish and Estonian and, therefore, no references can be provided, it is often held by laypeople and by linguists in Estonia that Estonians tend to understand Finnish better than Finns understand Estonian. Approached from a purely typological and structural point of view, the issue of directionality can be explained in the following way: compared to Finnish, Estonian has undergone more contraction and change, for instance the final vowel of the stem is lost in the nominative case, cf. the Estonian mets (‘wood’) and the Finnish metsä (‘id.’). On the whole, Estonian has moved further towards fusional morphology, for example possessive suffixes have become extinct in Estonian but not in Finnish: the Estonian minu isa-lt (I:GEN father-ABL, ‘from my father’) corresponds to the Finnish isä-ltä-ni (father-ABL-1SG) (on the development of Estonian from agglutinating towards the fusional type, see Sutrop, 1997). While all this is true, it is not the only explanation.

One has to keep in mind that the degree of intelligibility in either direction is influenced by extra-linguistic factors. Estonians in Tallinn tend to watch Finnish TV, whereas Estonian-language TV broadcasts are not available in Helsinki; it is Finnish tourists who come to Tallinn as customers and visitors and not vice versa. It is true that traffic between Tallinn and Helsinki has constantly been on the increase and that many Estonians work in Helsinki; however, this is a totally different sociolinguistic situation because foreign employees are expected to master the language of the country and cannot enjoy the same privileged treatment as foreign clients/tourists do. Thus, sociolinguistic factors cannot be separated from the discussion on intelligibility and its directionality. It would be highly instructive if research on understanding between the speakers of Latvian and Lithuanian existed, for in this case the status of the languages and communities involved is more or less equal.

To summarize, my position on semicommunication, mutual intelligibility and the directionality of understanding is as follows. Structural and material similarities (due to genetic affinity or to other factors such as intensive language contacts, etc.) provide an important advantage for a potential communicator and create a basis for spontaneous grammar (Jessner, 2008; Meissner, 2004). But if the communicator in question has not been exposed to a closely related variety and, therefore, has no experience in the formulating of a hypothesis on what is similar and what is not (in other words, if s/he lacks experience in the establishment of cross-linguistic equivalence), then all the similarities in question simply do not function for this speaker (recall Hufeisen & Marx, 2007 on cases of learners’ inability to discover correspondences without explicit instructions). Awareness and skills in the establishment of equivalences (for instance, the Estonian ee yields the Finnish ie, as in the Estonian tee [‘road’] vs. the Finnish [tie]), even at the risk of overgeneralization, is all-important for RM. Such awareness can be developed either through instruction in a formalized setting (i.e. language courses, etc.), through self-instruction or, probably, through unfocused acquisition (i.e. the speakers unknowingly know another variety to a certain extent – see Franceschini, 2003; elsewhere, Franceschini, 2008 lists unfocused language acquisition among under-researched topics in linguistics).

Analogically to the model representing the weight of different factors in contact-induced language change, proposed by Backus (2005), it could be claimed that individual factors (highly developed linguistic awareness, language aptitude, positive attitudes, experience in studying languages, exposure to a given variety) and sociolinguistic factors (the prestige of the languages involved, the language planning climate, linguistic purism or the lack thereof, etc.) are ultimate causes that make speakers exploit typological and genetic similarities for the purpose of RM communication. If these ultimate causes are not at work for a given speaker, the potential of cognates and similarities remains merely theoretical.

3 Historical context

It has been demonstrated in the literature that RM is not a new type of bilingual communication, characteristic of the present, but a phenomenon known throughout European history (Braunmüller, 2007; Rindler-Schjerve & Vetter, 2007). A study by Braunmüller (2007) concentrates on historical aspects of RM in the Scandinavian region. He emphasizes that prior to the emergence of nation-states and language nationalism movements, RM was not unusual. The rise of nation-states and, subsequently, of ideologies favouring monolingualism and a homogenous standard language, in his view, have marginalized ‘non-native’ varieties, flexibility and communicative/understanding goals in favour of the ‘national’ language.

This critique is justified in many cases; nevertheless, there are substantial differences between nation-states in Europe as far as the monolingual ideal is concerned. Although formulation and cultivation of the standard was considered a primary task in many countries, the ‘standard building’ (analogically to nation building) did not occur in identical circumstances throughout Europe. For instance, certain states, which emerged from former multinational empires such as the Russian or Austro-Hungarian Empire, provide a relevant example here. In Estonia, at the beginning of the 20th century, there existed the concept of the ‘three local languages’ (Estonian, German and Russian). Proficiency in the three local languages was an advantage for a person whose profession required communication with clients. This did not necessitate, however, a full command of the languages but rather an ability to understand and communicate in rather narrow domains (Ariste, 1981, pp. 33, 34). It is highly possible that some kind of RM communication took place at the very period of time when Estonian intellectuals were involved in the corpus and status planning of Modern Standard Estonian. To the best of my knowledge, no forces in society considered ‘the three local languages’ inferior to monolingualism, and the phenomenon existed further on in the Republic of Estonia up to the first Soviet occupation (1940–1941). In contrast to the ‘old’ nations, Baltic states officially acknowledged the existence of ethnic and linguistic minorities, i.e. equality was not understood as homogeneity: citizens of a given country can be speakers of different languages and receive secondary education (supported by the state) in their mother tongues. In Estonia and Latvia prior to the Soviet occupation in 1940, all schoolchildren, majority and minority alike, studied at least two foreign languages in school.

Speaking about attitudes towards the Finnish, in particular, it is relevant that the awareness of close genetic affinity (and, therefore, of structural and material commonalities between Finnish and Estonian) increased exactly during the period of and hand in hand with the national awakening and birth of the Estonian (proto-)elite. In the first half of the 19th century, Baltic German intellectuals and, somewhat later, the first Estonian scholars came to an understanding that Estonian was related to Finnish. In this spirit, a new grammar of the Estonian language created by Eduard Ahrens appeared in 1843 (with the enlarged edition in 1853) and served as a watershed between the previous tradition of grammar description modelled on the German language and the new one, oriented towards Finnish (for more on so-called Estophiles and Ahrens’s grammar, see Erelt, 2003, p. 380). The fact that Estonian and Finnish are ‘sisters’ is explicitly stated in the grammar and creates a new theoretical departure point. The principles of Modern Standard Estonian orthography are based on those of Finnish orthography (for instance, in the marking of a long vowel or consonant by doubling it), as well as on the introduction of the Essive case in Finland in the late 19th century – these are but some examples of the awareness of the close genetic affinity (Raag, 1999, p. 27 labels the tendency to see similarities as a ‘Fennic leaning’ of the developing Estonian intelligentsia).

Speaking of a broader intellectual context of the time, Elias Lönnrot (1802–1884), a renowned figure of the Finnish national awakening, a physician, philologist, editor of dictionaries, publisher of magazines and collector of folk oral poetry, based on which he composed the Finnish epos ‘Kalevala’, visited Estonia in 1844 and established personal contacts with the Estonian literati. His visit marked the birth of the so-called ‘Finnish bridge’. Further, the connection was fostered through contacts between Estonian and Finnish poets, writers, scholars and intellectuals. Significantly, during the period of Russification in the Baltic provinces of the Russian Empire (from the beginning of the 1880s to 1905) many Estonian intellectuals studied at Finnish universities and returned with new models for nation building and language planning.

Much later, during the Soviet occupation (1940–1941, 1944–1991), the acknowledgment of historical, linguistic and cultural affinity between the two peoples became one of the building elements of oppositional identity. The tacit (or, at times, not so tacit) support of Finnish intellectuals mattered a lot (on contacts between Estonian and Finnish academics during the Soviet occupation, see Kallio, 2004).

From the 1970s, the inhabitants of Tallinn and the northern coast of Estonia were able to watch Finnish TV programmes. Needless to say, under Soviet occupation this was semi-illegal and the authorities did their best to fight ‘the foreign influences’ and ‘hostile ideologies’, but with no success. A whole generation of Estonians in Tallinn grew up watching Finnish TV shows, news and films, acquiring at least some understanding skills and having access to the Western world. The impact of Finnish TV cannot be overestimated. For instance, at a time when face-to-face contact with Finns was extremely scarce, Loog (1991) attests to Finnish lexical items in school slang in Tallinn, the source being Finnish TV programmes. It was not only a source of information (in contrast to censured, biased and carefully selected news in the Soviet media) but also an excellent point of comparison for Estonians to reflect on Estonia’s identity. Figuratively speaking, watching Finnish TV was a political statement, symbolically restoring Estonia’s belonging among European nations and (re)constructing the connection between two peoples speaking closely related varieties (on the impact of Finnish TV on the assertion of Estonian identity during the Soviet occupation, see Lauristin & Vihalemm, 1997, p. 76).

4 EFC at a glance: Finnish and quasi-Finnish

Among the Estonian speakers who work with Finnish customers on an everyday basis, there are those who have had at least some instruction in Finnish. Unfortunately, to the best of my knowledge, there are no statistics on how many, in fact, have studied Finnish on language courses. It is likely that the majority of Estonian salespersons lack any formal instruction in Finnish; they rather draw on the input they receive from their Finnish clients and/or on what they may have unknowingly acquired via Finnish TV. In communication over the counter, one can rely on a substantial number of additional clues: the goods are on display, and labels with the names of products and prices are provided. Non-verbal communication may also be helpful.

People are working under the same roof whose degree of proficiency in Finnish differs radically. Most of my informants (i.e. Estonian salespersons) told me that they have never studied Finnish. They have some experience of Finnish TV and are exposed to Finnish they hear from their clients. For example, at a hairdresser’s near Rotermanni Keskus where I collected some data, one of the hairdressers spoke so well that her speech could be qualified as fluent Finnish. Of course, language labels are misleading and, as stated above, linguistic items cannot always be ascribed to a definite monolingual variety, but there was no doubt in the given case that the language spoken was Finnish. At the same time, her colleague, in whose speech I was mostly interested, admitted that she had had no instruction in Finnish and spoke ‘bad Finnish’. Still, my observations confirmed that she was a very efficient communicator and, in most cases, succeeded in getting her meaning across. For sure, she has acquired some, no matter how vague, idea about ‘how Finnish functions’, i.e. she is aware of certain case endings and sound correspondences as well. In other words, she has developed a kind of what may be called contrastive awareness. For instance, she has derived the ‘rule’ that in Estonian the final vowel of nominal stems has disappeared in many declension types in the nominative (apokope) and is present in the genitive, while it has, in most cases, been retained in Finnish. Thus, she ‘reconstructs’ what she thinks to be the Finnish nominative by adding a stem formant (SF), sometimes getting ‘target’ forms and sometimes failing to do so. In (2a), she recommends her client blow-dry her hair in a certain manner because the client’s face is small and thin:

(2a)

pikk- a peen- i kasvo-t

long-SF small-SF face-PL

‘long small face’

Compare to Finnish in (2b) and Estonian in (2c):

(2b) Finnish

pitkä-t piene-t kasvo-t

long-PL small-PL face-PL

‘id.’

(2c) Estonian

pikk väike nägu

long small face

‘id.’

The Finnish kasvot (‘face’) is a pluralia tantum noun but the speaker does not realize it (in Estonian, it is a regular noun). Hence, the number agreement is to be expected as in (2b). What is essential here is the transformation rule concerning the Finnish nominative the speaker has derived: ‘Estonian nominative + stem formant = Finnish nominative’ or ‘Estonian stem + i = Finnish nominative’. In this way, she obtains the compromise form pikka (Estonian nominative pikk: genitive pika), cf. Finnish pitkä (‘long’, nominative). As for peeni, which stands for Finnish pieni (‘small’), it can be analysed in two ways. One option would be to view it as a derivation (i.e. Estonian nominative + stem formant) from the Estonian peen (genitive peene, ‘fine, delicate, refined’) that is a cognate of the Finnish pieni (‘small’). Thus, she constructs peeni which is comparable to the Estonian peen (genitive peene, ‘thin’). Noteworthy, in the two languages, the cognates have different meanings. In Estonian, there also exists peenike (‘thin, narrow’), derived from the same stem. The difference is probably not as important because in this context both ‘small’ and ‘thin’ make sense. The second option is to consider peeni not as a ‘derivation’ of Estonian but as a Finnish item minus the diphthong. According to my observations, the speaker has not (yet?) arrived at a complete understanding of the correspondences between Estonian long vowels and Finnish diphthongs. In any case, being present during the interaction, I got the impression that (2a) was understood by the client. Despite her ‘bad’ Finnish, she is reasonably fluent and has no major communication problems.

Since society and the mainstream language pedagogy understand languages as standard monolingual varieties, speakers such as the hairdresser in (2a) would indeed describe their own proficiency in Finnish as ‘poor’, although, as we have seen, they have elaborated a sort of RM strategy that draws on deliberate approximation, negotiation and accommodation. Thus, RM cannot be equalled to merely passive knowledge of a variety. That is why I added the option of accommodation and conscious manipulation to the definition by Zeevaert and ten Thije (2007, p. 1), quoted in the introduction: while speaking to each other, the interlocutors use their respective first languages, modifying them when necessary (my emphasis).

Analogically to the notion of quasi-Italian, introduced by Franceschini (2002) for the characterization of Italian speech among some German speakers working as salespersons in Switzerland, I call this variety (or, more precisely, a cluster of such varieties) quasi-Finnish. The notion of quasi-Finnish should not be understood pejoratively or considered as of less value than monolingual Finnish, learner Finnish, etc. Quasi-Finnish, like quasi-Italian in Franceschini (2002), is a conscious creative manipulation of language resources. Material similarities between the varieties in question are recognized by speakers of Estonian. They are perfectly conscious that what they produce is non-target forms but they are not concerned with it. As they achieve their communication goals, quasi-Finnish is as valuable as any other communication strategy.

Of course, one can say that this is acquisition of Finnish of a sort. Still, it would be wise for methodological reasons to distinguish speakers who have had instruction in Finnish from speakers who have not received any instruction in Finnish and consciously experiment with the linguistic resources available to them.

It remains to be seen whether quasi-Italian and quasi-Finnish have anything in common in the sense of structural features preferred, pragmatic strategies used, etc., but both definitely share at least one trait: in both cases, the speakers, without any deliberate instruction in the respective languages, have ‘picked up’ some words, expressions, patterns and elaborated strategies for their effective use.

There is still another group of Estonian speakers who may have mastered some Finnish from textbooks, courses, etc., but who, for some reason, stopped learning (probably because their degree of proficiency was just enough for the needs of their occupation). However, what such speakers produce is, in fact, very similar to (2a). For instance, after I had asked her explicitly, the speaker in example (1a) told me that she knew Finnish; still, even if she thinks she is speaking Finnish, her speech looks very much like quasi-Finnish, i.e. there are compromise forms of stems, omission of vowel harmony in Finnish lexical items, Estonian stems with Finnish inflectional morphology added, and copies of Estonian case semantics onto Finnish forms.

Thus, Estonian speakers use at least two kinds of varieties that are not fully crystallized and have fuzzy boundaries: (1) varieties that are more or less approximating monolingual Finnish (including learner Finnish); (2) varieties that are the result of a conscious compromise between Estonian (L1) and Finnish (L2 or L3 or maybe even L4) that heavily draws upon cognates and the establishment of interlingual equivalences. Even those with some instruction in Finnish may think they are speaking Finnish but, in fact, they are producing something similar to (2).

The similarity between quasi-Finnish and learner Finnish would be a relevant topic for future research. The departure points for the speakers are different, even if what they produce is indeed similar. This opens us up to the perspective of a comparative research of different Estonian–Finnish language contact situations.

5 Transformation rules, negotiation and accommodation

In this section, I will at first concentrate on the ‘Estonian side’ of EFC. At this stage, my data contain more information on the speech of Estonians. Given the situations where EFC takes place, this is to be expected: we know that Estonians are more likely to have some experience in Finnish than vice versa and that they are more motivated to accommodate their clients. Still, it will be shown that there are some strategies of accommodation on the ‘Finnish side’ as well, albeit of a different nature.

RM implies that understanding and communication are more important than ‘target’ forms. The importance of ‘negotiation’ was stressed earlier. In the previous section, it was demonstrated that speakers have a kind of contrastive awareness and derive transformation rules, i.e. what in their L1 corresponds to L2 or, at least, what can be understood. From the point of view of modern contact linguistics, a bilingual speaker establishes cross-linguistic equivalences (or, in this case, transformation rules) that are subjective in their nature (Johanson, 2002, p. 57). Whether we deal with related or non-related varieties, it is likely that in any language contact situation the establishment of equivalences is fundamental (Heine & Kuteva [2005, p. 219] rightly observe that equivalence is a central notion in most of the works in contact linguistics). RM can be viewed as a language contact situation of a particular type and the above-mentioned claim applies here as well. It is a relevant research question as to how exactly such equivalences are established by the speakers/hearers. To put it figuratively, I refer to the words of my informant in my earlier study of Yiddish in Estonia. To the question of whether she understands or speaks other varieties of Yiddish, the lady fluent in Estonian, Yiddish, German and Russian answers confidently (Verschik, 1999, p. 55):

(3)

Nu, vos iz, avade ken ix redn anderš. Ven men zogt o onštot a, farvos ken ix eix nit zogn o onštot a?

‘So what, sure I can speak other way. When one says o instead of a, why couldn’t I say o instead of a as well?’

We have seen that Estonian speakers tend to elaborate certain formulas of accommodation. While more data is needed in order to provide a systematic overview of these formulas, what follows is by no means a complete list. These are strategies on the ‘Estonian side’ of EFC:

- Estonian stems + Finnish inflectional morphology;

- Finnish stems + Estonian inflectional morphology;

- Finnish lexical items [– vowel harmony];

- the establishment of sound correspondences and application thereof to Estonian lexical items (for instance, how to derive the Finnish nominative or to change certain Estonian long vowels into diphthongs in order to arrive at Finnish cognates);

- CS, probably including all three types proposed by Muysken (1995, 2000): insertion, alternation, congruent lexicalization. Since these are closely related varieties, the latter option is highly likely. Theoretically speaking, Estonian stems combined with Finnish inflectional morphology (or vice versa) fit under the heading of CS as well (i.e. content morphemes from one and system morphemes from the other variety). Still, CS is broader than just the mentioned phenomena. CS between closely related varieties and in EFC in particular requires a separate study and cannot be considered within the limits of the present article.

An example of an Estonian stem with Finnish markers was given in (2a). For Finnish stems with Estonian inflectional morphology, consider (4a) where a hairdresser explains that a client has a reservation:

(4a)

on tila -tud

be:3 order-PAST PRTC PASS

‘is ordered’

Compare with Finnish in (5b) and Estonian in (3c):

(4b) Finnish

on tila-ttu

be:3 order-PAST PRTC PASS

(4c) Estonian

on telli-tud

be:3 order-PAST PRTC PASS

Some Finnish lexical items and fixed expressions come as ready-made packages but their phonic rendition may be a compromise, such as the absence of vowel harmony (in Finnish, if the first syllable has a front vowel, the back vowel cannot occur in the following syllables). Compared to Finnish, Estonian has lost vowel harmony. It is likely that vowel harmony is omitted for the reason that, for an Estonian speaker, it is difficult to articulate. Thus, the Finnish ole hyvä (‘please’, sg.) becomes as in (5):

(5)

ole hüva

be good

‘please’

Monolingual Estonian does have hüva but with a different meaning – ‘fine, let it be’, and it is never collocated with ole (‘be’). However, the existence of the Estonian cognate and its different semantics is not relevant here. The speakers use (4) as the monolingual Finnish ole hyvä would be used. Another instance of vowel harmony is siella (‘there’) in (1a).

An important component in the accommodation rules is the correspondence between Estonian and Finnish sound combinations, some of which have been mentioned in earlier sections. From my data, I can conclude that different Estonian speakers exhibit different ‘knowledge’ of these correspondences: some are not aware of them and some overuse them. The following example (6) shows how an Estonian-speaking salesperson (S) in a cafeteria tries to derive the Finnish word for ‘cream’. She serves coffee and wants to know whether the client (C) would like some cream:

(6)

Kas kuori myös ?

Q ? also

‘(would you like) cream as well?’

This was a clear attempt to derive the Finnish word for ‘cream’ on the basis of the Estonian koor (‘id.’) and the correspondence between the Estonian oo and the Finnish uo (cf. the Estonian pool, ‘half’, nool, ‘arrow’ which are cognates of the Finnish puoli and nuoli respectively). In this particular instance, the relation is more complicated. Behind the Estonian koor, there stand two homonyms: (1) ‘cream’ and (2) ‘bark (of a tree), shell’. The application of transformation rules yields kuori, which does exist in Finnish but only in terms of the second meaning. For ‘cream’, Finnish has a completely different item, kerma:

(7)

Estonian Finnish

koor (‘cream’, kerma)

koor (‘bark, shell’, kuori)

In this particular context, the use of kuori appeared counterintuitive to the Finnish client. The problem was solved when the saleswoman helplessly looked at me and asked whether I knew the Finnish word for koor (‘cream’).

During the initial encounter, the establishment of correspondences needs some effort on the part of the speaker/hearer. My data on EFC provide some examples of explicit ‘translation’, as in (8). The interaction took place between the above-mentioned hairdresser (H) and her client (C).

(8)

C: Terve, halua-n leikkaus-ta. Kauan-ko kestä-ä?

hi want-1SG cut-PART long-INTER last-3SG

‘Hi, I would like a haircut. How long does it take?’

H: Pool tunt-i

half hour-PART

‘half an hour’

C: Pool tunt-i, aa, puoli tunti-a

half hour-PART o half hour-PART

‘half an hour, o, half an hour’

It is interesting to consider why the Finnish client uses terve (‘hello’), a greeting that in Finland is associated mostly with male speakers from certain areas. It may be an attempt at accommodation, i.e. something similar to the Estonian tere or tervist (‘hello’) which are rather universal greetings.

The hairdresser has no problems in understanding her Finnish interlocutor. In order to construct her answer, she makes some adjustment in pronunciation (cf. the Estonian tund, ‘hour’ and the Finnish tunti, ‘id.’). The form tunti can be analysed as the Estonian partitive with adjusted pronunciation (cf. the Estonian tundi, ‘hour’, PART); an alternative analysis would be the application of the formula ‘Estonian nominative + stem formant’: tund + i + phonetic rendition close to Finnish. However, she does not apply this formula to pool (‘half’), hence there is no compromise form like pooli and, as we already know, she has not arrived at the correspondences between diphthongs (recall peeni in (2a)), hence there is no puoli (the Finnish form). Apparently, the client requires some time to derive the meaning of the hairdresser’s utterance. From the Finnish point of view, the partitive marker is absent but this is probably not critical for understanding. The client repeats the compromise utterance and then ‘translates’ it into Finnish. Excerpt (6) demonstrates two things: first, that the derivation of equivalency rules may be explicit and, second, that, although it is Estonians who try to speak Finnish or quasi-Finnish, both sides derive transformation rules (i.e. RM efforts are not completely asymmetrical). At this stage, it seems that Estonians derive these rules in order to produce Finnish or quasi-Finnish, and Finns, as in (6), in order to understand quasi-Finnish. It remains to be seen whether this hypothesis is confirmed by more empirical research.

This is a relevant point which leads us to the directionality of accommodation in EFC. Although general sociolinguistic factors are important for the understanding of bilingual communication, micro-level data are needed in order to explain particular examples. While it is true that Finnish is a highly prestigious language and it is Estonian salespersons, not Finnish tourists, who are likely to accommodate and make a step in the direction of Finnish, it would be wrong to claim that Finns always speak monolingual Finnish. While some of them definitely do this, there are individual factors at play. I have shown elsewhere (Verschik, 2008, pp. 181–188) that, under certain circumstances, Russian speakers do not feel that it is ‘appropriate’ to use their L1 because this would require their Estonian to do all the job on their own, so to speak.

As Finns are less likely to have any command in Estonian than Estonians in Finnish, they mostly rely on different strategies of accommodation. This can be considered typical of tourist communication when tourists know some dozen words and expressions from the language of the country. A preliminary list is as follows:

- the use of Estonian greetings – words like ‘thank you’, ‘please’, etc. (in other words, insertional CS of certain lexical items);

- reading aloud signs in Estonian when asking for particular items or repeating after Estonian salespersons (this may be considered under the heading of CS as well);

- the fact that Finnish speakers are probably aware that there exist a considerable number of cognates (i.e. awareness of material and structural similarity) on which they can rely.

Example (9) contains both Estonian greetings used by Finns and the quoting of Estonian-language labels. The interaction took place in a cafeteria at Viru Keskus, a shopping centre frequented by Finns. The client, a middle-aged Finnish male, places his order while looking at the list of beverages and prices:

(9)

Tere. Üks/yks’ väike kohv ja vet-tä

hello one small coffee and water-PART

‘hello, one small coffee and some water’

The ambivalent items are as follows: the Estonian üks (‘one’) and colloquial Finnish yks (derived from yksi ‘one’) and the conjunction ja (‘and’) are identical in the two languages. The greeting and the collocation väike kohv (‘small coffee’) are unambiguously Estonian and vettä (‘[some] water’) is Finnish. The Estonian kohv is reminiscent of the Finnish kahvi (needless to say, these are lexical borrowings that may be considered internationalisms), while väike (‘small’) has no cognate in Finnish (cf. the Finnish pieni, ‘small’). The context of the interaction (i.e. the topic of conversation, the goods on display, etc.) is rather limited and the meaning can be derived from extra-linguistic clues. Even if the speaker would not recognize an isolated instance of väike, in this situation the price list helps where suur kohv (‘big coffee’) is mentioned (cf. the Finnish suuri (‘big, large’); although iso (‘id.’) would be more appropriate in this context, the link between suur and suuri can be easily established and väike kohv has a lower price than suur kohv.

All in all, it was demonstrated that both sides in EFC modify their speech, departing from their respective L1 and accommodating the other, albeit their strategies differ due to certain asymmetry as far as exposure to/experience in the other variety and roles in the EFC setting (i.e. customers vs. salespersons) are concerned.

6 Inherent vs. acquired RM

Russian speakers constitute slightly less than 50 per cent of the population of Tallinn (on the Russian language in Estonia in the post-occupation period, see Verschik, 2005, 2008, pp. 181–188). They too are involved in the service industry and in communication with Finnish customers in particular. Compared to Estonian speakers, of whom 16.6 per cent declared proficiency in Finnish, only 1.6 per cent of those whose L1 is Russian did that in the census of 2000. Unlike Estonians, as a rule, Russians do not have a several-decades-old habit of watching Finnish TV, neither do they have a history of oppositional identity that involves Finns and the Finnish language as a reference point. At the same time, young Russians receive compulsory instruction in Estonian as L2 (various models of bilingual and immersion education are available as well and an increasing number of ethnic Russians attend Estonian-medium secondary schools) and, working in the service industry, are more or less fluent in Estonian. Thus, Estonian, their L2, is related to Finnish (L3) and the nature of their encounter with Finnish is probably of the same type described by Jessner (2008), in terms of spontaneous grammar. Figuratively speaking, while ‘insiders’, i.e. speakers of closely related varieties, see differences, speakers of L1 unrelated to these varieties perceive the same type of varieties, whereas differences appear not as important.³

As discussed earlier, for Russian speakers this is acquired RM (through the prism of Estonian) as opposed to inherent RM in the case of Estonian speakers. Both empirically and theoretically, this is a fascinating instance of both types of RM in the same setting. At this point, I have very little data on Russian–Finnish communication (RFC) over the counter. So far, I have encountered two types of RFC: either Russians speak Estonian to Finns (probably assuming that their customers will understand) or a kind of quasi-Finnish that in some cases may have been acquired from their Estonian colleagues (frequently, Russian- and Estonian-speaking salespersons work side by side, as they do in Rotermanni Keskus). Whether and in what sense this quasi-Finnish differs from that of Estonian speakers, remains to be seen, as more fieldwork is needed to answer this question. Example (10a) differs from what an Estonian speaker would produce only inasmuch as the lexical item for the European currency ‘euro’ is a compromise version between the Russian jevro and the Estonian/Finnish euro (‘Euro’), namely jeuro (‘id.’). Otherwise, (10) is easily imaginable in the quasi-Finnish speech of an Estonian speaker:

(10a)

Se(e) on kümme jeuro-t sama sata- viisikümmen-tä

this is ten euro-PART same hundred-fifty

eesti kroon- u - a

Estonian kroon-SF-PART

‘this is ten Euros, same as one hundred and fifty Estonian crowns’

Compare to Finnish in (10b) and Estonian in (10c):

(10b)

Se on kymmenen euro-a, sama kuin sata-viisikümmen-tä

this is ten euro-PART same as hundred-fifty

eesti kruunu-a

Estonian kroon-PART

‘id.’

(10c)

See on kümme euro-t, sama nagu sada viiskümmend

this is ten euro-PART same as hundred fifty

eesti kroon-i

Estonian kroon-PART

‘id.’

In this utterance, the speaker can rely on a high number of ambivalent items (i.e. almost or entirely identical in the two varieties). The partitive case is used twice, once with the Estonian marker (jeuro-t, cf. Estonian euro-t) and the second time with the Finnish marker -a that is added to the combination of the Estonian nominative form kroon and the final vowel -u of the Finnish stem (cf. Finnish kruunu, ‘crown’ [currency]). The form kroon-u-a is a perfect instance of those compromise forms that emerge in EFC.

Thus, different types of RM (both involving Finnish) are present in Tallinn. This makes Tallinn a fascinating arena for RM fieldwork and research.

7 Conclusions: Directions for future research

The emerging field of RM research conceptualizes multilingual communication and speech as phenomena in their own right, that is, independent of respective monolingual norms. While case studies in RM concentrate on different aspects and use different approaches (intercultural pragmatics, contact linguistics, bilingualism, language pedagogy and language acquisition, to name but some), they all share the assumption that RM is not in any way inferior to a ‘proper’ L2 use.

Thus, Tallinn is a sociolinguistic arena where several different language contact situations can be observed: first, a more or less established practice of RM in EFC; second, both inherent and acquired RM are present. In this context, several research questions can be formulated:

- How do the speakers establish interlingual correspondences and apply transformation rules?

- What kind of compromise forms is understood and what is not, and why?

- We already know that accommodation in EFC is not symmetrical. How do the accommodation devices used by Estonian and Finnish interlocutors differ and what do they have in common? What factors affect the elaboration of accommodation strategies?

- To what extent are ‘usual’ contact-induced language change (for instance, in Ingrian Finnish of Estonian) and change produced in the course of RM communication the same or in what ways do they differ?

- Consider politeness strategies in EFC communication (see Hickey & Stewart, 2004 and especially Keevallik, 2004 for Estonia and Yli-Vakkuri, 2004 for Finland).

- What is the course of the development of acquired RM (i.e. Russian speakers communicating to Finns)? How do they acquire quasi-Finnish? To what extent do they rely on Estonian?

A further data collection and investigation of these questions may prove useful for the developing and expanding field of RM.

Footnotes

Acknowledgements

I wish to express my gratitude to Scott Jarvis (of Ohio State University) for his suggestions and comments.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

About the author

Anna Verschik is professor of general linguistics at Tallinn University and docent in Yiddish studies at Helsinki University. Her scholarly interests include Russian–Estonian language contacts, sociolinguistics of the Baltic countries and varieties of Yiddish in the Baltic region.

References

Ariste

(1981). Keelelkontaktid. Tallinn: Valgus.

Auer

(2007). Monolingual bias in bilingualism research – or: Why bilingual talk is (still) a challenge for linguistics. In Heller

(Ed.), Bilignualism: A social approach (pp. 319–339). Houndmills: Palgrave.

Backus

(1999). Mixed native languages: A challenge to the monolithic view of language. Topics in Language Disorders, 19(4), 11–22.

Backus

(2005). Codeswitching and language change: One thing leads to another? International Journal of Bilingualism, 9(3/4), 307–340.

Braunmüller

(2007). Receptive multilingualism in Northern Europe in the Middle Ages: A description of a scenario. In ten Thije

J. D.

Zeevaert

(Eds.), Receptive multilingualism (pp. 25–47). Amsterdam: John Benjamins.

Bullock

B. E.

Gerfen

. (2004). Phonological convergence in a contracting language variety. Bilingualism: Language and Cognition, 7(2), 95–104.

Bullock

B. E.

Toribio

A. J.

(2004). Introduction: Convergence as an emergent property in bilingual speech. Bilingualism: Language and Cognition, 7(2), 91–93.

Clyne

(1987). Constraints on codeswitching: How universal are they? Linguistics, 25, 739–764.

Clyne

(2003). Dynamics of language contacts. Cambridge: Cambridge University Press.

10.

Cook

(1995). Multi-competence and the learning of many languages. Language, Culture and Curriculum, 8(2), 93–98.

11.

Cook

(1997). Monolingual bias in second language acquisition research. Revista Canaria de Estudios Ingleses, 34, 35–50.

12.

Cook

(2006). Interlanguage, multi-competence and the problem of the ‘second’ language. Rivista di Psicolinguistica Applicata, 6(3), 39–52.

13.

Cook

(2007). Multi-competence: Black hole or wormhole for SLA research? In Han

Z.-H.

(Ed.), Understanding second language process (pp. 16–26). Clevedon: Multilingual Matters.

14.

Doetjes

(2007). Understanding differences in inter-Scandinavian language understanding. In ten Thije

J. D.

Zeevaert

(Eds.), Receptive multilingualism (pp. 217–230). Amsterdam: John Benjamins.

15.

Ellis

(2006). Monolingualism: The unmarked case. Estudios de Sociolingüística, 7(2), 173–196.

16.

Erelt

(2003). Study of Estonian. In Erelt

(Ed.), Estonian language. Linguistica uralica supplementary series, Vol. 1 (pp. 379–398). Tallinn: Estonian Academy Publishers.

17.

Franceschini

(2000). A multilingual network in the re-activation of Italian as the third language among German speakers: Evidence from interactions. Zeitschrift für Interkulturellen Fremdsprachen unterricht, 5(1). URL: http://zif.spz.tu-darmstadt.de/jg-05-1/beitrag/frances3.htm

18.

Franceschini

(2002). Umgang mit Fremdheit: Mixed style und Quasi-Italienisch bei Deutschschweizer Händlern im Gundeldingen (Basel). In Keim

Schütte

(Hrsg.), Soziale Welten und kommunikative Stile: Festschrift für Werner Kallmeyer zum 60. Geburtstag. Studien zur deutschen Sprache 22 (S. 217–252). Tübingen: Narr.

19.

Franceschini

(2003). Unfocussed language acquisition? The presentation of linguistic situation in biographical narration. Forum: Qualitative Social Research, 4(3). URL: http://www.qualitative-research.net/index.php/fqs/article/view/667/1442

20.

Franceschini

(2008). Multilinguismo: Prospettive per la ricerca. Multilingualism: Research opportunities. Vorschungsperspektiven Mehrsprachigkeit. Keynote speech at AILA 2008 conference ‘Multilingualism: Challenges and Opportunities’, Essen, 24–29 August.

21.

Grosjean

(2001). The bilingual’s language modes. In Nicol

(Ed.), One mind, two languages: Bilingual language processing (pp. 1–22). Oxford: Blackwell.

22.

Hassinen

(2002). Simultaaninen kaksikielisyys. Läheiset sukukielet viro ja suomi rinnakkain. Acta Universitatis Ouluensis. Series B. Humaniora 43. Oulu: Oulu University Press.

23.

Haugen

(1966). Semicommunication: The language gap in Scandinavia. Sociological Inquiry, 36, 280–297.

24.

Heine

Kuteva

(2005). Language contact and grammatical change. Cambridge: Cambridge University Press.

25.

Herdina

Jessner

(2002). A dynamic model of multilingualism: Perspectives of change in psycholinguistics. Clevedon: Multilingual Matters.

26.

Hickey

Stewart

(Eds.). (2004). Politeness in Europe. Clevedon: Multilingual Matters.

27.

Hinnenkamp

(2000). Constructing misunderstanding as a cultural event. In Di Luzio

Günthner

Orletti

(Eds.), Culture in communication. Analyses of intercultural situations (pp. 211–243). Amsterdam: John Benjamins.

28.

Hufeisen

Marx

(2007). How can DaFnE and EuroComGerm contribute to the concept of receptive multilingualism? In ten Thije

J. D.

Zeevaert

(Eds.), Receptive multilingualism (pp. 307–321). Amsterdam: John Benjamins.

29.

Jessner

(2008). Teaching third languages: Findings, trends and challenges. Language Teaching, 41(1), 15–56.

30.

Johanson

(2002). Structural factors in Turkic language contacts [With an introduction by Bernard Comrie]. London: Curzon.

31.

Kallio

(2004). Suomen ja Viron tiedesuhteet erityisesti Viron miehitysaikana vuosina 1940–1991. PhD dissertation, Tallinn University, Tallinn.

32.

Karam

(2000). Investigating mutual intelligibility and language coalescence. International Journal of the Sociology of Language, 146, 119–136.

33.

Kataja

Klaas

(2003). Elämisluvalla lopettaiaisiin. Viron suomalaisopiskelijoiden suomen kielestä. In Jönsson-Korhola

Lindgren

A.-R.

(Eds.), Monena suomi maailmalla. Suomalaisperäisiä kielivähemmistöjä (pp. 300–309). Helsinki: Suomalaisen Kirjallisuuden Seura.

34.

Keevallik

(2004). Politeness in Estonia: A matter-of-fact style. In Hickey

Stewart

(Eds.), Politeness in Europe (pp. 203–217). Clevedon: Multilingual Matters.

35.

Kloss

(1967). Abstand languages and Ausbau languages. Anthropological Linguistics, 9(7), 29–41.

36.

Lauristin

Vihalemm

(1997). Recent historical developments in Estonia: Three stages of transition (1987–1997). In Lauristin

Vihalemm

Rosengren

K.-E.

Weibull

(Eds.), Estonia: Return to the western world (pp. 73–126). Tartu: Tartu University Press.

37.

Loog

(1991) Esimene eesti slängisõnaraamat [the first Estonian slang dictionary]. Tallinn: Mai Loog.

38.

Martin-Jones

Romaine

(1986). Semilingualism: A half-baked theory of communicative competence. Applied Linguistics, 7(1), 26–38.

39.

Meissner

F.-J.

(2004). Transfer und Transferieren: Anleitungen zum Interkomprehensionsunterricht. In Klein

H. G.

Rutke

(Hrsg.), Neue Forschungen zur Europäischen Intercomprehension (S. 39–66). Aachen: Shaker.

40.

Muysken

(1995). Code-switching and grammatical theory. In Milroy

Muysken

(Eds.), One speaker, two languages (pp. 177–198). Cambridge: Cambridge University Press.

41.

Muysken

(2000). Bilingual speech: A typology of code-mixing. Cambridge: Cambridge University Press.

42.

Pavlenko

(2002). Poststructuralist approaches to the study of social factors in second language learning and use. In Cook

(Ed.), Portraits of the L2 User (pp. 277–302). Clevedon: Multilingual Matters.

43.

Pavlenko

Jarvis

(2002). Bidirectional transfer. Applied Linguistics, 23(2), 190–214.

44.

Praakli

(2007). Eesti keele kasutusväljad Tampere eesti kogukonna näitel. In Muikku-Werner

Kokko

Remes

(Eds.), VIRSU III. Suomalais-ugrilaisia kohdekieliä ja kontakteja (pp. 91–105). Joensuu: Joensuun Yliopisto.

45.

Praakli

(2008). Eestlased ja eesti keel Soomes. In Metslang

Günthal

(Eds.), Lähivertailuja 18 (pp. 157–175). Helsinki: Suomalaisugrilainen Seura.

46.

Raag

(1999). One plus one equals one: The forging of Standard Estonian. International Journal of the Sociology of Language, 139, 17–38.

47.

Rannut

(2007). The linguistic consequences of EU enlargement for Estonia. Sociolinguistica, 21, 22–35.

48.

Rehbein

(2006). The cultural apparatus: Thoughts on the relationship between language, culture, and society. In Bührig

ten Thije

J. D.

(Eds.), Beyond misunderstanding (pp. 43–96). Amsterdam: John Benjamins.

49.

Rehbein

Kameyama

(2003). Pragmatik. In Ammon

Dittmar

Mattheier

Trudgill

(Eds.), Sociolinguistics. Soziolinguistik. An international handbook of the science of language and society (pp. 556–588). Berlin: de Gruyter.

50.

Reinelt

(2008). Two types of receptive multilingualism in the East Asian Hanzi/Kanji area. Article presented at the Symposium on Receptive Multilingualism, part I (organized by J. Rehbein). 2008 conference ‘Multilingualism: Challenges and Opportunities’, Essen, 24–29 August.

51.

Remes

(1995). Suomen ja viron vertaileva taivutustypologiaa. Suomen ja saamen kielen ja logopedian laitoksen julkaisuja, 2. Oulu: Oulun Yliopisto.

52.

Riionheimo

(2002). How to borrow a bound morpheme? Evaluating the status of structural interference in a contact between closely-related languages. SKY Journal of Linguistics, 15, 119–143.

53.

Riionheimo

(2007). Muutoksen monet juuret. Helsinki: Suomalaisen Kirjallisuuden Seura.

54.

Rindler-Schjerve

Vetter

(2007). Linguistic diversity in Habsburg Austria as a model for modern European language policy. In ten Thije

J. D.

Zeevaert

(Eds.), Receptive multilingualism (pp. 49–70). Amsterdam: John Benjamins.

55.

Savijärvi

(1998). Cognate languages in contacts. Ingria – Where four Baltic–Finnish languages meet. In Niemei

Odlin

Heikkinen

(Eds.), Language contact, variation and change. Studies in languages (Vol. 32, pp. 269–286). Joensuu: Joensuu University.

56.

Schmid

(2004). Identity and first language attrition: A historical approach. Estudios de sociolingüķstica, 5(1), 41–58.

57.

Sulkala

(1996). Finnish as a second language fro speakers of related languages. In Martin

Muikku-Werner

(Eds.), Finnish and Estonian – New target languages. Proceedings of the Fenno-Ugric languages as second and foreign languages symposium (pp. 143–158). Jyväskylä: Soveltavan kielentutkimuksen keskus.

58.

Sutrop

(1997). Entspricht Estnisch dem agglutinierenden Sprachtypus? In Erelt

(Ed.), Estonian typological studies 2 (pp. 199–219). Tartu: Tartu University Press.

59.

Thomason

S. G.

(1997). On mechanisms of interference. In Eliasson

Jahr

E. H.

(Eds.), Language and its ecology: Essays in memory of Einar Haugen (pp. 181–207). Berlin, NY: Mouton de Gruyter.

60.

Thomason

S. G.

(2001). Language contact: An introduction. Washington, DC: Georgetown University Press.

61.

Van Bezooijen

Gooskens

Ch.

(2005). How easy is it for speakers of Dutch to understand Frisian and Afrikaans, and why? Linguistics in the Netherlands, 22, 13–24.

62.

Van Bezooijen

Gooskens

Ch.

Kürschner

Schüppert

(2008). Linguistic factors of mutual intelligibility in closely related languages. Article presented at the Symposium on Receptive Multilingualism, part II (organized by J. D. ten Thije), AILA 2008 conference ‘Multilingualism: Challenges and Opportunities’, Essen, 24–29 August.

63.

Verschik

(1999). Some aspects of the multilingualism of Estonian Jews. International Journal of the Sociology of Language, 139, 49–67.

64.

Verschik

(2005). The language situation in Estonia. Journal of Baltic Studies (Baltic Sociolinguistic Review, Special Issue), 36(3), 283–317.

65.

Verschik

(2008). Emerging bilingual speech: From Monolingualism to code-copying. London: Continuum.

66.

Werlen

(2007). Receptive multilingualism in Switzerland and the case of Biel/Bienne. In ten Thije

J. D.

Zeevaert

(Eds.), Receptive multilingualism (pp. 137–157). Amsterdam: John Benjamins.

67.

Yli-Vakkuri

(2004). Politeness in Finland: Evasion at all costs. In Hickey

Stewart

(Eds.), Politeness in Europe (pp. 189–202). Clevedon: Multilingual Matters.

68.

Zeevaert

(2007). Receptive multilingualism and inter-Scandinavian semicommunication. In ten Thije

J. D.

Zeevaert

(Eds.), Receptive multilingualism (pp. 103–135). Amsterdam: John Benjamins.

69.

Zeevaert

ten Thije

(2007). Introduction. In ten Thije

J. D.

Zeevaert

(Eds.), Receptive multilingualism (pp. 1–23). Amsterdam: John Benjamins.