Printed by

SLAVICA HELSINGIENSIA 46 FINNISH AND RUSSIAN AS LINGUA FRANCAS JOINT ACTIVITY IN CONVERSATIONS Merja Pikkarainen HELSINKI 2015 SLAVICA HELSINGIENSIA 46 Series editors Tomi Huttunen, Jouko Lindstedt, Ahti Nikunlassi Published by: Department of Modern Languages P.O. Box 24 (Unioninkatu 40 B) 00014 University of Helsinki Finland Copyright © by Merja Pikkarainen ISBN 978-951-51-0269-0 (paperback) ISBN 978-951-51-0270-6 (PDF) ISSN-L 0780-3281, ISSN 0780-3281 (Print), ISSN 1799-5779 (Online) Cover design: Laura Vainio Printed by: Unigrafia Abstract This study analyses the language use of non-native speakers of Finnish and non-native speakers of Russian by exploring the joint activity by two or more participants in authentic conversations. More specifically, the present focus is on word searches, collaborative productions and candidate understandings. Although these structures have been claimed to be similar phenomena, they also have different characteristics. The data of this study consist of approximately 11 hours of naturally occurring conversations in a range of everyday and institutional situations in Finnish and Russian. The methodological framework adopted for this analysis is interactional linguistics, which focusses on analysing the language use in naturally occurring interaction. Interactional linguistics can be defined as a part of linguistics that originates from different types of functional orientations (conversation analysis, functional linguistics, and anthropological linguistics). However, the main analytical tools for this study are provided by the methodology of conversation analysis. A typical collaborative production consists of a preliminary component that is produced without any specific hesitations or perturbations, and of a final component that fits syntactically, prosodically and semantically with the preliminary component. For example, a preliminary component can consist of a subject + verb combination and a final component contains a complement (in a broad sense). A typical word search can be initiated with pauses or hesitations, or with specific questions. In these cases, the coparticipant is considered to be a knowledgeable partner. With respect to both collaborative productions and word searches, either the initiator of the word search, or the speaker who produced the preliminary component, has an opportunity to evaluate the suggested word or the final component that the recipient produces. When comparing collaborative productions and word searches, the candidate understandings fall somewhere in the middle. Interactionally, candidate understandings are used to check the recipient’s interpretation of the prior and to maintain mutual understanding. All three types of joint activity emphasize the willingness of the participants to cooperate and the fact that although the language used is foreign to all of the participants, the interaction need not be problematic. On the contrary, it would seem that since no native speakers are present, the conversations become relaxed. The data shows that the resources of these two lingua francas are utilised somewhat differently. The Finnish data consist of proportionally more searches that concentrate on grammatical issues than the Russian data. On the other hand, candidate understandings are more common in the Russian data. These differences can be due to the participants’ different linguistic backgrounds. The lingua franca speakers of Russian may have a more native-speaker relation to the Russian language in that they do not consider the grammatical appropriateness to be as important as the lingua franca users of Finnish. Aluksi – Предисловие Tämä väitöstutkimus ei olisi ollut mahdollinen ilman monien ihmisten tukea. Kiitos teille kaikille, jotka olette olleet mukana matkalla! Suurin kiitos kuuluu ohjaajalleni, professori Arto Mustajoelle, jonka kanssa minulla on ollut ilo käydä vuosien varrella monta mielenkiintoista keskustelua ja ohjaustuokiota. Kiitos, Arto, että otit minut mukaan Suomen Akatemian lingua franca -projektiin. Kiitos, että olet antanut minun miettiä asioita rauhassa ja tarpeen vaatiessa ohjannut taas eteenpäin. Toivottavasti keskustelut kielten ja kommunikaation kiehtovista kiemuroista jatkuvat tulevaisuudessakin. Я хочу также выразить сердечную благодарность доктору педагогических наук Екатерине Юрьевне Протасовой за ценные советы, критические замечания и всевозможную поддержку во время проведения исследовательской работы. Emil Aaltosen Säätiölle esitän mitä syvimmät kiitokset vuosina 2012 ja 2013 myönnetystä tuesta, joka on mahdollistanut väitöstutkimukseen keskittymisen. Я выражаю искреннюю признательность своим рецензентам, доктору филологических наук Марике Калюга и профессору Николаю Борисовичу Вахтину за множество полезных советов, замечаний и рекомендаций на заключительной стадии работы. Само собой разумеется, что за все возможные ошибки и недостатки в этой книге отвечает сам автор. Я выражаю особую благодарность всем коллегам, с которыми я встречалась на разных этапах работы над диссертацией. В особенности хочу поблагодарить участников аспирантского семинара, а также всех сотрудников Кафедры русского языка и литературы. Kanssanne on ollut ilo tehdä töitä! Tutkimuksen teon alkuvaiheessa sain vertaistukea Venäjän kielen ja kirjallisuuden oppiaineen jatko-opiskelijoilta. Kiitos, meidän oma aspiranttiporukkamme Hanna Laitinen, Olga Pussinen, Anu Reponen ja Ilona Sammalkorpi! Slaavilaisen filologian puolella haluan esittää kiitokseni Max Wahlströmille, jolla on aina aikaa pohdinnoille ja Johanna Virkkulalle, joka jakoi vastaväitelleenä ystävällisesti tietämystään väitösvalmisteluista ja auttoi väitöskirjan teknisessä toteutuksessa. Ystävät kalliit, teitä on niin paljon! Sanni, olet maailman paras tauottaja. Kaisa, Satu, Anna ja Marina, olette vieneet ajatuksia muualle. Ehkä ehdin jatkossa taas keikoille ja muihin rientoihin. Venäjä ja venäjä ovat olleet läsnä elämässäni myös muiden kuin opiskelujen merkeissä. Kiitos ja kumarrus kaikille vuosien varrella menossa pyörineille. Дорогие друзья и в Финляндии и в России, спасибо вам всем поддержку! До скорой встречи! Suuri kiitos Kuhmo-osastolle! Kiitos äiti ja isä, että olette tarjonneet muutakin ajateltavaa kuin väitöskirja ja pitäneet huolta siitä, että Etelän ihmisen ei ole tarvinnut vapaa-ajallaankaan kärsiä tekemisen puutteesta. Mikko, Merja ja ihanat pikkutytöt, toivottavasti näemme tulevaisuudessa useammin. Pekka, sinuakin näkisi mielellään enemmän. Mari, olet tuki ja turva isosiskollesi. Kiitos, Jalmari-armaani! Sinä tulit elämääni ja jäit. Yksi kiiteltävistä on jo poistunut keskuudestamme, mutta hän elää edelleen muistoissamme. Omistankin tämän väitöskirjan edesmenneelle mummulleni Rauni Piiraiselle (o.s. Huusko, 1922–2009). Kanta-Hämeessä, Mommilassa 18.12.2014 Merja Pikkarainen Table of contents ABSTRACT ........................................................................................................ 3 ALUKSI – ПРЕДИСЛОВИЕ ................................................................................. 5 1 INTRODUCTION ............................................................................................ 10 1.1 The subject of the study and research questions ..................................... 10 1.2 The theoretical and methodological framework ..................................... 16 1.3 The structure of the study ...................................................................... 22 2 RESEARCH ON ENGLISH AS A LINGUA FRANCA AND NON-STANDARD FORMS OF RUSSIAN ........................................................ 23 2.1 English as a Lingua Franca .................................................................... 24 2.2 Diaspora Russian and Colloquial Russian .............................................. 30 3 PREVIOUS RESEARCH AND KEY CONCEPTS ................................................... 40 3.1 Organisation of turn-taking in relation to collaborative productions ...... 40 3.1.1 Compound-TCU, secondary TRP .................................................... 44 3.1.2 Formal aspects and functional aspects ............................................. 47 3.1.3 Co-constructions and increments ..................................................... 48 3.2 Repair organisation in relation to word searches .................................... 50 3.2.1 Who repairs and under what circumstances?.................................... 51 3.2.2 Candidate understandings – somewhere in between word searches and collaborative productions ................................................................... 57 4 DATA AND TRANSCRIPTION .......................................................................... 60 4.1 Finnish data ........................................................................................... 62 7 4.2 Russian data .......................................................................................... 65 4.3 Discourse transcription, interlinear gloss, and translation ...................... 66 5 COLLABORATIVE PRODUCTIONS ................................................................. 69 5.1 Previous research on collaborative productions ..................................... 69 5.2 Participants co-construct a phrase or a clause ........................................ 72 5.2.1 Co-ordinated words and phrases ...................................................... 72 5.2.2 Agreement within phrases ............................................................... 78 5.2.3 Subject NP and the predicate verb ................................................... 83 5.2.4 Objects, other complements required by predicate verbs, and predicatives .............................................................................................. 87 5.2.5 The semantic motivation that triggers collaborative productions: the Finnish verb mennä, ‘to go’ + direction .............................................. 98 5.3 Participants co-construct a complex sentence ...................................... 101 5.4 Summary ............................................................................................. 107 6 WORD SEARCHES ...................................................................................... 109 6.1 Previous research on word searches ..................................................... 109 6.2 How speakers indicate that there is a problem? ................................... 111 6.2.1 Pauses and hesitations that indicate a word search......................... 113 6.2.2 The use of the spoken lingua franca .............................................. 120 6.2.3 Loan word examples ..................................................................... 126 6.3 Actual resolving of the search.............................................................. 135 6.3.1 Direct offer, one word ................................................................... 135 6.3.2 Prolonged search sequences .......................................................... 138 6.4 Reaction to the resolution .................................................................... 146 6.4.1 The original speaker repeats or otherwise accepts the offered word ........................................................ 146 6.4.2 Original speaker continues directly ............................................... 151 6.4.3 Either the co-participant continues or the continuation is longer.... 153 8 6.5 Unsolved cases .................................................................................... 155 6.6 Summary ............................................................................................. 156 7 CANDIDATE UNDERSTANDINGS – SOMEWHERE IN THE MIDDLE? ................ 159 7.1 Full clauses .......................................................................................... 160 7.2 Noun phrases ....................................................................................... 170 7.3 Summary ............................................................................................. 174 8 CONCLUSION AND IMPLICATIONS FOR FURTHER STUDY............................. 176 8.1 Collaborative productions and word searches ...................................... 176 8.2 Further research ................................................................................... 181 REFERENCES ................................................................................................ 184 APPENDIX ..................................................................................................... 199 9 1 Introduction 1.1 The subject of the study and research questions The main objective of this dissertation is to describe how the non-native Finnish and non-native Russian languages are used in authentic conversations by non-native speakers. Even though the language adopted in these conversations is non-native to the participants, they are nevertheless able to interact and communicate. More specifically, I am interested in the interactional phenomena in these types of conversations, such as how participants in conversation co-construct different types of structures and how they themselves orient to these collaborations. For example, how does a participant indicate that he or she is missing a word and how do the other participants help in finding that word or item? Nonetheless, an exchange between speakers can also occur without any explicit sign of problems arising in their utterances. My aim is to observe and describe the relation between the linguistic and conversational structures. This study explores the structures that are jointly produced by two or more participants during their conversations. The main focus of this study is on word searches (for example, see Helasvuo et al. 2004, Kurhila 2006), collaborative productions (for example, see Helasvuo 2004; Lerner 1991, 1996, 2004a; Szczepek 2000a, 2000b) and candidate understandings (for more information, see Kurhila 2006: 153–217). Although these structures have been analysed as being part of the same phenomenon, they also have different characteristics. I will attempt to define to what extent these phenomena overlap or diverge. Whenever foreign language learners engage in conversation, one assumption is that an essential part of their conversations entails word searches. These searches can be seen as a part of language acquisition and as an imperfect knowledge of language. However, this view is challenged by other jointly produced structures, such as collaborative completions. For instance, Savijärvi (2011) observed how younger children learn Swedish as a second language in an immersion kindergarten. She noticed that children were able to produce Swedish co-constructions during the early stages of their immersion schooling. However, one challenge is how understandable the frame is and what part of the teacher’s utterance is missing. The teacher can also ‘trigger’ pupils to become active in their utterance production, but in some cases, pupils continue to produce an utterance that has already been completed. In such cases, the teacher cannot use syntactically incomplete 10 1 Introduction utterances because they are syntactically complete. (Ibid. 148–149, 152.) Teachers can also use what is referred to as Designedly Incomplete Utterances (DIUs; Koshik 2002, Margutti 2010). According to Koshik (2002: 279), teachers use DIUs, ‘to elicit self-correction of written errors’. A language that is used between non-native speakers can be referred to as a lingua franca. The lingua franca that is currently the most used in the world today is English (English as a Lingua Franca; ELF). The status of English as a global and international language means that there are different conditions and goals for its research in comparison to the more locally used languages such as Finnish (for instance, see the following edited volumes: Jenkins 2007, Mauranen & Ranta 2009). Another point worth noting is that non-native speakers have outnumbered the number of native speakers of the ‘core’ varieties of British and American English (Mauranen & Metsä-Ketelä 2005: 1, see also Crystal 2001: 61, Seidlhofer 2001: 141). This analysis focusses on two neighbouring languages that belong to different language groups. Finnish is spoken by approximately five million people and it is a member of the Finnic group of the Uralic family of languages. In contrast, Russian is a Slavic language that is a member of the Indo-European language family. According to traditional morphological typology, Finnish is an agglutinating inflectional language and Russian a fusional language. The phonetic systems of these languages differ. A rather simple explanation is that Finnish is a language with vowels and Russian, a language with consonants. (Bondarko 2009: 23–29, Iivonen 2009: 48–60.) Prosodically, the place of word stress in Russian is not fixed, whereas in Finnish, stress always falls on the first syllable (Bondarko 2009: 31, Iivonen 2009: 60). Finnish also has a phonological opposition of sound length that applies to both vowels and consonants, but it is not connected to stress (Iivonen 2009: 48, 59). However, the Finnish and Russian languages possess some characteristics that make them more similar in certain aspects than some of the European languages within the Indo-European family. Both languages share similarities, such as the absence of articles, comparatively free word order, an alternative structure for the verb to have, as well as rich word-formation possibilities. These two languages currently face somewhat different situations. Until recently, Finns have usually migrated to other countries and learnt foreign languages (see the edited volume by Jönsson-Korhola & Lindgren 2003) with little migration to Finland. However, increasing immigration to Finland has resulted in Finnish now being spoken as a lingua franca. However, this 11 1 Introduction language form and its usage have not been studied extensively thus far (see, however, Koivisto 1994, Lehtonen 2006, Mazeland & Zaman-Zadeh 2004, Siitonen & Martin 2001). Due to the history of Russian being used as a lingua franca during the Soviet era in the Soviet Union and in the countries of the Eastern Bloc, the situation of Russian as a lingua franca is somewhat different. However, the position and significance of Russian has decreased after the dissolution of the Soviet Union (Pavlenko 2006, 2008; Zaprudski 2007). When Russian lost its status as a supra-ethnic language following the collapse of the Soviet Union, this had a major impact on almost 36 million Russian speakers. This population of what is called the ‘Near Foreign’ countries is also often interchangeably referred to as the Russian diaspora, or simply as the Russian-speaking population, although only 25 million of these 36 million Russian speakers were ethnic Russians. (Pavlenko 2006.) However, in many post-Soviet countries, Russian continues to be used in interethnic communication (as in Tajikistan; Lemivaara-Khudoikulova 2011, Pavlenko 2006). Furthermore, the research agenda in Russia is currently dominated by language policy studies. That is, the micro-sociolinguistic studies of the regional varieties of Russian and of the actual language have attracted less attention (see, however, Lemivaara-Khudoikulova 2011, Verschik 2004). In the Russian Federation, the significance of Russian has increased because many of the minority language speakers either use Russian in many spheres, or have changed their language to Russian (for additional information, see the following edited volume: Zamjatin, Pasanen & Saarikivi 2012). The concept of a word search is used to refer to a speaker in conversation who encounters difficulties in finding a correct word or other expression, such as the correct noun form. Thus, the speaker shows uncertainty, or displays hesitancy, implying that he or she is having difficulties conveying their meaning in a way that the recipient would understand. This uncertainty or hesitancy leads to the progressivity of the turn being halted due to the speaker experiencing difficulties in formulating talk that would be understandable to others. (Kurhila 2006: 91.) According to Goodwin and Goodwin (1986), a prototypical word search is either preceded by sound stretches or other speech perturbations and a pause. Speakers conduct word searches in different types of conversations (for instance, see Helasvuo et al. 2004 for word-search sequences in the conversations of Finnish speakers with Aphasia; Hayashi 2003 for word searches in Japanese conversations). 12 1 Introduction When a speaker is unfamiliar with a concept, this might create problems in his or her utterance formulation and to illustrate this, I present an excerpt from my data. In this example, speaker N is encountering problems with a specific Russian concept. 1 For example, it is customary in Russian academic speech to construct compound nouns with a part of an adjective and a part of a noun, such as, filfak: filologičeskij fakul’tet, ‘Faculty of Philology’. In this particular example, the participants are discussing the educational system of Tajikistan and in particular, the doctoral training offered there. In line 1, N tries to form the appropriate word and initiates the search by stating v ètom med-, ‘in this med-’. The word med- is cut-off and the utterance ends with slightly falling intonation. In line 2, A acknowledges that the idea in N’s utterance was correct and confirms it with the affirmative particle da, ‘yes’. She then takes the whole construction and recycles it as a part of her own utterance: v medkolledž postupila, ‘I was admitted to the medical institute’. The word consists of the truncated adjective med- (from medicinskij, ‘medical’) and the noun kolledž, ‘college’. This type of truncation is typical of Russian colloquial speech in educational and vocational spheres (Zemskaja et al. 1981: 126–127). (1) TI 37. <8:32> 01 N: → 02 A: → 03 N: 04 A: .h A posle posle škol-y (.) posle vy srazu .h but after after school-GEN after 2PL immediately postupi-l-i v ↑èto-m (.) med--, be:admit-PST-PL PRE this-PREP (.) med-But right after school and then you immediately got admitted to this med, Da v medkolledž-ø [postupi-l-a. Yes PRE medical:institute-ACC be:admit-PST-FEM Yes I was admitted to the medical institute. [Èt-o by-l-o ↑vs-ë 3SG-NEU be-PST-NEU all-NEU po-rus-sk-i. PRE-Russian-ADJ-ADV It was all in Russian. .h Èt-o by-l-o da: vs-ë, .h 3SG-NEU be-PST-NEU yes all-NEU It was yes all, Lerner (1991, 1996) has made the major contribution to the subject of ‘collaborative productions’, and he refers to the phenomenon as ‘sentencesin-progress’. It was originally Sacks (1992a) who commented that they were ‘collaboratively built sentences’. By comparison, Helasvuo (2004) and Ono 1 The data and transcription principles are described in chapter 4. 13 1 Introduction & Thompson (1995) classify collaborative productions as ‘coconstructions’. Szczepek (2000a, 2000b) has examined both the functional and the formal aspects of collaborative productions in English conversation. Although different researchers emphasize slightly different aspects of these collaborations, the production of these joint structures is usually interpreted as being an interactional achievement. Thus, when one speaker completes an utterance by another speaker, this reveals aspects of an interactionally relevant syntax in that participants analyse syntactic structures and orient to them. (Auer 2009: 4; Helasvuo 2004: 1334; Lerner 1991: 441, 445; Pekarek Doehler 2011.) The following excerpt illustrates a typical conversational example. T and S are discussing the different types of civic organisations that are financed by the Finnish Ministry of Education and Culture. T suggests (line 2) that perhaps it would be better if there would be some coordinating organisation that would receive a greater sum of money to be shared among the smaller organisations. 2 However, T also acknowledges that this might eventually lead to arguments or in-fighting. In line 3, S adds the word keskenänne, ‘with each other’, and completes the utterance: (2) MV 41. <01:19:49> 01 T: 02 → 03 S: → 04 T: Mielummin joku <kattojärjestö> Rather some:kind roof:organisation kene-llä annet-aan ne parikymmentä who-ADE give-PASS those twenty, Rather some kind of coordinating organisation that would get those twenty, ((thousand euros)) ja sitten te (.) vo-isi-tte tappel-la ↑siellä, And then 2PL can-CON-2PL fight-INF there And then you could argue there, Keske[nä-nne (-). With each other-PX.2PL With each other. [Keskenä-nne niin paljon kuin h(h)alua-tte. With each other-PX.2PL as much as want-PL2 With each other as much as you want. In addition to the word searches and to the collaborative productions that enable the joint activity in conversations, it is also possible that either the speaker or the recipient may, instead of providing the next relevant turn, 2 In line 1, there are a few mismatches in terms of Finnish grammar. Firstly, speaker T uses the animate pronoun kuka,‘who’, although the referent is the inanimate kattojärjestö, ‘coordinating organisation’. Secondly, she uses the inappropriate form, kene-llä, ‘which-ADE’, although the correct case would be kene-lle, ‘which-ALL’. I have transcribed the conversations based on their actual language use so the language has not been corrected. 14 1 Introduction shift the focus of the conversation to certain elements in the prior speech. These types of examples in which a co-participant articulates his or her interpretation of the previous turns, can be referred to as candidate understandings. A speaker can repair a problem they felt that the previous speaker had, but he or she can also check his or her understanding of the problematic utterance. In other words, the speaker can articulate an understanding of the other speaker’s meaning. Thus, rather than merely modifying the prior, the articulations take it one step forward from the problematic situation by presenting a potential interpretation of the trouble turn. (Kurhila 2006: 153–217.) It may be argued that word searches and collaborative completions are connected. According to this assertion, word searches allow collaborative completion and they can be described as being grammatically designed for conditional entry by recipients (Lerner 1996: 261–263). Word searches provide an opportunity for the recipients of talk to contribute to the search and to collaboratively complete it. Moreover, a collaborative completion is typically designed so that only the next word being sought is produced. After the search, the primary speaker may continue his or her turn. Lerner (ibid.) noted that the search is often placed near the end of the unit, which consequently provides a place for the co-participant to complete the unit in progress. Furthermore, a word-search sequence can provide a sequential environment for deciding who will deliver the story in a situation that involves several potential tellers (Lerner 1992). My approach is somewhat different from this. I shall attempt to demonstrate that word searches and collaborative productions have specific positions in conversations and that participants orient to them differently. As was evident in the first two examples, the word search was accompanied by hesitancy markers and other perturbations. During the collaborative production, the second speaker added the word at the end of the utterance and this added word immediately followed the previous speaker’s utterance. In both cases, the initiator of the word search or the previous speaker accepted the suggested or added word by repeating it. My aim is to observe how word searches and collaborative productions are distributed in conversation. One pertinent question is what characterizes the utterances that can be described as either collaborative productions, or as word searches? As I have mentioned earlier, these structures can be defined as being a part of the same phenomenon so that they form a type of continuum. Candidate understanding might also serve as a bridge-builder 15 1 Introduction between these two activities. My aim is to determine the linguistic and interactional features word searches, collaborative productions and candidate understandings. Other factors that will be taken into account are syntax and other grammatical features, as well as how participants orient to them on a conversational level. Another set of research questions in this dissertation is connected to the two languages that will be analysed. That is, I am interested in determining whether speakers of the Finnish and Russian languages construct jointly produced structures differently, or whether the proportions of instances differ. This is not purely comparative research, but it reflects the type of resources these languages access in order to solve word searches or to jointly produce different types of structures. In brief, this analysis adopts a cross-linguistic perspective. 1.2 The Theoretical and Methodological Framework This paper explores the interrelationship between linguistic forms and the different ways of organising social interaction by focussing on both linguistic and conversational structures. The methodological framework for this is interactional linguistics, which is interested in analysing language use in naturally-occurring interaction (for more information, see the following edited volumes: Hakulinen & Selting 2005, Ochs et al. 1996, and Selting & Couper-Kuhlen 2001). Interactional linguistics can be defined as a part of linguistics, originating from different types of functional orientations (conversation analysis, functional linguistics, and anthropological linguistics). However, the main analytical tools are provided by the rigorous methodology of conversation analysis (CA; see, e.g. Heritage 1996 [1984], Psathas 1995, Sacks et al. 1974). Since the beginning of the 1990s, an increasing number of researchers have adopted the premise that language structure and use are inseparable from interaction. This means that any linguistic phenomenon must be examined in relation to its social-sequential context of occurrence. (For example, see the following edited volumes: Couper-Kuhlen & Ford 2004, Couper-Kuhlen & Selting 1996, Ford et al. 2002, Hakulinen & Selting 2005, Ochs et al. 1996, Selting & Couper-Kuhlen 2001.) A fundamental tenet of conversation analysts is that casual conversation is the basic or primordial form of all forms of talk (for instance, see Heritage 1996 [1984]: 234–235, Sacks et al. 1974: 730, Schegloff 1987). However, during the last few decades, CA-based research has begun to examine a 16 1 Introduction wide range of interactions, which are predominately located within institutional settings and practices such as courtrooms, journalistic interviews, and political speeches (see Clayman 1992, Drew & Heritage 1992, Heritage 1985, Sammalkorpi 2006). What then follows is that, on the one hand, the structures and the grammar of the language have formed to fulfil the needs of interaction (see, e.g. Couper-Kuhlen & Selting 2001: 3, Sacks et al. 1974: 722, Schegloff 1989: 143–144, Schegloff 1996: 54–56). On the other hand, when interpreting linguistic expressions, the most essential factor when interpreting linguistic expressions is the placement of the expression in the sequential context of occurrence (see, e.g. Couper-Kuhlen & Selting 2001: 3–5, Fox 2007: 304, Schegloff 1996). According to Schegloff (1996: 56), ‘grammar stands in a reflexive relationship to the organisation of a spate of talk as a turn’. Conversation occurs in context and, of course, context is not stable, but the participants co-construct it continuously, on a turn-by-turn basis. The important part is the local interactional context. In other words, the content or the form of the turns cannot be characterized primarily by the participants in the conversation, or by the themes or other external factors. Thus, turns need to be analysed with respect to previous and following turns. This means that the participants’ linguistic and other behaviour are not merely reactions to the context (context-shaped), but they are also one means of constructing context (context renewing). (Goodwin & Heritage 1990: 289.) In short, conversation does not take place in a vacuum. From the viewpoint of the context and the internal structure of the conversation, jointly produced structures are at a junction because participants construct and shape context together. If a turn is constructed by two or more speakers, this reflects that participants can process utterances in real-time and produce continuations that are suitable both syntactically and from the viewpoint of the conversation. (Helasvuo 2004; Sacks 1992a: 651, 654; Sacks 1992b: 57–58.) CA has its roots in ethnomethodology, which is a sociologically oriented field of study. Ethnomethodology focusses on studying how people manage in everyday activities. Unlike strictly linguistic discourse analysis, CA-based research tries to avoid using categories that are defined in advance. Thus, research questions and categories originate from the data. With respect to this study, although the data collection is aimed at obtaining spoken Finnish and Russian lingua franca material, the actual research theme was not selected until after some recordings were already collected. 17 1 Introduction One can also talk about the ‘emic’ relevance of the different category memberships. The emic perspective means that the participants’ own perspectives are considered when conducting a research. (Firth & Wagner 1997: 286.) In a way, CA-oriented research leaves the decision making to the participants. This means that the researcher does not need to make hypothetical classifications, but rather he or she can rely on the actual linguistic behaviour of the participants. (Kurhila 2006: 32.) This also means that all the details of interaction can be important, and nothing can be left out of the analysis as irrelevant or accidental (Ford 2001: 59–60, Heritage 1996 [1984]: 237–238, Psathas 1995: 2, Schegloff 1997: 502). This approach is suitable the present study since little research has thus far been conducted on Finnish or Russian in lingua franca interaction. What, then, does interactional linguistics offer to a study of lingua franca communication? As a starting point, interactional linguistics emphasises the analysis of individual languages as well as the analysis of cross-linguistic and cross-cultural aspects (Couper-Kuhlen & Selting 2001: 3). One goal of interactional linguistics is a better understanding of the nature of the influence that languages and interaction have on each other (ibid.). Regarding typologically different languages, interactional linguistics research suggests ‘that the way common problems are dealt with in interaction may be shaped the distinct linguistic resources which a language provides’ (ibid. 8). Thus, research on lingua franca communication adds one layer to the interactional linguistics research. Although the participants are not native speakers of Finnish or Russian, all the resources of their mother tongues and the other languages they know are present in every conversation. They also have knowledge about different conversational situations. For instance, they know the appropriate way to answer questions or how to conduct oneself in institutional settings. (Canagarajah & Wurr 2011.) As mentioned above, CA-based research attempts to avoid using categories that are defined a priori. Thus, the notion of ‘lingua franca interaction’ is somewhat problematic. This could be compared to categorisations such as ‘women’s interactions’ or ‘elderly persons’ interactions’. However, it should be taken as an analyst construct, potentially without procedural relevance for the participants (Firth 1996: 241, Schegloff 1991). Procedural relevance refers to the category being demonstrable in the talk itself and the participants themselves making it relevant. That is also the case in relation to the participants of this study. For example, the participants 18 1 Introduction often mention that they do not know the Finnish language well enough or that it is not their mother tongue (see also Firth 1996: 241). If the notion of ‘lingua franca’ is taken as a conceptual categorisation, it does provide some advantages. This language use can also be described as being ‘foreigner talk’, ‘interlanguage talk’, or some type of ‘learner interaction’. However, as Firth (1996: 241; emphases original) defines: For in contrast to these latter-mentioned categorizations, the term 'lingua franca' attempts to conceptualize the participant simply as a language user whose realworld interactions are deserving of unprejudiced description, rather – as these latter categories – than as a person conceived a priori to be the possessor of incomplete or deficient communicative competence, putatively striving for the 'target' competence of an idealized 'native speaker'. In the 1970s and 1980s, the conversation analytic tradition was monolingual and researchers analysed their own mother tongue. The monolingual tradition has been validated by asserting that researchers should be able to utilize all their native intuitions, suggesting that they should only work in their native language (Wong & Olsher 2000: 114–115; see also Seedhouse 1998 for criticising monolingual research paradigm). However, researchers currently extend their attention to situations that involve non-native speakers, speakers with different speech disorders and communication between parents and children (Helasvuo et al. 2004; Kurhila 2003, 2006; Laakso 2006). The research on lingua franca interactions challenges CA’s working assumptions (for example, orderliness in talk) and tests their applicability. It can be said that concerning the research on lingua franca interactions, CA provides a basic methodology for describing how these interactions are both sequentially and socially constructed. Furthermore, consideration of the data type actually allows us to define better CA’s methods and working assumptions. (Cf. Firth 1996: 240, 253.) It can therefore be concluded that the CA-based approach is also suitable for participants conversing in a foreign language. CA emphasises the cooperation between participants. Even though it might ostensibly seem that the participants have nothing in common – they do not have the same mother tongue or cultural background –the shared feature that they are using a foreign language to conduct a conversation (cf. Hülmbauer 2009). The emphasis on cooperation means that analysis is not merely guessing about a speaker’s intentions, but attempting to explain how and in what ways participants attempt to achieve mutual understanding. Understanding is 19 1 Introduction considered to be a continuous and dynamic process that is constructed and modified in and through interaction (Kurhila 2006: 19). In the CA tradition, achieving mutual understanding is referred to as intersubjectivity (Heritage 1996 [1984]: 249–257). Intersubjectivity does not attempt to deny a person’s individuality or uniqueness of experiences, but it works on the assumption that people’s experiences are sufficiently similar for them to act together and communicate (ibid. 64–70). When intersubjectivity faces the risk of breakdown, participants have means to remedy the situation (Schegloff 1992). For example, participants may return back in their conversations and perform repair initiations (Schegloff et al. 1977). Researchers on the Russian language have focused on the factors that might result in miscommunication (kommunikativnaja neudača, ‘communicative failure’; Ermakova & Zemskaja 1993; Mustajoki 2012, 2013b). One reason for certain situations leading to communicative failure is that the participants’ mental worlds differ (Mustajoki 2012: 218). As a consequence, participants must take into account these possible differences and design their utterances to accommodate the fact that the other participants’ mental world may be different and this is called the ‘recipient design’; for example, see Mustajoki 2013b: 7–10). Although the focus of the Russian research is somewhat different compared to Western research, the Russian research has revealed that miscommunication or communicative failures are just as common in interaction between friends and family members as in intercultural interaction (Ermakova & Zemskaja 1993: 31– 32). Temporality shapes both conversation and the formation of linguistic structures (for example, see Auer 2005; Schegloff 1996: 55–56, 102). Temporality affects the grammatical structure of a conversation. This phenomenon can be called Emergent Grammar (Hopper 1987, 1998, 2011). Hopper (1987: 141) argues that grammar ‘must be viewed as a real-time, social phenomenon, and therefore is temporal’. Hopper (2011: 26) claims that dialog does not consist of syntactic structures that are generated by rules rather than assembling familiar fragments. This position envisions emergence as a synchronic process, with syntactic structures unfolding in real time (Hopper 2011: 27, see also Auer & Pfänder 2011: 3). The emergence of clauses as syntactic units can also be achieved through the participants engaging in mutual activity (Helasvuo 2001b: 33–35). 20 1 Introduction Another feature, which is connected to the temporal sides of interaction, is called ‘projection’. The potential course of linguistic structures is often anticipatory so that the recipient can interpret whether the utterance is complete or still unfinished. (For instance, see Auer 2005, Hopper & Thompson 2008: 116, Sacks et al. 1974: 702.) The temporal aspect does not only include the anticipation of linguistic structure of the turn, but the interpretation of ongoing action as well. The recipient must continuously monitor the relation of the ongoing turn to the previous conversation and be aware of how the turn affects the continuation of the conversation (for example, see Schegloff 1996: 100). My research does concern how speakers learn language. If we examine more closely the traditional Second Language Acquisition (SLA) research, we notice that the research is predominantly based on premises that learning is an action that emphasizes individual effort (Firth & Wagner 1997). Thus, when speakers encounter problems in interaction, they are often approached as a language-specific problem for the learner rather than a problem between interlocutors. This type of approach can be called a mentalist SLA perspective. (Kurhila 2006: 9.) For example, when studying word searches, second language studies have focussed primarily on the strategies that nonnative speakers adopt when encountering problems in their language productions rather than observing their difficulties as an interactional phenomenon (Kurhila 2006: 94). Lilja (2010: 30) defines this type of research tradition as mainstream and uses the term ‘individually-oriented second language acquisition research’. With special reference to lingua franca research, Hülmbauer (2009: 326) has argued that research concerning a lingua franca is interested in interactive language use, whereas interlanguage research concentrates on “linguistic development on the individual level.” 3 The aim of this thesis is to observe language and conversational structures as they occur in real conversation. That is, my goal is not to compare the language that is used by non-natives to native speakers’ language. In a traditional sense, I do not accept the assumption that the problems that arise in interaction are caused solely by the foreign language. In fact, the participants do not necessarily even pay attention to grammatically unsuitable forms (for example, see Kurhila 2006: 31 for the conversations involving both native and non-native speakers of Finnish). Kurhila (ibid. 132–133) states: “The NNS orients to its linguistic aspect: the 3 See also Ranta 2009. 21 1 Introduction aim of the search is to bring the target word to a grammatical completion. The NS, for her part, orients to the interactional aspect: she interprets the aim of the search as achieving a state of mutual understanding.” 1.3 The structure of the study This study is organised as follows: chapter 2 introduces previous research concerning lingua franca studies as well as research on diaspora and colloquial Russian. The key concepts of CA and Interactional Linguisticbased research are provided in chapter 3. Chapter 4 presents the data consulted in this study. Chapters 5, 6 and 7 will comprise an extensive analysis of the jointly produced structures in the Finnish and Russian data. In chapter 5 concentrates on the collaborative productions and provides an analysis of the general lexico-semantic, syntactic and prosodic features of the utterances. Chapters 6 is devoted to word searches and chapter 7 focusses on candidate understandings by analysing the linguistic and local sequential aspects of the phenomena. Chapter 8 will discuss the results and conclude the study. I will also outline the possible research subjects for the future. 22 2 Research on English as a lingua franca and non-standard forms of Russian In this chapter, I will first introduce the research concerning English as a Lingua Franca (subsection 2.1). I will then discuss the situation of the Russian language from different perspectives (subsection 2.2). The collapse of the Soviet Union had an immense impact on the Russian population who remained outside Russia; they are referred to as the Russian- speaking diaspora. Naturally, this diaspora has contacts with populations that use languages other than Russia. Finally, I will discuss the research on the colloquial Russian that is spoken in Russia. There are several for grouping together the English and Russian languages and I will discuss the research on them in the same chapter. As for lingua francas, English is the obvious number one because it is the most widespread and widely spoken lingua franca and most of the LF research has focussed on English (for instance, see Meierkord 2006 for historical background). Discussion concerning the use of ELF (‘whose language is it’ or ‘who owns the language’; see, for example, Haberland 2011, Seidlhofer 2001) has been rather polemic. On the one hand, the situation of the Russian language is somewhat similar to that of ELF, but on the other hand, Russian can be regarded as completely different. During the Russian Empire and the Soviet era, Russian was used in the grand duchies and in the Soviet Socialist Republics. During the Soviet era, Russian also played a significant role in other Eastern European countries. In the Soviet Union, Russian was the language of interethnic communication (jazyk mežnacional’nogo obščenija) or an intermediate language (jazyk-posrednik). Even though the significance of the Russian language has changed after the dissolution of the Soviet Union, the geographical enormity of Russia as well as the great number of minorities living within its boundaries and using Russian in everyday life make it clear that the question of Russian as a lingua franca is far from being solved. (For instance, see Daniel et al. 2010 for the language situation in Daghestan; Hakimov & Trusova 2010 in the Udmurt Republic; Vahtin et al. 2010 for an overall description of Russian ‘languages’.) Although the Russian language has been used as a language of interethnic communication or as an 23 2 Research on English as a lingua franca and non-standard forms of Russian intermediate language, research on the subject remains scarce (Mustajoki 2011:16–18)4. When taking into account the methodological approach adopted for this study, it becomes clear that previous research concerning the Russian language is somewhat limited. This is not to say that there are no studies concerning the use of colloquial Russian or the language form used abroad. However, Western scholars are usually not familiar with the findings of their Russian colleagues. This might be due to the language barrier: Russian scholars tend to write in Russian, and since Russian is not commonly spoken among Western scholars, Russian research is not accessible for them (however, see, for example, Mustajoki 2010). On the other hand, Russians might have a similar language barrier with respect to English or other Western languages, or they might not want to use English. This might, unfortunately, result in a situation where ideas never cross the language barriers and two or more scholars may address with same issues without being aware of each other. 2.1 English as a Lingua Franca I will now introduce research concerning English as a Lingua Franca. As noted earlier, the number of non-native English speakers has surpassed the number of native speakers of the ‘core’ varieties of British and American English (Mauranen & Metsä-Ketelä 2006: 1, see also Crystal 2003: 61, Seidlhofer 2001: 141). One way of drawing distinctions between the different usages of English is to describe them as belonging to three different circles. The inner circle includes the native English used in Britain, the United States and Australia. The outer circle, in turn, consists of countries in which English is spoken as a second language due to the countries’ colonial history (for instance, India, South Africa; see, for example, Foley 2006 for Singapore). Finally, the expanding circle comprises countries that do not have a history of colonisation with the members of the inner circle but which recognise the importance of English as an international language (for example, China, Japan, Israel, and the Scandinavian countries). (Kachru 1992 [1982]: 356, see also Crystal 2003: 4 Fedorova and Gavrilova (2010) examined how native Russian speakers interact with non-native Russian speakers and they proposed three categories. They concluded that while in St. Petersburg, native speakers tend to behave as hosts, on the Russian-Chinese border, native speakers served as bosses, and on the Finnish-Russian border, as pupils. As hosts, the Russians take responsibility and dominate the dialogue, as bosses they speak fast and use colloquial forms, and as pupils, they explain grammatical rules by referring to the uniqueness of the Finnish language (ibid. 52, 56, 60–61). 24 2 Research on English as a lingua franca and non-standard forms of Russian 60–61.) In this type of international usage, English can also be defined as an international language (for English as an International Language, EIL; see, for example, Jenkins 2000, 2002; Seidlhofer 2000). However, Jenkins (2006, 2009) has criticised the dominance of Standard British and Standard American English as the premise of language learning and the use of Received Pronunciation (RP) as the ideal. Regarding pronunciation, other issues also emerge, such as whether it is appropriate for a non-native speaker to use a regional accent (Jenkins 2009: 11)? Research on ELF began in the late 1980s. At first, the subject of study was English as an international language (EIL), but as early as in 1996, Firth first spoke of ELF. In 1998, Meierkord studied the discourse structure and politeness phenomena during small-talk conversations and compared them to native speakers’ conversations. On the one hand, the ELF employed by non-native speakers shared similarities with the standard varieties. Those similarities were speech-related features such as the length of turns and simultaneous speech. On the other hand, the ELF spoken in those conversations also exhibited characteristics that are generally attributed to learner language. For example, speakers tended to prefer safe topics such as meals or university classes. In order to establish a friendly and cooperative atmosphere, they also made use of laughter to substitute for verbal back channels. However, as Meierkord (1998) states: Conversations are built up collaboratively and speakers used a comparatively high amount of sentence completions and restatements. These characteristics do not result from interferences with the speakers' mother tongues. Rather, they are specific phenomena of this interlanguage lingua franca. The pragmatic studies conducted by Firth (1996) and Meierkord (1998) are predominantly descriptive and small-scale, but they provide important information on ELF. Jenkins’ (2000) study on the phonology of ELF is acclaimed as seminal and it has significantly affected ELF research of ELF. Jenkins’s study was the first major description of ELF and it is the first to consider ELF as a language form in its own right not only as a deficient form of Standard English. 5 To facilitate ELF research, several corpora have been collected. One of these is the Vienna-Oxford International Corpus of English (VOICE) that aims to provide a general basis for the analyses of ELF talk on all linguistic levels (Breiteneder et al. 2006; Seidlhofer 2001, 2005, 2011). The study of 5 Traditional views of language proficiency are still in effect. For example, if the rules of a language are somehow not followed in the conversation, the evaluation depends on the speaker. Thus, native speakers are merely bending the rules, whereas non-native speakers break them. (for example, see Prodromou 2007: 23.) 25 2 Research on English as a lingua franca and non-standard forms of Russian ELF in Finland has focussed on the ELF that is used in academic settings. The Department of English Philology at the University of Helsinki has collected a corpus of this type of language use, The ELFA Corpus (see, e.g. Mauranen 2006b, 2009b; Mauranen & Metsä-Ketelä 2006). One current, heated debate specifically concerns academic ELF. Although a large number of international universities use English, university language policies and practices are governed by national English (British and North American) norms (Jenkins 2011: 926). This is evident, for example, in the academic journals in which “non-native English academics are obliged to have their English ‘corrected’ by native speakers before their articles are accepted for publication in international academic journals” (ibid. 928). In journals specializing specifically in ELF, other policies are implemented (cf. Mauranen & Metsä-Ketelä 2006: 6). According to Mauranen & Metsä-Ketelä (ibid.), this means that although native speakers have the right to submit their articles to the journal, they are not assumed to have the ultimate authority regarding linguistic correctness or comprehensibility. Thus, the contributions of the non-native users of ELF have not been proofread by native speakers. Suviniitty (2012: preface) stated in the preface of her dissertation that her thesis did not undergo language revision even though it was highly recommended by the university. Suviniitty (ibid.) argued that the principle of language revision by a native speaker of English would have collided with the whole notion of ELF. There has been a strong research history on the pragmatic features of ELF (for example, see House 2002, Mauranen 2006b). Mollin (2006) has considered ELF in relation to Kachrus’s circles. However, ELF is not usually included in the Expanding Circle. The problem of ELF is that it appears to lack coherent features that delineate it from other varieties. However, ELF might be conceptualized as a functional phenomenon that emerges from the strategies of lingua franca communication. (Ibid. 55.) If structural similarities were to be found in ELF, this would imply that “the Kachruvian model would have to be re-written at the level of varieties and standards, granting the Expanding Circle the same right to its own standard as the other circles” (ibid. 43). As Mollin (2006: 45) points out, speakers of ELF do not necessarily use language daily with the same people. What then follows is that the speaker community can be constantly in flux rather than remaining stable and fixed (see also Seidlhofer 2011: 81–88). Indeed, speakers need to consider different linguistic and cultural backgrounds and to accommodate their 26 2 Research on English as a lingua franca and non-standard forms of Russian speech accordingly. However, is it even possible to negotiate a common standard in these conversation situations that are ever-changing? Mollin (2006: 45) hypothesizes that the common features of ELF would be rather surprising. The results of her study confirm this hypothesis. Firstly, speakers tend adhere to the standard usage conventions of native-speakers. The errors that they make therefore depend on their mother tongue and their English competence. In brief, the errors by the ELF speakers were individual; there were very few common features that united lingua franca speakers. (Ibid. 48.) Mollin’s position is in direct contrast to other studies that have found that similarities occur. One of the interesting features is the controversial thirdperson singular –s, which is being used somewhat differently than it is in Standard English (Cogo & Dewey 2006). The present study combines both pragmatic and lexicogrammatical features and demonstrates that the non-use of –s depends on both the type of verb that it is attached to and on situational parameters (ibid. 77). Furthermore, the zero –s is more likely to occur on main verbs, whereas on auxiliary verbs, –s is more likely to be present. If the situational parameters are taken into account, the use of zero is a result of the efficiency of communication and exploited redundancy. (Ibid. 87.) Mollin’s (2006) and Cogo & Dewey’s (2006) studies reveal something essential of ELF and its speakers. Even though the results of those studies tend to contradict each other, they emphasise variability in terms of speaker proficiency, and openness to an integration of forms of other languages. Lingua franca interactions are flexible in the sense that they allow participants to negotiate the language use and to take into account different backgrounds. (Jenkins 2003.) In addition to the above-mentioned alternation between zero and –s, other features also commonly occur in ELF. Those features include confusing the relative pronouns who and which, and omitting the definite and indefinite articles where they are obligatory in Standard English and inserting them where they do not occur. Speakers also tend to insert redundant prepositions and overuse certain words of high semantic generality (do, have, make, etc.). (Björkman 2009: 231-236; Dewey 2009: 63; Meierkord 2004: 121; Seidlhofer 2004: 220, 2011: 107–109.) One of the interesting features is the frequent and divergent use of the – ing verb form (Ranta 2006). Non-native speakers often overuse this ‘attractive’ progressive (ibid.) and it has been described as a problem for non-native speakers in their second-language acquisition studies. However, 27 2 Research on English as a lingua franca and non-standard forms of Russian as Ranta (ibid. 112) observes, the overuse of the progressive did not create trouble in the communication in the data. For this reason, it might not even be justified to mention the overuse of the progressive because the results presented by Ranta suggest that the vast majority of the uses resembled the standard users. Thus, on-native speakers do know the semantics and use it appropriately most of the time. (Ibid. 111). Moreover, the ‘attractiveness’ of the progressive is connected to the grammatical form itself. Ranta (ibid. 112) states: What I mean by this is that adding the ending -ing and the auxiliary BE to a verb (any verb for that matter) gives the verb more prominence and salience in the speaker’s utterance. It makes the verb stand out, so to speak, and draws the interlocutor’s attention as a ‘heavier’ periphrastic structure. Ranta (ibid. 100) also discovered that this feature is attested in the English of speakers who have typologically very different first languages and this makes it a candidate for a true ‘ELF feature’ and this calls to question the common assumption that this usage of the progressive is either a transfer from the speaker’s mother tongue, or a result of the instruction that the speaker has received. As mentioned earlier, when non-native speakers communicate, they do not encounter more misunderstandings or problems than in the communication between native speakers (Mauranen 2006a, Meierkord 2004, Thije 2003). In her recent dissertation, Suviniitty (2012) examined the lectures conducted in ELF that were offered in the International Master’s Program- A thorough analysis of the transcripts and surveys revealed that ‘broken’ or ‘bad’ English does not have a negative impact on a student’s reception of teaching. It is more important to note that teachers currently use interactional features during lectures that encourage the students to participate during their lectures (Suviniitty 2012: 13). For the past twenty years, ELF has emerged as a prestigious field of research, but its studies and results have not achieved universal approval. Some researchers are concerned that the use of ELF and non-standard forms will ultimately influence English as a native language (ENL) and consequently lead to a deterioration in the English taught as a foreign language. This position maintains that only a monolithic and standardised model of English ensures comprehension among those who speak English. Some researchers, however, claim that the English language belongs equally to those who use it as a foreign language as well as to those who speak it as a mother tongue (Widdowson 1994). 28 2 Research on English as a lingua franca and non-standard forms of Russian The situation of the English language is somewhat divided. Firstly, English has been considered as a threat to local languages and cultures. Indeed, English is often the first and only foreign language that someone learns. In theory, people know foreign languages, but they only know the same languages. Secondly, as was stated earlier, the global use of English might threaten Standard English. (Mauranen 2009b: 1.) However, the question is how this will affect native speakers. Will they have to learn foreign languages, or is their mother tongue sufficient? One could also argue that English can be a restrictive factor in their lives because native speakers of English can form their own identities as monolinguals. (Chesterman 2005.) More generally speaking, multilingualism as we know it today is not a new tendency. According to one estimation, “at the end of the 20 th century, one or another form of multilingualism affect 60 per cent of the world’s population” (Lüdi 2006: 20). According to Lüdi (ibid.), monolingualism can be conceived of as a boundary case of the multilingualism that originates from very specific cultural conditions. For most people, English is a visible lingua franca that is used globally in the sense that it surrounds us all the time.6 People might also use other languages as lingua francas, but they are used more locally or are used in certain spheres (for example, French in the European Union; Wright 2006; or German in the Middle Ages; Darquennes & Nelde 2006). In addition, lingua franca usage might pose a threat to Standard English, but on the other hand, using English as a lingua franca also ensures that there are other speakers in addition to native speakers. ELF might also replace the language that previously has been used as a lingua franca. For example, Russian has lost its status in Eastern Europe as a language of inter-state communication. This is due to the former Soviet Union and its language policy that led to conflicts between the native 6 This visibility may also indicate that there is a need for the varieties of English as a global language to be more accurately described. Phillipson (2008: 250) uses the following terms: English as “a lingua economica (in business and advertising, the language of corporate neoliberalism), a lingua emotiva (the imaginary of Hollywood, popular music, consumerism, and hedonism), a lingua academica (in research publications, at international conferences, and as a medium for content learning in higher education), or a lingua cultura (in the literary texts of English-speaking nations that school foreign language education traditionally aims at, and integrates with language learning as one element of general education). English is definitely the lingua bellica of wars between states (aggression by the US and its loyal acolytes in Afghanistan and Iraq, building on the presence of US bases in hundreds of countries worldwide). The worldwide presence of English as a lingua americana is due to the massive economic, cultural, and military impact of the USA.” Phillipson also acknowledges the harm that monolingualism brings to linguistic diversity. Language policies tend to favour lingua frankensteinia (emphasis original) and this leads to linguicide. (Ibid. 251.) 29 2 Research on English as a lingua franca and non-standard forms of Russian languages and Russian. Russian was treated as a hegemonic language that could have suppressed local languages. (House 2003: 561.) The replacement of earlier lingua francas with ELF might result in another interesting consequence that is rather paradoxical consequence. The spread of ELF may encourage members of minority languages to cherish their local languages and as a result, strengthen local varieties and cultural practices. (Ibid.) For those of us who use the English language as a lingua franca, it can be only a means of communication. It is a foreign language that does not bear negative connotations for Finns, perhaps because Finns have not been colonialised by countries from the inner circle. The choice of which language we use might be influenced by the idea that some languages are regarded merely as a means of communication while others are a means of constructing one’s identity (House 2003: 559–562, see also Mustajoki 2011). Generally, native languages are considered to be connected to identity and lingua francas to mere communication. Lingua-franca interactions may remain neutral and equal, because ELF is often nothing more than a functional tool without a heavy emotional burden. However, ELF is also used in more intimate intercultural relationships and friendships, therefore making it safe to assume that people do use ELF to show signs of affection and dislike, too. 2.2 Diaspora Russian and Colloquial Russian As mentioned earlier, the status of the Russian language has changed since the collapse of the Soviet Union. However, the present situation varies among the countries and regions where Russian is spoken. According to the HSE Institute of Demography (IDEM), in 1989, when the last census of the USSR was conducted, the total population was approximately 286 million. Of these, an estimated 145 million were Russian by nationality, which constitutes 50.8% of the population. However, owing to the language shift, 57.2% of the population spoke Russian as their mother tongue 7. The first official census of the Russian Federation was conducted in 2002. According to this census, the population of Russia was approximately 145 million people, of which 116 million, or 80%, were Russians. 8 In other post-Soviet republics, the number of Russian speakers has declined. According to the State Statistical Department of Russia, in 1996, less than 23 million ethnic Russians remained in the Near Foreign countries 7 8 http://demoscope.ru/weekly/ssp/sng_nac_89.php, accessed 18.1.2013. http://demoscope.ru/weekly/ssp/rus_nac_02.php, accessed 18.1.2013. 30 2 Research on English as a lingua franca and non-standard forms of Russian (Zevelev 2001: 117–118). Moreover, the Near Foreign countries have adopted different types of language policies and practices. Following the 1995 referendum, Belarus is thus far the only country that has upgraded Russian to an official language (Aref’ev 2012: 40–48, Zevelev 2001: 106). The policies range from the aforementioned Belarusian policy to the legislation of the Baltic countries, which determines that the titular language is the main language of the country (Alpatov 2003). Other countries are positioned more or less between these two extremes. For example, in the late nineteen-nineties and in the early decades of the twenty-first century, Uzbekistan went through a series of reforms. The Russian language was removed from streets signs as well as from signs showing place names. In addition, Russian was deleted from the names of institutions, and the official language was transferred from the Cyrillic alphabet to the Latin alphabet. The status of Russian likewise began to change in 2005 when Russia and Uzbekistan signed a strategic partnership. (Aref’ev 2012: 120–121.) For the Baltic countries, the term derussification has been used to refer to the tendency of Estonian, Latvian, and Lithuanian societies to replace the Russian language with their titular languages, in the areas such as government, administration, education, and science. In some cases, the process of derussification is being accelerated (Alpatov 2003: 428). The status of the Baltic countries is somewhat different from other post-Soviet countries. Pavlenko (2006) writes: This derussification course – unlike that taken by any other post-Soviet country – is best understood through a combination of historic and sociopolitical factors. Annexed by the Soviets in 1940, Estonia, Latvia, and Lithuania already had an experience of independent statehood and returned to it after a little more than four decades of the Soviet rule. They had spent the least time in the Soviet Union and their citizens had high levels of national consciousness and language loyalty, vehemently opposing russification. These states also chose to orient themselves toward the West economically, socially, and politically. However, as Pavlenko (2006) points out, the initial stages of this derussification process were the most dramatic and painful for the Russianspeaking population. Today independent states understand and acknowledge the usefulness of the Russian language, and Russian has regained its previous status as a language of interethnic communication. Russian culture, both high and popular, also plays an important role in these countries and contributes significantly to the process of maintaining the Russian language. As mentioned before, Belarus has upgraded the status of the Russian language by adopting it as an official language. The situation of the Belarusian language is therefore peculiar in comparison to the other Slavic 31 2 Research on English as a lingua franca and non-standard forms of Russian languages. Taking into account the number of speakers of Belarusian, it may be regarded as a middle-sized Slavic language. However, as Zaprudski (2007: 98) observes: In spite of the considerable demographic potential of the Belarusian language in comparison with some other Slavic languages, there are indicators that are not as positive. There are many important communicative domains in which Belarusian is either not used at all or only to an insignificant degree. These include science, higher education, the armed forces, legal proceedings, large-scale business, legislature, and so on. Given its communicative power, Belarusian in Belarus undoubtedly yields to Russian in many key spheres. The Belarusian language has undergone a series of russification processes that have affected aspects such as the orthography of Belarusian (ibid. 108– 109). The resulting mixed code that includes both Belarusian and Russian elements is referred to as trasjanka, which literally meansʻa mixture of hay and straw’ (ibid. 111).9 In the European Union, Russian is the fifth most widely spoken foreign language (Eurobarometer 243: 5). Especially television programs and radio news in Russian are understood in those European Union members that were previously part of the Soviet Union (Lithuania, Latvia, and Estonia; ibid. 30). However, generally speaking, the knowledge of Russian has declined in countries such as Bulgaria, Slovakia, Poland and the Czech Republic (ibid. 16). In those countries, Russian is no longer taught at school to the same extent as it was after World War II (ibid. 142). In addition to the above-mentioned Near-Abroad Russian diaspora, the ethnic Russians have also mass migrated abroad. There has also been an interest in studying and observing that language usage (Glovinskaja 2001, Zemskaja 2001). However, these studies often concentrate on those features that indicate, for example, the degree of preservation or loss that the Russian language has undergone (Zemskaja 2001; see also Avina 2000 & 2002 for the Russian language in Lithuania; Dubinina & Polinskaja 2013 in the USA; Zhdanova 2012 in Germany). When the social and historical background information of the diaspora speakers is taken into account, one could use the term ‘heritage speakers’ to refer to them. The term ‘heritage language speakers’10 refers to people who are raised ‘in a home where one language is spoken who subsequently 9 In the Ukraine, the mixed code is called suržik, which literally means the ‘flour or bread made from mixed grains’. 10 The term can be translated into Russian using different types of words or structures. One possibility is to use referents such as domašnyj jazyk, ‘home language,’ or semejnyj jazyk, ‘family language’. With special reference to Russian heritage speakers, Dubinina & Polinskaja (2013: 13–14) suggest the term ljudi s unasledovannym russkim which translates,ʻpeople with inherited Russian’. 32 2 Research on English as a lingua franca and non-standard forms of Russian switch to another dominant language’ (Polinsky & Kagan 2007: 368). This research has predominately focussed on English, meaning that the dominant language examined in the research is usually English and the heritage language could be, for example, Russian (ibid.; see, however, Kopeliovich 2010 for the situation in Israel). One of the main features of heritage languages is that they have not been completely acquired or they are affected by the dominant language (Kopeliovich 2010: 407–411). Heritage languages are used at home, which means that the speakers must switch to the dominant language when they are outside family settings. Of course, the knowledge of the heritage language varies among its speakers, and one must also consider the enormous variation within the language itself. (Polinsky & Kagan 2007: 369–370.) Observing the structure of the language of the Russian diaspora, Zemskaja (2001: 200–208) has analysed the increase in analytic structures that occur in the speech of émigrés. These analytic structures include aspects such as indeclinable nouns and adjectives. However, the indeclinability of nouns and adjectives is somewhat dependent on the individual who is speaking the language. For instance, the noun kèš, ‘cash’, is usually indeclinable, but some speakers may decline it (ibid. 202). Another interesting phenomenon is the mixing of words that have a similar pronunciation, but that sometimes overlap with the meaning (Glovinskaja 2001: 458–459). Speakers confuse two words that sound very similar: uže – už, ‘already – so’. Research has focussed primarily on the actual language and its forms, but the use of language as a part of naturally-occurring interaction has attracted less research interest. One of the areas that has been studied is functional sentence perspective (abbreviated as FSP). The term FSP means that a sentence structure is analysed in terms of textual elements rather than grammatical elements. Textual elements include, for example, the theme– rheme structure. In practice, sentence structure analysis is more or less identical with word order analysis. (Hakulinen & Karlsson 1995: 41.) The theme of a sentence refers to the subject of discussion while the rheme refers to what is being said about the subject. Glovinskaja (2001: 424–426) observed that diaspora speakers tend to produce erroneous structures including both theme and rheme. Typically, these mistakes involve an unusual word order. For example, the rheme might be placed incorrectly (the borderline between the theme and the rheme is marked with //): kvartira emu // prinadležala, ‘the flat him // 33 2 Research on English as a lingua franca and non-standard forms of Russian belonged’. In this case, the theme is kvartira emu and the rheme, prinadležala. The appropriate order of the words would be kvartira prinadležala // emu, ‘the flat belonged // to him’. The sentence presupposes that there was a flat and it belonged to someone, and the new information is that it belonged to a certain man. Due to the close geographical and historical relations between Finland and Russia, Finland has also played, and continues to play, a significant role in the Russian migration as well as the research on language use (see, for example, Protasova 2004 for an overview of the Russian migration to Finland). The former USSR and Russia comprise the largest group of foreign origin among the Finnish population (OSF). 11 The Russian migration has occurred in waves. Traditionally, five waves are considered to have taken place (Protasova 2004: 6–7). The first wave occurred before Finland became a part of the Russian Empire. During the Grand Duchy of Finland 1809–1917, Russian soldiers, officials, merchants and priests immigrated to Finland. Many of those first- and second-wave families are still recognised in Finland, such as the brewery family of Sinebrychoff. Finland became a transit country for many people who fled Russia after its two revolutions. Some of them continued their flight to Prague, Berlin and Paris, while others remained in Finland. The fourth wave occurred during World War II when Ingrian Finns along with some representatives of other nationalities migrated to Finland, but most of them were sent back to the Soviet Union after the war. Between World War II and the dissolution of the Soviet Union, immigration consisted predominantly of those Russians and other those of nationalities of the Soviet Union who married Finns. This type of immigration is not usually considered to constitute a specific wave. The latest wave began in the early 1990s when the remigration of the Ingrian Finns became possible. Most of these Ingrian Finns use Russian as their mother tongue. However, the remigration queue was closed in 2011. 12 In addition, other Russian speakers have immigrated to Finland from Russia, Estonia as well as from other parts of the former Soviet Union. Research has thus far focussed on the morphological and morphosyntactic features of Finland Russian. Leisiö (2001) has studied the convergence and integration in the morphosyntactic aspects of Finland Russian. The departure point for Leisiö’s research is that language change 11 It is not a surprise that the same observation is also valid when considering the informants of this study. The largest group consists of Russians and apart from Finnish, Russian is also the most used language in the word-search sequences. 12 Mononen 2013. 34 2 Research on English as a lingua franca and non-standard forms of Russian that is contact-induced entails a process that is affected by an interplay of social, linguistic and interactional factors (ibid. 15). For example, the Finnish and Swedish languages affect the word order of the Russian genitive construction, and this influence is most clearly discernible in the sphere of the determiner genitive, and more specifically, the genitive of the possessor and the genitive of the subject (ibid. 96). Kaivapalu (2005) examined how a source language (in this case, Estonian and Russian) can contribute to foreign language acquisition, concentrating on morphological aspects, especially on the inflection of the plural forms of Finnish nouns. Estonian learners used two strategies when producing the plural forms of Finnish nouns: analogy with the target language and analogy with the source language. Russian learners began their inflectional process by applying rules, only using analogy when processing by rules did not succeed. One of Finland’s neighboring countries, Estonia, has a different historical background with Russia. In the 20th century, Estonia was independent for a while between the world wars, but after World War II, it became a part of the Soviet Union. Estonia regained its independence in 1991. A fourth of its population is Russian speaking, including both the people who have lived in the area of present-day Estonia for hundreds of years (so called Old Believers) and the people who either immigrated voluntarily to Estonia or were forced to immigrate due to population transfers. (Hennoste et al. 1999, Verschik 2007.) Since Finnish and Estonian are kindred languages, albeit grammatically somewhat different, the somewhat forced language contact between Estonian and Russian is an interesting phenomenon (see, for example, Verschik 2004, 2006; Zabrodskaja 2009). Contact between these languages has been two-way so that contact-induct changes have occurred in both languages. For example, the system of local cases in Estonian has had an impact on the Russian language (Verschik 2006). Estonian and Finnish, have three interior and three exterior cases to express directional, static and separate meanings. Russian has no specific category of local cases. Thus, prepositional phrases are used to express spatial relations. (Ibid. 388.) The respondents in Verschik’s study claimed that some Russians might say, kupit’ iz aptek-i, ‘to buy out drugstore-GEN’ (ibid. 398). However, the appropriate grammatical form in the Russian language would be kupit’ v aptek-e, ‘to buy in the drugstore-PREP’. The genitive case in the erroneous Russian structure 35 2 Research on English as a lingua franca and non-standard forms of Russian is equivalent to the Estonian structure osta apteegi-st, ‘to buy drugstoreELA’. Verschik has also conducted studies in Tallinn on receptive multilingualism (RM). Although Finns as tourists are not always cordially welcomed by their southern neighbours, their importance to the Estonian economy is significant. Receptive multilingualism refers to cases ‘in which interlocutors use their respective mother tongue while speaking to each other’ (Zeevaert & ten Thije 2007: 1). RM has been studied, for example, in Scandinavia and Switzerland (for instance, see Doetjes 2007, Werlen 2007). The Scandinavian languages are closely related to the Germanic languages, whereas in Switzerland, the situation is somewhat complex because it entails both the Germanic languages and Romance languages. In the FinnishEstonian context, receptive multilingualism results in Finnish customers and Estonian salespersons using their own respective mother tongues that are mutually intelligible to some extent, but differ from each other rather significantly (Verschik 2011). In the actual conversational situation, the Finnish and Estonian participants have multiple ways of making their language use comprehensible and they produce forms that do not belong to either of the monolingual varieties (ibid. 269). A possible word formation can consist of an Estonian or Finnish stem to which Finnish or Estonian inflectional morphology is subsequently added (ibid. 278). Although Estonian salespersons are more likely to accommodate their speech to make it sound more like Finnish, Finnish tourists do not always speak only monolingual Finnish, either (ibid. 280). In fact, Finns are usually aware of the differences between Finnish and Estonian and they might be cautious of using only monolingual Finnish. Following the dissolution of the Soviet Union, many Russians returned to Russia. For example, from the years 1990 to 1997, approximately 2.4 million Russians returned to Russia from what is called Near Foreign countries (examples are Central Asia, Transcaucasia; Zevelev 2001: 122). However, one cannot assume that this type of repatriation could occur on a larger scale. Many of the ethnic Russians living in the Near Foreign countries in the diaspora were born in those countries, have lived their more than ten years, or they have intermarried (Avina 2000: 52). As Zevelev (2001: 123) states this succinctly: There is no reason to believe that most Russians will ‘return’ to their ‘historic homeland.’ Many Russians consider Ukraine, Belarus, Kazakhstan, or the Baltic States to be their homeland because several generations of their ancestors lived there. Besides, there are other powerful factors restraining migration, such as a 36 2 Research on English as a lingua franca and non-standard forms of Russian lack of adequate affordable housing in Russia. Therefore, one may conclude that Russian diasporas will remain an important feature of all newly independent states of Eurasia for the foreseeable future. This new linguistic situation requires new approaches because it complicates the research area. For example, to what extent do Russians remain Russians with respect to their language use and traditions? Or, is it even necessary to adhere to those concepts? However, the situation needs to be observed with the help of large studies that combine linguistics (e.g., contact linguistics) and area and cultural research. 13 For example, the future of Russian language can be twofold in Central Asia. On the one hand, the language might be preserved but on the other hand, changes in its form and grammar might lead to consideration whether a new variety of Central Asian Russian language will emerge (Podporenko 2001). The research of the colloquial Russian language that is used in Russia has focussed on describing the different regional varieties or the language that are spoken by different social groups (Zemskaja et al. 1981, see also Mustajoki 2013a for an overview). One of the research areas of colloquial Russian is the language form that occurs in cities. The term used for this type of language use is ‘prostorečie’ or ‘gorodskoe prostorečie’, which can be translated as. ‘urban vernacular’, ‘common speech’ or ‘colloquial language’. However, this term can be defined more broadly as a form of the Russian language that is used by uneducated strata of society (D’jačok 2003: 103). If the status of this colloquial language is compared to the literary language, it can be said that it is non-normative and socially constrained urban speech that is situated outside of the literary language (Kapanadze 1984: 5). Unlike regional varieties or dialects, this urban speech does not have a particular region or locality where it is used. In contrast to the literary language, this language form differs in many respects. Urban speech has different phonetic and lexical characteristics, such as different intonation patterns, the use of curse words, and obscene vocabulary. (D’jačok 2003: 105–106.) 13 The above-mentioned Kachruvian Circles could also be applicable to the situation of the Russian language. Inner circle includes native Russian used in Russia and, at least to some extent, in Belarus and Eastern Ukraine. The Near Foreign countries belong to the outer circle countries that have been a part of the Russian Empire or the Soviet Union. The expanding circle includes those countries that acknowledge the economic importance of the Russian language. This approach, however, requires additional research. For example, considering the history of Finland, it could be part of the outer circle, but it currently belongs to the expanding circle because the status of Russian was diminished after Finland became independent in 1917. 37 2 Research on English as a lingua franca and non-standard forms of Russian To facilitate the research on spoken Russian, corpora are also available. For example, the Russian National Corpus 14 also includes a corpus of spoken Russian. This corpus consists of speech that is both public (52%) and private (10%). The corpus mostly consists of transcripts of films. (Grišina & Savčuk 2009: 134.) Lexical, morphological and semantic searches are possible and an individual researcher can also construct a subcorpus by using sociological parameters (ibid. 137–138). In St. Petersburg, the researchers have collected a vast amount of spoken material in the framework of a project entitled Odin rečevoj den’, ‘One speech day’. Basically, the informants of the corpus had a dictaphone on for a whole day. Therefore, a range of communicative situations were taped. (Asinovskij et al. 2010.) These situations included shopping, doctor’s appointments, cafés, bars, hobbies, as well as other contexts (Bogdanova et al. 2010: 51). Diaspora Russian shares some characteristics with the above-mentioned urban vernacular. Both are considered to be the uncodified domain of the spoken urban speech (Zemskaja 2001: 207). However, there are also some major differences between them. The speech of the Russian immigrants contains indeclinable nouns and adjectives that are not used in Russia. Speakers of urban vernacular decline words that are indeclinable in the literary language, such as declining the word pal’to, ‘coat’, even though it is indeclinable: bez pal’ta; ibid.) It is also possible that old émigrés can have certain elements in their speech that could be characterised as resulting from the urban vernacular. This could be due to their visits to Russia after the dissolution of the Soviet Union. (Zemskaja 2001: 254–255.) As mentioned previously, the research conducted in CA or the research on the Russian language from an interactional linguistic perspective has not acquired a lot of attention. Makarov (2003) presents the key concepts of conduction research on spoken discourse. In addition, Isupova (2002) introduces the basic methods of CA as well as the techniques that are used in CA-based research. These books and articles are written in Russian, but they do not cover spoken discourse in Russian. Nevertheless, there are some interesting studies concerning CA-based research on the Russian language in conversation settings. For example, Paukkeri (2006) analysed the particles da, ‘yes’, nu, ‘well’, and tak, ‘so’, as the actions by the recipients in spontaneous Russian conversation. Paukkeri concluded that the particles ‘nu’ and ‘tak’ were used in an informing context, while ‘da’ was used in affective contexts. Bolden (2004) examined the onset and the offset of 14 http://www.ruscorpora.ru/en/index.html 38 2 Research on English as a lingua franca and non-standard forms of Russian reported speech that are marked in ordinary talk. In her analysis, Bolden discovered that in most cases, quotations are separated from other talk for many reasons. The separation occurs at the beginning of quotes and at the end. This is achieved through a variety of devices including grammatical framing, re-anchoring devices, prosodic shifts for their onset as well as several repositioning devices and sequence organisation practices for their offset. 39 3 Previous research and key concepts This chapter will present an introduction to the key concepts for jointly produced structures. Subsections 3.1 and 3.2 specify the research questions of the present study and construct the ways that the data are analysed in the study. Subsection 3.1 is devoted to collaborative productions and subsection 3.2 concentrates on word searches. 3.1 Organisation of turn-taking in relation to collaborative productions This chapter is devoted to an overview of how turns in a conversation are constructed as well as how they are allocated in interaction. When two or more participants produce a structure together, this chapter examines how they accomplish this and how they orient to this process. Thus, my aim is to define my research questions from a broader perspective, to specify them and explicate how the selected methods and approaches are appropriate for this type of analysis. The focus of this kind of analysis is on the organisation of turn-taking (Sacks et al. 1974). The basic unit of turn-taking is the turn-constructional unit (TCU), which refers to sentential, clausal, phrasal or lexical constructions (ibid. 702). A turn can contain one or more TCUs. TCUs can constitute potentially complete turns and on their possible completion, a change in speakers can become relevant (Sacks et al. 1974: 702–704; Schegloff 1996: 55; Selting 2000: 478, 480). A change in speakership depends on a complex interplay of grammatical, prosodic and pragmatic cues (Couper-Kuhlen & Selting 1996; Ford & Thompson 1996; Ford et al. 1996; Sacks et al. 1974; Schegloff 1996; Selting 1996, 2005). A projection of the possible turn completion in conversation is thus crucially dependent on cues that are predominantly linguistic in nature (Couper-Kuhlen & Selting 2001: 5–6). The linguistic units of a conversation are defined as a part of interaction and their boundaries are negotiated throughout a conversation. For this reason, an application of traditionally syntactic terms (such as the clause or phrase) is not straightforward means of describing those units. Instead, the terms used for these purposes are based more on dialogue and on context than on pure syntactic premises. Those terms include turn, TCU and utterance. For example, when a change in speaker occurs, the ongoing turn 40 3 Previous research and key concepts has reached its end. Therefore, a turn is defined by a speaker change. (VISK § 1004.) Koivisto (2011) focussed her analysis on conjunctions in her dissertation and she differentiates a twofold between a turn and a TCU (ibid. 26). While a turn is something that can be observed retrospectively as a part of interaction, it is a materialized contribution of interaction (see Goodwin 1981: 20, Selting 2000: 511). Nonetheless, a TCU is a hypothetical, possible turn (see also Schegloff 1996: 55). The boundaries of a turn are defined by way of organisation of turn-taking, whereas a TCU is interpreted as completed if certain sequential and functional criteria are met. A TCU has to be an appropriate contribution in context. (See also Sacks 1992b: 43.) For example, Koivisto (2011: 29–30) defines her scope of research that she studies those TCUs that end in a conjunction. After this conjunction, the transfer of speakership can occur, but it is not necessary. The same speaker can also continue his or her TCU. A change of speakership can occur at the end of every TCU, which is called a ‘transition relevance point’ or TRP. One of the essential factors that influences the organization of turn-taking is syntax. The ending points of different syntactic structures are also possible turn completions. (Hakulinen 1997: 42; Sacks et al. 1974: 704, 721.) In one of the groundbreaking studies, Oreström (1983: 68) established five linguistic features that operate as signals of the possible completion of a turn. Those features are prosody, syntax, semantics, loudness, and the silent pause. The combination of prosodic, syntactic and semantic features forms a major juncture for a change in speakership. (See also Ford & Thompson 1996: 153–154, Tiittula 1985 for Finnish conversations.) Furthermore, turn size is usually locally managed as well as interactionally achieved therefore the turn size is not prearranged (Sacks et al. 1974: 700–701). Turns are allocated according to certain principles and rules. One of the most important rules is that turns are coordinated to minimise the gap and overlap so as to make the transition smooth. Three rules can be applied during conversation. Firstly, the current speaker can select the next speaker, in which case the selected speaker has the right and is obliged to take the next turn. Secondly, if the current speaker does not select the next speaker, self-selection occurs. Furthermore, whoever begins to talk first acquires the right to a turn. Thirdly, if the turn-so-far does not involve any of these scenarios, the current speaker may continue. If the first two rules have not been in effect at the initial transition-relevance place, and the current 41 3 Previous research and key concepts speaker has continued his or her turn, then these three rules are re-evaluated at the next transition-relevance place until the transfer has occurred. (Sacks et al. 1974: 704, see also Duncan 1972.) It is also possible for the firstspeaker’s turn to be interrupted by another speaker, so that the first speaker can produce a delayed completion (Lerner 1989). A point of transition relevance can be projected. At the beginning of a turn, projecting the TRP is often difficult or even impossible, but it becomes easier when the linguistic structure of the turn-in-progress becomes more apparent. Moreover, different languages differ in terms of the type of projections they allow. For example, English allows early projection, whereas Japanese does not (Thompson & Couper-Kuhlen 2005: 485–488). If we examine the notion of projection more thoroughly, it can be understood as an individual action or part that foreshadows another. In order to understand what is being projected, participants need some degree of knowledge about how actions (or action components) are typically sequenced, that is, how they follow each other in time. (Auer 2005: 8.) Auer (ibid.) continues on describing limitations that apply to projections: Each projection has a time span. [---] An action (or action component) may project onto the timing slot immediately following it and make some next activity (component) expectable in this slot. But it may also allow other things to happen ‘in between’, before the projected unit legitimately can or should occur, and it may project more than one ‘next’ in a sequence. While it might seem highly likely that projection is used to complete utterance completing, projection might also be used for other purposes. For example, focussing on French, Chevalier and Clift (2008) have reported that participants in French conversations do not necessarily complete unfinished turns. Instead of completing the turns, ‘recipients provide responses fitted to the unfinished turns, instead of completing the turn’ (ibid. 1736). In such cases, the projection is not dismantled. This is possible due to the French syntax. Traditionally thinking, French is a Subject + Verb + Object (SVO) language (as are Finnish and Russian), but the direct and indirect object pronouns have to be placed before the verb. In other words, the verb occurs at the end of the sentence, resulting in the number of possible verbs becoming narrower. (Ibid. 1735.) It is also possible that the current speaker uses incomplete structures on purpose, to indicate that the turn has not yet reached a transition-relevance place and to reserve the turn for him- or herself (Sacks et al. 1974: 702–703, 719–723; Schegloff 1996: 92–93; Routarinne 2003: 23–24). The possible sentence can also become rather long and the syntactic projection can extend 42 3 Previous research and key concepts far. However, on the syntactic level, only the possible sentence can be projected. For example, in multi-unit-turns, stories have to be initiated by announcements or story prefaces. Thus, syntax can be used as a turn-holding device only concerning the unit under way. (Local & Walker 2004, Schegloff 1982: 75–77, Selting 1996: 374–375.) Prosody as a means of managing the organisation of TCUs and turns is used locally, that is, to project and contextualise relations between items (ibid. 378). This does not mean, however, that syntax and prosody can be placed in a hierarchical order, in which syntax is number one and prosody number two. In some cases, prosody might have differentiating power. (Ibid. 384.) Projection is connected to the temporal sides of interaction. Thus, the course of linguistic structures is often anticipatory and the completeness of utterances can usually be interpreted. However, the strength of projections varies. From the perspective of sequentiality, the strongest projection occurs in adjacency pairs. (Auer 2005: 16.) Adjacency pairs somewhat resemble collaborative productions, but the pairs are produced in a slightly different manner. Adjacency pairs consist of two parts, the first pair part and the second pair part, which are pronounced by two participants (Schegloff & Sacks 1973: 295–296). These parts are adjacent, and the type of the first pair part is relevant to the form of the second pair part (ibid.). Typical examples of adjacency pairs include the exchange of greetings as well as sequences that include a question and response (Duranti 1997: 251–252, Raevaara 1997: 76). The similarities and dissimilarities of collaborative productions and adjacency pairs have not been studied systematically or in depth. Some differences are observable regarding the syntactic and prosodic features of the utterances. If we take a closer look at the syntax, we can see that after the preliminary component, collaborative productions are syntactically unfinished and they require a final component to be syntactically complete. In most cases, the prosodic pattern of the utterance is also not completed following the preliminary component. In adjacency pairs, the first-pair part is usually syntactically and prosodically completed, but that pair needs the second pair part to be recognized as a conversational action. Thus, while an adjacency pair is a part of sequential organization, a compound TCU is a 43 3 Previous research and key concepts resource for turn-taking (Schegloff 2007: 9, see also Lerner 2004b: 151– 152).15 3.1.1 Compound-TCU, secondary TRP The conversational structure of collaborative productions can be defined by how the organisation of turn-taking proceeds (see, Sacks et al. 1974). As mentioned earlier in this thesis, it is often assumed that a speaker is entitled to one turn constructional unit (TCU). A completion of a TCU constitutes an initial transition-relevance place (TRP), which is a place where a transfer of speakership can occur (ibid. 703). The unit of collaborative productions is the compound turn constructional unit (compound TCU; Lerner 1996: 240) and participants construct these together. These units consist of a preliminary component uttered by one speaker, and a final component uttered by another speaker (Helasvuo 2004: 1316). In terms of turn taking, the above-mentioned transition-relevance place typically occurs at the end of the final component. The preliminary component then projects the possible form of the final component in the TCU (Lerner, 1996: 240). Typical instances of compound TCUs consist of sentences with conjunctions (when-then, if-then) (Lerner 1991: 442). Between the components there may be a pause, an inhalation, or another element that allows speaker B to participate in completing the utterance. Pauses of this type can be referred to as intra-turn pauses16. They are also located in the middle of a turn and not between them (Lerner 1996: 260; see also Kurhila 2006: 100, 104). A transition can also occur without a pause or components can slightly overlap. The format of a compound turn-constructional unit is primarily a resource for turn-taking. As a resource, a compound TCU can be used to project the next appropriate place for a speaker change. That is, ‘it provides for the sequential possibility of anticipatory completion’ (Lerner 1991: 445, emphasis original). If another participant produces an anticipatory completion, it can initiate a collaborative turn sequence. The proffered completion can then be either accepted or rejected and this action becomes relevant as a responsive action. (Lerner 1996: 241, 2004a: 229.) 15 See also Ford (2001) for findings concerning turn and sequential organisation. According to Ford, the current speaker can continue speaking after the TCU has been completed, such as elaborating what was previously said (ibid. 60). 16 If the pause is situated after a possible completion of a turn-constructional unit (that is, inter-turn silence), it can be referred to as a delay (see, for example, Wong 2004: 114). 44 3 Previous research and key concepts According to Lerner (2004a: 231), the collaborative turn sequence occurs within one speaker’s turn space, not in the space of two turns separated by a TRP. Since the transfer of speakership occurs within one speaker’s turn space, it collides with the common turn-taking practice. When this occurs, the next speaker begins to talk within the turn space that is allocated to the prior speaker (Lerner 1994: 21). Traditionally, the speaker change occurs at the possible completion of a TCU. In compound TCUs, interactants construct the transition-relevance point together (Helasvuo 2004: 1316, Lerner 2004a: 240). The transfer of speakership therefore occurs within a compound TCU and this makes it a secondary place for the transfer. Although the place is secondary, it is still a systematic place for the transfer. (Lerner 1991: 454.) Final components can be pre-emptive in nature, meaning that they are ‘produced as a rendition’ of what the other was going to say’, but are not composed as a guess (for instance, with a try-marker) that would explicitly invite acceptance or rejection’ (Lerner 2004a: 229). However, the original speaker can consider them as being candidates that can either be accepted or rejected (ibid. 229–230). It is also possible to compare the organisation of the collaborative turn sequences with the repair phenomenon. According to Lerner (2004a: 232), both pre-emptive completion and other-initiated repair locate the just prior speaker’s talk as the object on which the turn sequence operates. They both systematically select that just prior speaker as next speaker. The prior speaker then has the right and the opportunity to control a turn to its completion. Lerner (ibid.) argues the following: For next-turn repair initiators this means having a right to repair any trouble even though it might be located by another participant. For collaborative turn sequences, this means maintaining control over what the completion will look like even in the face of a second participant making a bid for speakership within their turn space. The next action after the pre-emptive completion is allocated to the original speaker. Both the speaker of the original turn-in-progress and the speaker of the pre-emptive completion orient to it. (Ibid.) Let us consider the notion of the compound TCU in the following example. There are two collaborative productions, but the example is not exceedingly long. In first excerpt, the subject under discussion is mincemeat, which the participants consider to be very typical food in Finland. At the beginning (lines 1–2), the current speaker, S, suggests that perhaps immigrants also prepare some type of mincemeat sauce, which is 45 3 Previous research and key concepts something that Finns would prepare themselves. In line 3, speaker A suggests that another possible dish could be meat loaf. As we can see, the completion is placed near the end of the unit (cf. Lerner 1996: 261–263). Syntactically, this addition is enabled by the coordinating conjunction vai, ‘or’, which is pronounced at the beginning of the turn. In the second excerpt (from line 21), the participants are discussing Finnish sausage, which refers both to a trademark as well as to a specific word in colloquial Finnish (sininen lenkki, ‘blue sausage’). Although not everyone is certain that all immigrants consume this Finnish delicacy in their everyday diet they suggest that, for example, Japanese immigrants may put wasabi on the top of it. The original speaker in this example, S, likewise pronounces the added word and confirms it suitability. It is evident that both the speaker of the preliminary component and the speaker of the final component orient to the turn after the final component as though it belongs to the original speaker (Lerner 1996: 232). (3) MV 43., 44. <01:00:10> 01 S: 02 → 03 A: → 21 S: 22 T: 23 S: → 24 T: → 25 S: >Ni ni< valmista-a-k-s se sii-- teke-e-k-s Well well prepape-3SG-Q-CLI 3SG it make-3SG-Q-CLI se sii-t jauhelihakastikkee-n. 3SG it-ELA mincemeat:sauce-GEN Well does he prepare it does he make mincemeat sauce of it. Vai teke-e-k-s se sii-t jotai .hh [(.) Or make-3SG-Q-CLI 3SG it-ELA something Or does he make something of it [Muruke. Meat loaf. ((17 LINES REMOVED)) [Ja sit se, And then 3SG And then it, [J(h)oo. Yeah. »>laitta-a sii-hen jotain< [.hh put-3SG it-ILL something puts there something [päälle. PRT on top of it. vai jotai päälle. Or something on top of it. The transition can occur without a pause and can continue the prosodic pattern of the utterance (Szczepek 2010). In some cases, the final component can slightly overlap with the preliminary component. However, as Sacks, 46 3 Previous research and key concepts Schegloff & Jefferson (1974: 708) define a smooth transition (from one turn to a next), it covers those transitions with no gap and no overlap as well as transitions that are characterized by a slight gap or overlap This also applies to collaborative productions because they are usually formed so as to not disturb the ongoing conversation. 3.1.2 Formal aspects and functional aspects Collaborative productions have been analysed from a variety of perspectives. Different researchers emphasise slightly different aspects that range from strict syntactic analysis to examining the social nature of the phenomenon. Sacks (1992a, 1992b) touches upon the phenomenon in his lectures on conversation analysis and discusses them under several headings (for example, collaboratively built sentences, 1992a: 144–149; see also Sacks 1992b: 56–66). Lerner (1991, 1994, and 1996) refers to them as sentences-in-progress and his work in this area may be called groundbreaking. Both Sacks and Lerner are interested in characteristics that enable the participants to construct collaborative sentences. The social nature of these sentences provides a syntactic resource for the interlocutors to do something together. (Lerner 1991: 441, Sacks 1992a: 145.) Helasvuo (2004) and Ono & Thompson (1995, 1996) focus on the syntactic aspects of the phenomenon and they use the term co-construction. As Ono & Thompson (1995: 221) emphasise, ‘the realization of syntax is both locally managed and extremely dynamic’. Szczepek (2000a, 2000b) and Local (2004) adopt the broader term collaborative production, which does not adopt a strong stance in relation to the nature of the phenomenon. Within the wider framework of interactional syntax, co-constructions can be divided into two groups according to their syntactic types, as extensions and completions (Grenobl’ 2008: 25, Ono & Thompson 1995: 227–228, Ono & Thompson 1996: 75; see Sacks 1992a, on completers and recompleters). The differences between these types can be detected in their names. The first type, extension, refers to a situation in which speaker B expands what speaker A has stated into a new syntactic unit. For example, this could be achieved by adding a prepositional phrase to an already complete clause (Ono & Thompson 1996: 78). The second type, completion, refers to speaker B completing (or being offered an opportunity to complete) a syntactic unit that speaker A has left unfinished (Ono & Thompson 1996: 75, 1995: 227–228). In extension, the addition is accepted by the first speaker (Lerner 2004a: 236, Ono & Thompson 1995: 227). 47 3 Previous research and key concepts Helasvuo (2004: 1317) analysed the co-constructions in Finnish conversations and focussed on the completions that had the preliminary and the final parts that were syntactically more closely intertwined. These completions involved the co-construction of one clause and in addition to the analysis of their syntactic structure they were also analysed in terms of their interactional organisation within the sequence. Helasvuo argues that completions can occur both at constituent boundaries and within constituents (within a noun phrase) (ibid. 1317). The preliminary part determines the form that the final takes, such as personal deixis (ibid. 1332). Helasvuo (ibid. 1317) arrives at the conclusion that ‘syntax is shared and jointly produced, from the smallest units up to larger sequences’. Szczepek (2000a: 3) considers collaborative productions as means to form a social group within a conversational setting. During those productions, interlocutors use syntax as a linguistic resource for social aims. Szczepek (2000a, 2000b) discusses both the formal and functional aspects of collaborative productions in her two articles. Formally, Szczepek (2000a: 5– 9) analyses productions with respect to their syntactic, prosodic, and interactional units. However, as Szczepek (2000a: 16) emphasises, that type of division of linguistic levels is feasible only analytically. For interlocutors, however, all the levels are constantly interwoven, and this includes the prosody, pragmatic content, interactional activities and syntax. Functionally, collaborative productions are used to accomplish four broad, conversational actions. These actions include duetting, displaying understanding, borrowing, and eliciting information. Interactional co-operation occurs when the participants collaborate in telling a story, when they help one another in their word searches or in when they display understanding one another. (Schegloff 1984, Szczepek 2000b.) 3.1.3 Co-constructions and increments In addition to the above-mentioned criteria, turn-taking and syntax are also connected. Syntax is a means of controlling turn-taking, but turn-taking and the recipient’s action influence the syntactic form of the turn. If the recipient does not take the turn at the first possible point of transition relevance, the current speaker may either continue with a new TCU, or extend his or her turn by using a syntactically appropriate extension. (Couper-Kuhlen & Ono 2007: 513–514; Ford et al. 2002; Sacks et al. 1974: 704, 726–727; Schegloff 1996: 59). In other words, a turn can be constructed incrementally and the end of a turn can either be possible (turn-so-far), or it can be the actual end 48 3 Previous research and key concepts of a turn that involves a transfer of speakership (Ford & Thompson 1996: 143, Schegloff 1996: 55). The term ‘extension’ can also be applied when the current speaker somehow extends the turn after the possible transition-relevance point. These extensions can be referred to as increments (Schegloff 1996: 90–91, see also Fox 2007: 308). 17 An increment creates a new TRP in which the transfer of speakership can occur (Ford et al. 2002: 16). Some researchers also consider the cases in which the extension is not placed in the TRP; these extensions are called non-add-ons (Couper-Kuhlen & Ono 2007: 515). Turn extensions are usually grammatically dependent and backward oriented, extending the prior action (Couper-Kuhlen & Ono 2007: 547). If the extension is prosodically integrated with the prior unit without a TRP, it is a non-add-on. If the extension is placed after a TRP, it is an add-on. If the extension repairs some part of the host, it is a replacement. Non-repairing extensions can be divided into two categories in terms of their grammatical suitability. For example, glue-ons are grammatically fitted to the end of the host. In contrast, insertables do not properly fit the end of the prior unit, but they have some characteristics that allow them to be interpreted as belonging somewhere within it. Retrospectively, these new elements can be considered as constituting a part of the prior unit. (Ibid. 515.) Another type of increment is the free constituent or the unattached NP increments (Couper-Kuhlen & Ono 2007: 525, Ford et al. 2002: 26). Unattached NPs do not extend the syntax or the action of the previous unit. According to Ford, Fox & Thompson (2002: 26), ‘the speaker comes to a place of possible completion and then adds an NP that cannot be interpreted as a syntactic part of the just possibly-completed turn’. The distribution between the different interpretations is twofold. Firstly, unattached NP increments create a new point (an additional TRP) ‘at which a recipient could display appropriate responsiveness (after lack of uptake)’ (ibid.). Secondly, unattached NP increments also establish a stance or an assessment of a referent that was produced in the previous unit (ibid.). If the classification above is applied to collaborative productions, certain similarities can be detected. Although the term ‘increment’ usually refers to the extensions that the original speaker makes, an increment does share some characteristics with co-constructions and the way that co-constructions are built. However, a typical co-construction possesses features that are 17 If the ongoing TCU is halted and a short remark is added, this can be called a parenthetical sequence (Mazeland 2007, Duvallon & Routarinne 2001). 49 3 Previous research and key concepts somewhat contradictory. The second part of a typical co-construction is usually a non-add-on, which means that it is prosodically integrated with the prior unit without a TRP. In addition to this, the second part is usually also non-repairing in nature but rather a grammatically fitted glue-on or somehow belonging to the prior unit which would be then analysed as an insertable. The final component of a co-construction can be termed differently, depending on which of its aspects are adopted as a starting point. For example, continuation is intended to be neutral; it suggests that the coparticipants of a conversation are producing something that could be interpreted as a continuation to what was previously said. It does not assume a strong position with respect to syntactic or conversational structures. These aspects could, however, also be taken into consideration when examining continuations. One might also observe the direction of the final component. Turn extensions are usually analysed as being backwardoriented because they continue the previous unit. If a new unit initiates a new action, it is forward-oriented. (Couper-Kuhlen & Ono 2007: 513–514.) However, this position has been challenged. For instance, Koivisto (2011: 63) classifies syntactic extensions as forward-looking, while explanations and post-extensions are backward- looking. If a turn initiates a new syntactic and functional entity, it is referred to as a change of direction. 3.2 Repair organisation in relation to word searches When analysing word searches, I will adopt Kurhila’s (2006: 91) suggestion to regard word searches as part of the repair organisation. This organisation can be understood as a an umbrella term for the various procedures that interactants use to deal with problems they encounter in producing, hearing, or understanding talk18 (Schegloff et al. 1977: 361, Sorjonen 1997: 111). Repair sequences consist of two parts: a trouble-source turn (the repairable) and a repair turn, which may be preceded by a repair initiation. The repairable may possess some specific features (for example, errors or 18 Some researchers find the original definition of repair too narrow. Other-initiation of repair involves displaying a problem within the domains of speaking, hearing, and understanding. Svennevig (2007: 336), for example, regards the formulation “problems of speaking” as too narrow: “However, to my mind the formulation ‘problems of speaking’ evokes a too narrow conception of the nature of the problems, namely as being just a matter of ‘misspeaking’. But the problems may concern many aspects of producing an utterance that is recognized as a valid or felicitous social action in a given situation. In addition to linguistic problems (pronunciation, vocabulary, syntax, etc.) it also includes acceptability problems, such as saying something ‘wrong’ in a wide sense, that is, untrue, inappropriate, irrelevant, etc.” 50 3 Previous research and key concepts mistakes in speech production) which enable the repairs, but that is not necessary (Schegloff et al. 1977: 363). I have decided to focus on those word-search sequences that involve both the self and the other. That means the self-initiations that lead to self-repair are left outside the analysis. Kurhila (2006: 96) focusses on ‘substantial’ word searches in which the speaker attempts to involve the recipient in the search process: “By addressing the recipient, the speaker treats her as being the knowledgeable participant and creates an expectation that she will contribute to finding a solution.” 3.2.1 Who repairs and under what circumstances? Schegloff el al. (1977) were the first researchers to describe repair organisation in conversation for the first time. For them, there is a difference between the terms ‘repair’ and ‘correction’, and they see them hierarchically organized. ‘Repair’ is a more general term, whereas ‘correction’ represents a sub-group of a repair (ibid. 363). The term 'correction' is commonly understood to refer to the replacement of an 'error' or 'mistake' by what is 'correct'. [---] Some occurrences, clearly in the domain with which we are concerned, do not involve the replacement of one item by another. For example, a 'word search', which can occur if an item (e.g. a word) is not available to a speaker when 'due', is in the domain which we address, but is not a 'replacement' or a 'correction. In conversation, repairs can be self-initiated or other-initiated (ibid. 365). Self-initiation means that the repair is initiated by the current speaker, whereas other-initiation indicates an initiation by the recipient or by the interlocutor (for example, see Fox et al. 1996). Either participant can then make the actual repair. If the repair is performed by the current speaker, it is referred to as self-repair, and if the repair is performed by the interlocutor, it is known as other-repair (Sorjonen 1997: 114–118). In terms of the social organisation of interaction, neither of these types is to be treated as an independent type of possibility or event. They are not structurally equivalent, equipotential, or equally valued. Rather, self-repair and othercorrection are organisationally related. Self-correction is preferred to othercorrection. (Schegloff et al. 1977: 362, 370, 375–376.) It is common for the current speaker to interrupt his or her turn and self-repair it. In these cases, the linearity and uni-directionality of the spoken language become visible. This means that what has been said cannot be retracted. If one wants to rephrase something that one has said, he or she can perform a self-repair and begin again. (Fox 2007, Auer 2009.) In situations where the repair is other51 3 Previous research and key concepts initiated, the first speaker usually concludes the repair. Both self-initiated repairs and other-initiated repairs allow self-correction. (Schegloff et al. 1977: 377, Sorjonen & Laakso 2005.) In principle, a repair might be relevant to any sentence (Schegloff 1979: 269). Although my aim is not to analyse cases in which the initiation and the actual repair is self-performed, the following example illustrates a typical instance of this. In this example, a self-initiation leads to self-repair (see, for example, Fox & Jasperson 1995). (4) MV <7:56> 01 M: → Lena Lena ko-me-Lena tule-e huomenna koska mei-llä on huomenna come-3SG tomorrow because 1PL-ADE be.3SG tomorrow kov-- ko↑kouspäivä? mee?-- meeting:day comes tomorrow because we have tomorrow a meeting day? The participants in this example are discussing a new staff member who is scheduled to begin work at their workplace later that month. Speaker M is attempting to understand why this new staff member, Lena, will come to work the following day even though her contract has not begun yet. She suggests that the reason for Lena’s early visit could be related to their having another meeting that day. M self-initiates the search, cuts off twice (ko-, kov-) and finally finds the correct word (kokouspäivä) without any help from the other participants. As Lerner (2004a: 262) observes, ‘repair organisation seems to divide the opportunity to contribute to the search into ‘immediate’ and ‘delayed’ contributions’. Thus, if the word- search candidate is delayed, it provides the current speaker with an opportunity to self-repair (ibid.). Often the correction is directed at a lexical unit, such as word choice, but the speakers also correct single sounds or case endings (Schegloff et al. 1977, Sorjonen & Laakso 2005: 244). The search can also be self-initiated, but only when the actual repair is undertaken by the other. In this example, M is trying to decline the noun tapa, ‘habit’, in line 3, and faces problems with the consonant gradation p and v. (5) MV <00:05:00> 01 M: >Minä kunnioita-n< ↑erilais-i-a pien-i-ä 1SG respect-1SG different-PL-PTV small-PL-PTV ma-i-ta. country-PL-PTV I respect all kinds of small countries. 52 3 Previous research and key concepts 02 03 → 04 S: → 05 M: Ol-i kysymys vain sii-tä se on kulttuuri-sta ö, Be-PST.3SG question only it-ELA 3SG be.3SG culture-ELA VOC It was a question only about that. It is cultural, Se tapa kysymys tapa-sta? 3SG habit question tapa-ELA It habit a question about habit. tava-sta, habit-ELA About a habit. >tava-sta tava-sta.< habit-ELA habit-ELA About a habit a habit. Consonant gradation is a relatively common phenomenon in Finnish, occurring in word stems. It relates to the stop consonants p, t and k when they are preceded by voiced sounds, that is, vowels or voiced consonants. In Finnish, both the quantity (tt:t) and the quality of stops (t:d) varies. (VISK § 41.) In the example above, the gradation between the consonants p and v seems to be problematic, which is understandable because it is more common for the consonant t to alternate than for the consonant p (ibid.). When describing these types of instances, the term ‘correction’ is applicable. The term ‘repair’ can be used when the speaker produces an alternative expression for something the other speaker has said, whereas ‘correction’ refers to cases in which an alternative version is produced to replace a linguistic unit that can be regarded as erroneous according to the norms of standard language. (Kurhila 2006: 20, Schegloff et al. 1977: 363.) In the previous example, an incorrectly declined noun was replaced with a normative form. Both self-initiations and other-initiations are positioned in a systematic manner in relation to the trouble source, and they clearly differ from each other in terms of this position (Schegloff et al. 1977: 365). Self-initiations have three main types of position. First, they may be placed within the same turn as their trouble source. Second, they may be placed in the transition space of the trouble source turn. Third, they may be placed in the third turn after the trouble-source turn, that is, in the turn subsequent to that which follows the trouble-source turn. By contrast, other-initiations only occupy one main position: the turn just subsequent to the trouble-source turn. (Schegloff et al. 1977: 366–367, Schegloff 1992.) If we examine the sequential context of initiations, it becomes clear that self-initiations occur temporally earlier than other-initiations. This assumption is also confirmed by Kurhila’s data (2006: 96): 53 3 Previous research and key concepts However, it is quite common in my data that both types occur during a search sequence: the speaker is first doing a self-directed search, and if this does not lead to a result, the search then is directed to the recipient. In other words, I have left out cases where the speaker disrupts her speech in the middle of a TCU but finds the target word without (seeking) help from the recipient. As we observed earlier, self-repairs and other-repairs are not equivalent. In self-initiated repairs, the initiator can cut off a word and then either produce the correct alternative him- or herself or ask another participant to assist. Other-initiations also differ from each other in relation to how they are initiated. It has been suggested that the other who initiates the repair will likely first try the least feasible solution, preferring to address the problem as being caused by difficulties in hearing rather than in understanding or acceptability. (Svennevig 2008: 333, 347; see also Mustajoki 2012: 222.) Curl (2004) examines other-initiated repair sequences in American English from a phonetic viewpoint, and states that turns treated as trouble sources can be either fitted or disjunct. According to Curl, fitted turns ‘are appropriately designed and placed to fit within the structure set up by the prior turn or sequence’ (ibid. 8), whereas disjunct turns ‘are sequentially or interactionally inappropriate in relation to the prior turn and/or sequence’ (ibid. 10). Phonetically, fitted turns are louder and have longer durations than disjunct turns (ibid. 41). Disjunct turns are not louder and have similar articulatory settings (ibid. 19). Reproduction of lexico-syntactic features can be used in order to solve possible hearing problems, while phonetic cues display the co-participant’s orientation to the sequential appropriateness of the turn being repaired (ibid. 1). According to Schegloff (2007: 101), the least specific other-initiation takes the form of vague queries such as ‘Huh’ or ‘What?’, after which specificity may be increased in order to locate the trouble (for instance, ‘Who’ or ‘Where’). The initiator may also repeat the target repairable and, in the most specific of type of other-initiation, offer for confirmation some formulation of how the trouble-source was heard or understood. One of the ways in which a speaker can initiate a repair is pausing his or her turn. Other ways include separate search sounds (‘mm’, ‘uh’, and ‘er’). At a later stage, speakers may also make the search more explicit with questions such as ‘What is it’ (Schegloff et al. 1977: 363, 367). Mazeland and Zaman-Zadeh (2004) have studied other-initiated wordclarification repair in Finnish lingua franca interactions. In their data, nearly all instances of other-initiated word-clarification repair occur after an initial attempt at a hearing repair had failed. This leads them also to establish a rule 54 3 Previous research and key concepts of ‘non-firstness’ in word-clarification. Sorjonen’s (1996) study of repeats and responses in Finnish-language conversation also reports on a common pattern in which the requests for clarification and disaligning actions are preceded by repeats + confirmation. In some cases, word-search sequences resemble collaborative productions. Word searches allow collaborative completion and they can be described as being grammatically designed for conditional entry by recipients (Lerner 1996: 261–263). Furthermore, word searches provide an opportunity for the recipients of talk to participate in the search and to collaboratively complete it collaboratively. A collaborative completion is typically designed so that only the next word sought is produced. Following the search sequence, the primary speaker may continue his or her turn. Lerner (ibid.) noted that the search is often placed near the end of the unit, and this provides a place for the co-participant to complete the unit in progress. In the following example, M is talking about reality shows that she dislikes. This type of word-search sequence could be interpreted as one that enables collaborative completion. (6) MV 38. <01:34:41> 01 M: 02 S: 03 M: 04 S: 05 T: 06 S: 07 M: 08 S: Kyllä se on se on kiinnostava asia. Yes 3SG be.3SG 3SG be.3SG interesting thing Yes it is a very interesting thing. Se on roska-a, 3SG be.3SG rubbish-PTV It is rubbish, Mutta ihmiset katso-vat si-tä. But people watch-3PL 3SG-PTV But people watch it. Ihmiset ↑halua-vat si-tä. People want-3PL 3SG-PTV People want it. Miksi, Why, [LEIPÄ-Ä JA] sirkushuvi-a. Bread-PTV and circus-PTV Bread and circuses. ((from Latin: panem et circenses)) [Kaksi vaihto]ehto-a, Two alternative-PTV Two alternatives, Nii-PRT Yea- 55 3 Previous research and key concepts 09 M: → 10 T: → 11 M: Ei ole vaih-- vaihtoehtoʔ vaihto-a, NEG be.3SG alt-- alternative change-PTV There is no alt- alternative- change, (1) Vaihtoehto-a. ((tavuttaen)) Alternative-PTV (There is no) alternative. ((syllabifying)) Vaihtoehto-a. Alternative-PTV Alternative. Speaker M faces difficulties in finding the correct word (line 9). M has pronounced and declined the word appropriately in line 7, but due to the loud voice and the overlap with speaker S, M displays uncertainty regarding the noun choice she has made. In line 9, she begins with the cut-off word vaih-, then pronounces another cut-off word, vaihtoehto ʔ (Although this could also be interpreted as the nominative case of the word, it is pronounced with a glottal stop at the end; glottal stops or cut-off words are sounds that signal problems in clause formation. [Schegloff et al. 1977: 367].) Finally, she pronounces the word vaihtoa, ‘change’, which has the same beginning as the compound noun vaihtoehto, ‘alternative’. After this, she pauses for one second. T helps her in line 10 and produces the right form which M then confirms. As can be seen in example 6, the syntax of word-search sequences can be fragmented. The speaker begins with one syntactic construction, but when he or she is unable to find an appropriate word, the construction is left incomplete and a new one is begun. (Helasvuo et al. 2004: 1.) This differs from collaborative productions analysed in this study because their syntactic structure is formed differently. Although the preliminary component is syntactically incomplete, the final component is produced so that it completes the structure. It is also possible for the recipient to produce more than word in order to resolve the search sequence. In such cases, the recipient can produce the correct word as a part of his or her own utterance. In the following example, speaker N attempts to construct the Russian compound noun gossanitar, ‘state orderly’, which consists of the truncated adjective gos- (from gosudarstvennyj, ‘state’) and the noun sanitar, ‘orderly’ (see, for instance, Zemskaja et al. 1981: 126–127). In line 3, speaker E produces the correct word and accompanies his turn with the affirmative particles da, ‘yes’. 56 3 Previous research and key concepts (7) TI 19. <6:12> 01 N: → 02 03 E: → 04 N: A èto-t (-) ili kto-to mne ska↓za-l-ø vy VOC 3SG-MASC (-) or someone-CLI I.DAT say-PST-MASC 2PL rabota-l-i (.) esli ja pravil’n-o ponja-l-a? work-PST-PL (.) if 1SG correct-ADV understand-PST-FEM A that or someone said to me that you worked if I understood correctly? Èt-o by-l-o taka-ja rabota čto vy 3SG-NEU be-PST-NEU such-FEM work that 2PL proverja-l-i (.) kak san-- [é, check-PST-PL (.) like san- VOC It was that kind of word that you checked like sani-, [È: san-- gos[sani (-) da da. VOC san-- state:orderly (-) yes yes State:orderly yes. [Èt-o t-o (.) horos-o. 3SG-NEU 3SG-NEU (.) good-ADV It it okay. Although the ongoing word-search sequence might cause disturbances in the sequential context and hence in the progressivity of the utterance, the search is usually organised to show that participants are attempting to continue the TCU. However, a halt in the TCU progressivity allowed another participant to produce an opportunistic completion for the TCU. They are built and treated as turn-completing actions. Usually only the searched-for next word is actually produced. (Kurhila 2006: 155; Lerner 2004a: 229, 261.) 3.2.2 Candidate understandings – somewhere in between word searches and collaborative productions I have emphasised that collaborative productions and word searches form a continuum and that their interpretation depends on thorough sequential and linguistic analysis. In collaborative productions, the progressivity of a turn need not be disrupted, whereas in word searches, the need to secure the intersubjectivity surfaces the conversational level as well. However, the disruption caused by word-search sequences does not necessarily have to extend to a longer sequence. In some cases, it is also possible for either the speaker or the recipient to shift focus to certain elements in the prior speech instead of providing the (projected) next relevant turn. These types of instances involving a co-participant articulating his or her interpretation of the previous turns can be referred to as candidate understandings (Kurhila 2006: 153–217). 57 3 Previous research and key concepts Sequentially, a candidate understanding functions as the first pair part of an adjacency pair because it projects the confirmation of the next turn (Kurhila 2006: 155; Lerner 2004a: 229, 261; Schegloff & Sacks 1973: 295– 96.) Although this does not constitute a typical instance of adjacency pairs (such as the exchange of greetings or question-answer sequences; Duranti 1997: 251–252, Raevaara 1997: 76), it is a sequentially recognisable action. In relation to compound TCU’s, the sequential context of candidate understandings is somewhat different. A compound TCU is a resource for the organisation of turn-taking and an adjacency pair for the sequential organisation (Schegloff 2007: 9, see also Raevaara et al. 2001: 15–16). A turn produced with the help of a compound TCU only comprises one turn, even though two speakers are producing the parts of the compound TCU. When a speaker experiences something as problematic in the prior turn, the speaker can either repair the problem or check his or her understanding of the problematic utterance. In other words, the speaker can articulate his or her understanding of the other speaker’s meaning. While such articulations shift focus on the previous turn(s), they need not be solely backwardoriented (as is the case with corrections). As the speaker uses his or her own words to describe what he or she thought the other speaker was saying, he or she often introduces new elements into the conversation. So, rather than just modifying the prior turn, articulations may advance from the problematic situation by presenting a potential interpretation of the trouble turn. (Kurhila 2006: 153–217.) The following example illustrates a candidate understanding that is projected rather than being backward-oriented. This example concerns immigrants who live in Finland, but they are examined from a different perspective: (8) PM 6. <40:55> 01 02 03 D: Se on mu-n oma mielipide. 3SG be.3SG 1SG-GEN own opinion. That is my own opinion. syyt-- yhdistet-ään sii-hen teidä-n oma seminaari prin-- connect-4 3SG-ILL 2PL-GEN own seminar ei-kö me men-nä vähän pois sii-tä meiä-n teema-sta. NEG.3SG-Q 1PL go-INF little away it-ELA 1PL-GEN theme-ELA But if we connect that to your own seminar do not we go a little far away from our theme? 58 3 Previous research and key concepts 04 F: 05 D: 06 F: → 07 F: 08 D: → Ei: ei (-) koska se on se .h ö <päätavoite> että NEG NEG (-) because 3SG be.3SG 3SG .h VOC main:goal that me-kin puhut-aan vapaaehtoisuut-ta ja maahanmuuttaja että 1PL-CLI talk-4 voluntary-PTV and immigrant that miten se vapaaehtoisuus vo-isi palvel-la (.) tai how 3SG voluntary can-CON.3SG serve-INF (.) or palvel-la vai m(h)iten [sano-a se-n että .h serve-INF or how say-INF 3SG-GEN that .h No no because it is a main goal that we also talk about volunteering and the immigrant that how that volunteering could serve or serve or how to say it that [Joo. Yeah. miten maahanmuuttaja vo-isi (.) ä miten se ↑autta how immigrant can-CON.3SG (.) VOC how 3SG help.3SG maahanmuuttaja kotoutumise-en. immigrant integration-ILL How an immigrant could how it could help an immigrant to integrate. Että: se on se, That 3SG be.3SG 3SG That it is that, Sii-hen ↓toiminto-on. 3SG-ILL action-ILL To that action. In terms of conversational repair, these candidate understandings can be categorised as other-initiations of repair: they present a possible understanding of the trouble turn which can be either confirmed or disconfirmed by the speaker of the trouble turn (cf. Schegloff et al. 1977). What has been stated previously is paraphrased and summarised, and the actual candidate understanding can begin with a particle (Kurhila 2006: 161). According to Kurhila (ibid. 155), ‘candidate understandings are used to check the level of shared knowledge between the participants’. Any unclear relations that arise between separate turns can be clarified by elaborations. The current speaker can also check his or her interpretation of the prior speaker’s turn. From an interactional perspective, producing candidate understandings demonstrates, albeit tentatively, that participants are co-operating, and this emphasises the non-problematic nature of the interaction. (Ibid. 181–182.) 59 4 Data and transcription The data of this study comprise approximately eleven hours of naturally occurring conversation, ranging from everyday situations to institutional settings in Finland and Russia. The data were either audio-taped or videotaped. Some conversations were recorded in both ways. I have collected the Finnish data myself, but the Russian data have been partially obtained from colleagues 19. My goal was to obtain as versatile material as possible. Of course, would I have concentrated on specific situations, I would have obtained material that was more similar. Nevertheless, very few studies have been conducted on both Finnish and Russian as lingua francas, so it is important to observe different types of conditions of use. The basic assumption of CA-based research is that utterances are always multifunctional and they cannot be analysed without a context. For this reason, a quantitative approach to the subject does not suit conversation analysis. (Lappalainen 2004: 20, see also Seidlhofer 2009: 49 on ELF research.) This study is not quantitative, but the generalisations presented are based on an analysis of an extensive number of examples. The number of instances is provided in the table below: Table 1. The number of examples. Collaborative Word searches productions Candidate understandings Finnish 32 37 8 Russian 24 14 15 Total 56 51 23 A total of 130 examples were examined. In addition to these, an analysis was carried out on several borderline cases that occurred both outside of these categories and within them. The different groups will be presented more thoroughly in chapters 4, 5, and 6. In some of the categories, the 19 I am extremely grateful to Jalmari Auvinen, Nina Dobrushina and Noora Lemivaara-Khudoikulova for sharing their data with me. 60 4 Data and transcription distribution of the examples between those two languages might seem disproportional for some activities, such as word searches, but their occurrences can be motivated through other factors. Although to some extent I observe the language use of my informants in relation to colloquial Finnish and Russian, the data is insufficient for charting regional and social variation. In certain conversations, aspects such as the standard forms and colloquial forms of personal pronouns (mina-mä, minun-mun) are used slightly differently depending on the theme and the coparticipants in the conversations. With respect to the Russian examples, the participants use the colloquial forms of certain adverbs, for example sejčas – sčas, ‘now’, prodlënka, ‘continuation of lesson’ (for instance, see Ermakova 1984, Kapanadze 1984). However, my database is rather heterogeneous. As my data consist of approximately 60 non-native speakers with a wide range of language skills and linguistic backgrounds, it was not possible to focus on a particular ethnic identity. In addition to recordings, I also recorded notes on situational factors, such as who sits where and whether participants moved during the recording. The video camera was usually placed to the side of the actual conversation or meeting, but a dictaphone was placed on the table among the participants. In some of the recordings, the video camera was relocated to a more suitable place during the recording. All participants were asked for their permission to use the recorded material for this study. In addition, the names of the participants in the examples have been changed. The transcription of the data is time-consuming, but on the other hand, this suggests that the researcher is very well aware of the nature of his or her data and also of its shortcomings. Concerning this study, this means that not all the material is applicable, for example, for phonetic measurements. This is because the recordings were collected at different times and with a different set of equipment. Many of the recordings also contained various types of additional noise (cows mooing, children crying, sounds of cooking, etc.) that would make a phonetic analysis difficult and complicate the comparison of the results. Therefore, when analysing the prosodic features of the conversations, I can only observe them with the same accuracy as the actual participants, that is, without the help of electronic devices. (cf. Lilja 2010: 25.) Next, I will introduce my data with greater detail and describe the transcription and glossing conventions used in this study. 61 4 Data and transcription 4.1 Finnish data The Finnish conversations were taped in a variety of places and situations. The conversations took place in workplaces, in different meetings and gettogethers. Furthermore, the different types of meetings of non-governmental organisations were taped. One resource for informants and natural conversations was the community colleges and the Finnish courses they offered. The participants have very different linguistic backgrounds: more than 20 mother tongues are represented and many of the participants describe themselves as either being bilingual or multilingual. The most frequent mother tongue mentioned is Russian, but many participants spoke either Arabic, Somali, Kurdish, or Turkish. Overall, a high number of my informants reported that they know Russian either as a second language or as a first foreign language. These informants were born either in the Soviet Union or came from the Commonwealth of Independent States (CIS) and they learnt Russian at school. When I recorded the material, this selection of Russian-speaking participants was not a conscious choice, but it was understandable. According to Statistics Finland 20, at the end of the year 2012 more than 60,000 Russian speakers lived in Finland and they comprise a quarter of the whole foreign-language speaking population21. One of the conversations consists of two Finns interacting. If the notion of lingua franca interaction is interpreted in its purest sense, then it refers to those communicative situations in which the participants do not have a common language (House 1999, Seidlhofer 2001). However, some researchers also study contexts in which both native and non-native speakers are present (Mauranen 2009a, Suviniitty 2012). For example, Suviniitty (ibid. 36) observes that “as with all language use, a certain degree of fuzziness has to be accepted.” Even though differences exist between the different participants’ language skills, I presume that we cannot reduce the conversation in a foreign language to characterize it as being merely a collection of erroneous 20 The Official Statistics of Finland (OSF): Population structure [e-publication]. ISSN=1797-5395. 2012, Appendix figure 2. The largest groups by native language in 2002 and 2012. Helsinki: Statistics Finland [referred: 5.3.2014]. Available at: http://www.tilastokeskus.fi/til/vaerak/2012/vaerak_2012_2013-0322_kuv_002_en.html. 21 According to the Constitution of Finland, there is no official language in Finland, but the national languages are Finnish and Swedish. The constitution also mentions the Sami as an indigenous people of Finland, and the Roma and other groups have the right to maintain and develop their language and culture. Sign Language is likewise mentioned in the constitution. 62 4 Data and transcription utterances. In fact, the participants tend to not even pay attention to obvious grammatical errors, possibly because they focus more on problems that arise and these become the object of conversation through their cooperation. Moreover, speakers who know a language better would be unlikely to adopt a teaching role in order to instruct the unknowing participant in acquiring better language skills. However, it is interesting that the participants comment rather frequently either on their own mother tongue, or on the others’ mother tongues. They usually do this by comparing their languages. For instance, they comment on the Finnish language having or not having some particular feature that may or may not occur in their mother tongue. The descriptions of the overall conversational settings and participant relationships are provided below. Nonetheless, the figure provided in the table does not necessarily reflect the actual duration of the whole recording or the speech event. Table 2. Finnish material. Name of the segment, abbreviation used in examples Monivisio, MV Duration 120 min 1 recording Intensive course, IC 15 min 1 recording Conversation course, CC 42 min 2 recordings Women’s Night, WN 30 min 2 recordings Finnish language course 60 min acquaintances, FC 7 recordings Project meeting, PM 30 min 1 recording Total 4 hours 6 minutes 63 4 Data and transcription Monivisio, one recording (120 min) A staff meeting of a multicultural radio and television project. Seven participants are discussing future activities concerning television and radio programmes. The television and radio programmes are broadcast in five languages: Finnish, Russian, Turkish, Persian and Somali. Two Finnishspeakers are participating in the meeting. Intensive course, one recording (15 min) An intensive course in the Finnish language. Seven participants of Russian, Danish, German, and Portuguese descent are cooperating in group tasks during a lecture. Conversation course, two recordings (42 min) A Finnish conversation course. Two groups of three participants are discussing different subjects together, such as how a sense of humour is perceived in different countries. The first group consists of speakers of English, Spanish and Lithuanian, the second group has one Lithuanian and two Russian participants. Women’s night (Cuba & Mexico), two recordings (30 min) These recordings contain two discussions, one involving three Cuban women who have moved to Finland many years ago. In addition to the Cuban women, other immigrant women participate in this discussion. These participants are comparing the different ways of life with the Finnish participant. This includes another discussion that concerns a Mexican photojournalist who has lived and worked in Finland for several years. In addition to this photojournalist, immigrant women are also participating in this discussion. The participants discuss the Finnish immigrant policy and how immigrants could become active players in Finnish society. Finnish language course acquaintances, seven recordings (69 min) The interlocutors in these recordings have met each other in their Finnish course. The number of participants varies from nine to eleven. The participants are from Somalia, Estonia, Morocco, Albania, Russian, Japan, and Turkey. Their Finnish teacher is originally from Russia. The recordings were taped before their lectures and during their breaks. The group also 64 4 Data and transcription allowed me to participate in their homework sessions. The conversations range from grammatical debates to discussion about their weekend plans. Project meeting, one recording (30 min) This meeting is a gathering of representatives (Russian, Persian, Romanian, Somali, and Slovenian) from different non-governmental organisations. They are discussing events such as seminars that they are planning to organise. Certain organisations, such as the Finnish Red Cross, have different types of projects that are organised for immigrants. Usually these projects are short-term, as they consist of the launch, the actual working and a closing seminar. 4.2 Russian data The Russian conversations were taped in Russia, Tajikistan, and Dagestan. One of the conversations was taped in Finland. The conversations took place at workplaces and informal get-togethers. In addition, different types of meetings of associations were taped. In one of these conversations, two Russians participate in the conversation. Table 3. Russian material. Name of the segment and the abbreviation used in the examples Interviews in Tajikistan, TI Duration 100 min 4 recordings University meeting, UM 30 min 1 recording Dagestan meetings, DM 74 min 2 recordings Students in Ivanovo, IS 97 min 2 recordings Total 5 hours 1 minute 65 4 Data and transcription Interviews in Tajikistan, four recordings (100 min) Four Russian interviews were conducted in Tajikistan. The same interviewee asks four Tajik people about their language use and preferences. University meeting, one recording (30 min) This is an unofficial, informal get-together after a seminar. The professors, teachers, and students are discussing educational issues and research perspectives in a more relaxed atmosphere. The eleven participants are from Japan (4), from Finland (5), and from Russia (2). Dagestan meetings, two recordings (74 min) Two Russian interviews in Dagestan with speakers of Northeast Caucasian languages such as Avar, Lezgian, and Tabasaran. The interviews touch upon the traditional livelihood of Dagestan, the language use and the language histories of different speakers. Students in Ivanovo, two recordings (97 min) These recording involve a Finnish exchange student who conducts two interviews with other exchange students in Ivanovo in the Russian Federation. One of the students is from the Ivory Coast, the other from Angola. Due to their ethnic background, they both have encountered problems in Russia. During the interviews, the Finnish student attempts to determine how these students have observed Russian society and how they have experienced discrimination. 4.3 Discourse transcription, interlinear gloss, and translation The examples in this study have been transcribed in accordance with the conversation analytic conventions that were developed by Gail Jefferson (as formulated in Seppänen 1997).22 The transcription and glossing symbols are explained in the Appendix. The turns in the examples are mainly represented in three lines: the first line is the original Finnish or Russian utterance; the second is the gloss line; the third line appears in bold and is an approximate English translation. 23 Because the linguistic structures of Finnish and 22 Russian researchers tend to use other ways of transcribing (for example, see Kibrik & Podlessakaja 2003). 23 It should be noted that the glossing is based on the markedness of the items. The following categories are treated as unmarked and left unglossed: the singular form, the nominative case, the present tense and the active voice. The different types of infinitives and participles that occur in Finnish and Russian are not specified. They are labeled as the INF and PCP, respectively. 66 4 Data and transcription Russian differ from English, translating them may sometimes present problems. It is true that many ‘non-native speech features’ that are produced by the NNSs of Finnish and Russian surface at the morphological level in connection with such linguistic aspects as case endings. I have transcribed conversations based on the participants’ actual language use so that the language in the examples has not been corrected. Ungrammatical instances will not be included in the analysis unless the participants themselves make them the topics of their speech. 24 As not all of the conversations have been video recorded, I have been unable to observe gaze and gestures systematically. Furthermore, in some video recordings, some of the participants have their back to the video camera and for this reason, their gaze and gestures cannot be observed. Hence, I have decided to concentrate on the verbal construction of jointly produced structures. Still, whenever the recordings have made it possible to analyse gaze and gestures in accordance with the individual examples, I have utilised this possibility. I will not, however, make generalisations based on the analysis of these individual examples. All conversations have been thoroughly transcribed in accordance with sequential organisation, pauses and overlaps. With the help of this type of preliminary transcription, I was able to locate most cases in which the participants jointly produce structures. These turns and sequences were then more thoroughly transcribed. But how does one transfer a spoken form of a language into its written form? Languages have different orthographic conventions and this affects the transcription process (for example, see Markee 2000: 59). It can be problematic to transfer non-native Finnish or Russian into a written form, since a non-native speaker can pronounce certain speech sounds in a way that differs from standard Finnish or Russian pronunciation. However, I have decided to use the basic Latin alphabet with certain Finnish letters (ä, ö). The Russian examples have been transliterated from the Cyrillic script into the Latin alphabet in accordance with the scientific transliteration of the Cyrillic alphabet (see table 4). 24 Although the used language form is not the primary theme for research, I have noted some of the possible tendencies in language use and provided the corresponding examples. Two observations are needed beforehand. Firstly, the possible cases of ungrammatical language use are occasional in the sense that not every speaker shares the same forms or uses the language similarly. Secondly, as mentioned previously, these are only possible tendencies, meaning that I have not paid attention to every single instance. All in all, these cases might turn out to be useful when planning future research. 67 4 Data and transcription I have also used actual words in the gloss line whenever possible. This means that the prepositions or postpositions are translated with the corresponding English words (for instance, kanssa, ‘with’, krome, ‘except’). Moreover, different types of discourse particles or continuers are expressed by using similar English expressions (mhy/ugu, ‘uhhuh’, aga, ‘aha’; see, for example, Schegloff 1982, see also Lukina 2005 [2003], Sorjonen 1999). Finally, vocalization markers are used to indicate hesitation sounds. Table 4. Cyrillic letter а б в г д е ё ж з и й к л м н о п р с т у ф х ц ч ш щ ъ ы ь ю я Tranliteration to latin letter a b v g d e ë ž z i j k l m n 68 o p r s t u f h c č š šč ″ y ′ ju ja 5 Collaborative productions I will now focus on the structures that the participants produced collaboratively in conversation (for example, see Ferrara 1994; Grenobl’ 2008; Helasvuo 2004; Lerner, 1991, 1996, 2004a, 2004b). In these cases, a pre-emptive completion of one speaker’s turn-constructional unit by a subsequent speaker can operate on that unit to transform its production into a sequence that can be referred to as a collaborative turn sequence. (Lerner 2004a.) We can also examine the structure of the conversational units from a wider perspective, observing how one clause is co-constructed in interaction (Helasvuo 2004: 1315). One may also emphasise that grammatical structure is not an artefact of linguistic inquiry that has isolated language form from its actualisation as talk-in-interaction (Lerner 1996: 238). For example, Grenobl’ (2008) discusses jointly produced turns and divides them into two different categories: extensions and completions. In general, extensions are used to expand the first speaker’s turn, while completions serve to complete the turn, whether it is completed syntactically or semantically (ibid. 25). Although the emphasis may be placed on slightly different aspects, the production of these joint structures is usually understood as an interactional achievement. Thus, a speaker completing the utterance of another speaker reveals aspects of an interactionally relevant syntax. Participants can therefore be said to analyse syntactic structures and to orient to them. (Lerner 1991: 441, 445.) 5.1 Previous research on collaborative productions The conversational structure of collaborative productions can be defined in terms of the organisation of turn-taking. The unit of construction is a compound turn-constructional unit (compound TCU; Lerner 1996: 240). This concept is adopted from the general structure of conversation. The basic assumption is that each speaker is entitled to one turn constructional unit (TCU) at a time. With reference to the English language, unit-types can be sentential, clausal, phrasal, and lexical constructions (Sacks et al. 1974: 704). In addition, the phonetic form and intonational packaging play an important role in forming TCUs (Schegloff 2007: 3–4). In addition, compound TCUs (TCU-in-progress) are co-constructed by the participants in the conversation. These units consist of a preliminary component that is uttered by one speaker, and a final component by another speaker. 69 5 Collaborative productions Moreover, the preliminary component typically projects a possible form of the final component of the TCU (Lerner, 1996: 240). Typical instances of compound TCUs consist of sentences with conjunctions (when-then, if-then; cf. in Russian esli-to) (Lerner 1991: 442, see also Grenobl’ 2008: 30). The main categorisation can be established with the help of projection. If a compound TCU format is projectable, it therein provides resources for a completion by another speaker (Lerner 1991: 450). A compound TCU consists of two components. In the course of its formation, the first component is a preliminary component and ‘it foreshows both a place where a final component could begin and a form that such a final component can take’ (Lerner & Takagi 1999: 53, emphasis original.) What follows, then, is that the preliminary component provides an opportunity for the coparticipant to step in and to contribute in the compound TCU formation (ibid.). Hopper and Thompson (2008) have studied projectability and clause combining in interaction. The term used by Hopper and Thompson is ‘joint utterance production’ (2008: 114). These productions contain two clauses. The first clause typically projects the form of the second clause, which may not exhibit any consistent syntactic structure. It is an interdeterminate stretch of discourse. (Ibid: 99.) What makes the candidate completion possible is grammatical, prosodic, and social projectability (ibid. 115). According to Hopper and Thompson, projectability is affected by both social factors and linguistic structures. The structure of collaborative productions can be described as consisting of three separate stages. Firstly, the current speaker produces a preliminary component which does not necessarily display a particular problem. Secondly, the recipient produces a final component which might slightly overlap with the preliminary component. Thirdly, the speaker who produced the preliminary component has an opportunity to react to the final component. Usually, the progressivity of the conversation is not interrupted when a collaborative production is produced unless the speaker of the preliminary component specifically reacts to the final component. The final component can be analysed as being pre-emptive. When two participants are engaged in a joint activity, the recipient can respond to the prior speaker, ‘not by waiting until completion to act, but by pre-empting that completion as a method of responding’ (Lerner 2004a: 22, see also Schegloff 1984: 42). Thus, the first and second speaker can jointly produce a collaborative turn sequence (Lerner 2004a: 225). Both the speaker of the 70 5 Collaborative productions original turn-in-progress and the speaker of the pre-emptive completion orient to the turn after the pre-emptive completion as a turn allocated to the original speaker. Lingua franca researchers have focussed on candidate completions – or to be more precise – on the lack of them. Firth (1996: 246) states that there were few candidate completions in his data, even though in some cases, a completion would have been feasible or viable. One possible reason for the lack of completions could be that Firth’s data consisted of telephone calls (ibid. 241); although grammatical and prosodic projectability are also still in effect in telephone calls, some of the social projectability might be lost when the participants cannot see each other. When collaborative productions are analysed syntactically, the choice of terminology can be challenging. Towards this end, one could use terms such as ‘syntactic structures’ although this could be interpreted as reflecting a tendency towards analysing how written language structures are described (Selting 1996: 361). For example, Auer (1996: 59) adopts the term ‘syntactic gestalt’ and validates his arguments by referring to the idea of grammatical rules that produce syntactic structures. Auer (ibid.) observes that, ‘we are less interested here in syntactic structures as the potential output of some abstract grammatical system, than as communicatively and cognitively real events in time’. A closed syntactic gestalt can form a possible syntactic completion point, such as a potential transition-relevance place. For the production and recognition of possible completion points, contextualisation cues play an important role. These cues are syntactic, prosodic, semanticopragmatic and visual parameters. While their meaning is not stable, they must be interpreted in a local environment and they tend to co-occur. (Ibid. 58; for emergent grammar, see also Hopper 2011: 27; Auer & Pfänder 2011: 3.) Where the linguistic structure of collaborative productions is concerned, I will use the term syntactic structure even though I am aware of its shortcomings. However, as the analysis of single examples requires a thorough examination of the turn-organisational and linguistic factors, this also means that the terminology needs to be clarified for the analysis. In other words, as is evident in examples, although the basic concept is syntactic structure, the actual categorisation is more subtle (see also Helasvuo 2001b, 2004). With regard to emergence, it is noteworthy that syntactic units can be achieved through the participants’ mutual activity (Helasvuo 2001b: 33–35). 71 5 Collaborative productions Table 5 shows the number of examples in relation to the different categories. Apart from the coordinated words and phrases, the final components can be divided into two groups. Thompson & Couper-Kuhlen (2005: 492) argue: ‘what second speakers tend to add to a first speaker’s contribution is either (1) the second clausal component of a multi-clausal unit or (2) the last word or two of a mono-clausal unit’. Table 5. Collaborative productions Collaborative productions Finnish (32) I Coordinated words and phrases Russian (24) 2 1 29 19 a) Agreement 6 5 b) Subject NP + predicate verb 5 8 c) Objects, other complements 14 6 d) mennä ‘to go’ 4 - III Complex sentences 1 4 II Phrases and clauses 5.2 Participants co-construct a phrase or a clause 5.2.1 Co-ordinated words and phrases In this chapter, I will first introduce co-ordinated structures in which two or more syntactically equal words, clauses or sentences that are combined. This co-ordinated item has the same syntactic function as its parts. (VISK § 1079.) In other words, they comprise a special case of co-constructions. Subsequently, I will move on to the actual co-constructioning of phrases and clauses. An analysis will be presented on agreement phenomena, the subject + predicate cases and the collaborative production. With special reference to the Finnish data, there are also cases in which the collaborative production is enabled by semantic motivation. In addition to syntactic or semantic cohesion, another concept that plays a crucial role in analysing emergent constructions is togetherness (for example, see Barth-Weingarten & Couper-Kuhlen 2011 for conjoined verb phrases: VP1 and VP2). According to Barth-Weingarten and Couper-Kuhlen 72 5 Collaborative productions (ibid. 269), ‘in one case a speaker projects prosodically at the end of the first part that more is to come and this ‘more’ turns out retrospectively to be another VP linked to the first by and’. The first speaker who produced the first VP as well as the second speaker who produced the second VP form a compound turn-constructional unit. When this occurs, there is a prosodic boundary marker that denotes incompleteness (ibid. 271). In the first example, M and J are talking about how to film video material for different purposes. M usually works as a video photographer and she is known for filming too much material for even short television programmes. At the beginning of the excerpt, both participants are trying to establish the common principles for filming. M begins her turn in line 1 and J steps in after the conjunction ja, ‘and’, pronouncing the final component. The conjunction is placed at the end of the preliminary component: (9) MV 37. <01:37:50> 01 M: 02 J: 03 M: 04 Normaali-a viis minuutti-a ↑ohjelma-a (.) ja, Normal-PTV five minute-PTV ohjelma-PTV and Normal five minutes of programme and, Tunti kuvaus-ta. Hour filming-PTV An hour of filming. Kolme tunti-a ku↑vamateriaali-a, Three hour-PTV picture:material-PTV Three hours of video material, Viis minuutti-a ohjelma-a. Five minute-PTV programme-PTV Five minutes of programme. Syntactically, co-ordinated noun phrases function as the subjects of a copulative clause. In this particular case, the copula is zero marked. The noun phrases are syntactically and semantically identical and the head of the noun phrase is a numeral. It is interesting that in lines 3–4, speaker M produces another variant of the possible combinations of filming and programme. M’s turn is syntactically and structurally identical to the first collaborative production, but the order of the elements is reversed. M also extends the amount of the film material needed to three hours. The structure of the collaborative production is recognisable, and although M is accepting it as a model for her second turn, she manages to formulate it to fit her own purposes. In the second example, the participants are discussing an upcoming football match. They are all football enthusiasts who follow different types of football leagues. In this excerpt, the subject of the conversation is the 73 5 Collaborative productions Premier League. In line 9, R mentions the name of the first team to participate in the match along with a coordinating conjunction. In line 10, S completes the turn by producing the name of the second team. The form of the second team’s name, Manu, is colloquial (short for Manchester United). (10) FC 108. <09:15> 01 S: 02 B: 03 S: 04 S: 05 R: 06 B: 07 S: 08 R: → 09 S: → 10 B: Aika >on on on< (-) kello: ä: (1) Time be.3SG be.3SG be.3SG (-) clock VOC (1) The time is o’clock Sei-SevVartti:-a vaille (1) kummenen. Quarter-PTV PRT (1) ten Quarter to ten. (2) Vartti-a vaille (.) ku-Quarter-PTV PRT (.) teQuarter to te↑Kymmene-ltä. Ten-ABL At ten o’clock. <Yhdeksän yhdeksän> kolme neljäkyt viis, Nine nine three forty five, Ei lauantai ei, No Saturday no, Arsenal ja, Arsenal and, Manu. ManU. Manchester Arsenal. Manchester Arsenal. In line 10, speaker B, who was not a part of the first co-construction, produces yet another possible structure that has parts that are ordered in reverse. From the perspective football match protocol, only one of the orderings is accurate because one of teams is the home team and the other is the visiting team, and it is customary to declare the teams in their home– visitor order. However, from a practical standpoint, the order does not actually play an important role. The second example is similar to the first example. In both cases, the conjunction is placed at the end of the preliminary component. In addition, the words or clauses are coordinated symmetrically, as their order can be reversed (VISK § 1090). When participants are engaged in a collaborative production of the elements in conversation, they are constantly interpreting the actions of the other participants. If the preliminary part of the compound TCU ends in a conjunction, the TCU can project continuation in the final part. However, 74 5 Collaborative productions this does not always occur. For example, Koivisto (2011: 58) reported that although a turn ends in a conjunction, participants do not interpret the turn as being incomplete. Walker (2012: 149) states that: ‘participants do not systematically attempt to produce the talk which might otherwise be projected by the conjunction (---): rather, the talk ending with the ‘trail-off’ conjunction is treated by participants as informationally complete’. Co-ordinating conjunctions that require a certain type of continuation usually occur in story-telling sequences. In addition, they are placed at the end of a prosodic unit, and intended to denote that the story has not yet been completed. (Koivisto 2011: 89.) Syntactically, this type of use of conjunctions such as ja, ‘and’, cannot be interpreted as being coordinating because the order of the elements is temporarily fixed, not exchangeable. In the first example, the first two lines are somewhat disorganised. Filming and making a television programme are temporarily organised but in this particular example, J and M express that the order of the elements is exchangeable. In some cases, the participants may construct lists that consist of two or more parts; usually they are three-parted (Jefferson 1990, Lerner 1994). Participants orient to this three-parted structure as a factor that controls the sequential organisation. If only one of the participants constructs the list, the recipient interprets that the first possible place of transition relevance occurs at the end of the third part (Jefferson 1990: 73–77). According to Jefferson (ibid. 81–82), a list can be constructed by more than one speaker, and a link term (such as ‘and’ or ‘or’) can be used to achieve collaborative construction. The three-partedness can be being attributed to the requirements that conversational interaction imposes. As Lerner (1994: 22) states ‘List construction, as a situated social achievement in conversation, is shaped by the social coordination systems that organize conversational interaction’. A construction list is being kept as short as possible even though the participants continue to perform some type of interactional activity that requires listing. (Ibid.) In the first two examples, the constructions had certain features that connect them to list constructions despite their not being three-parted. In both cases, the link term ‘and’ was used to achieve co-operation between participants. Both cases also required social knowledge of the situation. Football matches by nature involve two competing teams. Therefore, although the list structure does not have three parts, the participants were performing a certain type of interactional work that required listing. In both 75 5 Collaborative productions cases, an anticipatory completion initiated a collaborative turn sequence in that the first speaker either accepted or rejected the turn-constructional unit completion by another speaker (cf. Lerner 1994: 26). However, the interplay between acceptance and rejection can be rather delicate. In the first example, M accepted the structure of the construction, but she managed to formulate to it for her own purposes and to extend the amount of film material that was needed for a television programme. When the preliminary component ends with a conjunction, it strongly projects that the second component will bring to completion the turnconstructional unit (for example, see Lerner 1994: 26). However, that does not necessarily mean that each action could be explained in terms of intentionality, or that an explanation is even mandatory. It is more important to locate and identify the specific social constructions and sequential circumstances that somehow provide the opportunities for collaborative productions. (See also Lerner 1994: 31.) Naming football teams is one of these social constructions, as it is identifiable and therefore provides the occasion for specifiable action. In comparison to the previous examples, the following example is somewhat equivocal. The participants in this excerpt are discussing the pharmaceutical industry in Tajikistan, or to be more precise, the lack of it. Speaker N states that she has seen pharmaceuticals that were made in Tajikistan but labelled in Russian. In line 6, speaker E lists other countries (Russia, China) that manufacture the pharmaceuticals prescribed in Tajikistan. Speaker E produces the prepositional phrase iz Rossii, ‘from Russia’, and conjoins it to another prepositional phrase by using the conjunction ili, ‘or’. However, the second prepositional phrase does not contain the actual preposition, as the noun is only declined in the genitive. The collaborative construction is subsequently completed by speaker N by using the conjunction i, ‘and’, to conjoin the prepositional phrases. The preposition iz is also absent from this third phrase, but the noun is declined. Thus, there are three co-ordinated prepositional phrases, even though a preposition is present only in the first phrase. The conjunctions used in the example are also interesting. The first conjunction is the disjunctive ili, ‘or’, and the second is the additive I, ‘and’. 76 5 Collaborative productions (11) TI 14. <13:30> 01 E: 02 N: 03 E: 04 N: 05 E: 06 N: 07 → Sčas u nas v: Tadžikistan-e èto-t Now PRE 1PL.GEN PRE Tajikistan-PREP 3SG-MASC farmacevtiče-sk-ij nik-ogo netu ni če pharmaceutical-ADJ-MASC nobody-GEN NEG NEG PRT tadžik-sk-ij [n:e ↓dela-jut. Tajik-ADJ-MASC NEG do-3PL Nowadays in Tajikistan nobody produces Tajik pharmaceuticals. Net daže ja vide-l-a taki-h le↑karstv-ø NEG even 1SG see-PST-FEM such-PL.GEN pharmaceutical-PL.GEN čto čto tam (.) èt-o proiz↑vodstvo Tadžikistan-a. that there 3SG-NEU production Tajikistan-GEN No. I even saw such pharmaceuticals that it is production of Tajikistan. [Nu: sam d-- da, Well myself d-- yes Well I myself yes, [Tadžik-sk-oe le↑karstvo no vse-taki na rus-Tajik-ADJ-NEU pharmaceutical but neverthless PRE Rus-[na rus-sk-om jazyk-e. PRE Russian-ADJ-PREP language-PREP Tajik pharmaceuticals but neverthless in Russian. [Da sam (-) iz Rossi-i ili Kita-ja [m--, Yes myself (-) PRE Russia-GEN or China-GEN m-Yes I myself (-) from Russia or from China m-, [I Uzbekistan-a mne tože vide-l-a. .h And Uzbekistan-GEN 1SG.DAT also see-PST-FEM And from Uzbekistan I have also seen. No ja udivi-l-a-s’ to čto tadžik-sk-ie But 1SG surpise-PST-FEM-REFL that that Tajik-ADJ-PL le↑karstv-a .h na upakovk-e rus-sk-ij pharmaceutical-PL PREP package-PREP Russian-ADJ-MASC jazyk sto-it (.) počemu rus-sk-ij počemu ne:t language stand-3SG why Russian-ADJ-MASC why NEG tadžik-sk-ij. Tajik-ADJ-MASC But I was surprised that it was Tajik pharmaceuticals but in the package is Russian language why Russian why not Tajik. Although the conjunctions produced in this example are somewhat controversial, their use can be analysed in terms of the overall structure of the conjoined prepositional phrases. The discussion centres on the lack of a pharmaceutical industry in Tajikistan and they begin to name countries that could possibly provide Tajikistan drugs. First, speaker E names two obvious sources, Russia and China. After that, speaker N adds the neighbouring country of Tajikistan, Uzbekistan, even though it is not usually considered to be a country that provides pharmaceutical products. 77 5 Collaborative productions 5.2.2 Agreement within phrases Let us now turn to the agreement phenomena within noun phrases from the viewpoint of the collaborative production. In the NP, the agreement codes the head – modifier relations. Agreement also codes the relationship between the subject and the verb in the clause. These cases will be analysed in the next subsection. A noun phrase (NP) can consist of a head; a noun: a pronoun; or a cardinal (VISK § 442, AG-80 § 1881). Neither the VISK nor the AG-80 uses the concepts pronoun or cardinal (numeral) phrase. This is due to the fact that pronouns and cardinals regarded as nouns when functioning as the head of a phrase. Ordinal numbers are considered to be adjectives (VISK § 771, AG-80 § 1366). Thus, even though both the cardinals and ordinals have the same origin in that they are words connected to counting, they differ grammatically. 25 In the following example, R is describing his home town in Morocco. R states that the town is very beautiful and the climate is pleasant. In line 7, R produces a head, in this case, the numeral kaksikymmentä kahdeksan, ‘twenty eight’, and L adds the modifier astetta ‘degrees’. R accepts the completion by repeating L’s word. (12) FC 63. <20:40> 01 R: 02 L: 03 R: 04 L: 05 R: Minu:-sta on kaunis kaupunki. 1SG-ELA be.3SG beautiful town I think it is a beautiful town. Joo. PRT Yeah. Ö: (1) ö: m: on koska on lähellä ↑ranta-a. VOC (1) VOC m: be.3SG because be.3SG close beach-PTV Because it is close to the beach. Mere-n [ranta-a. Sea-GEN beach-PTV To the seashore. [>Mere-n< mere-n ranta-a. Sea-GEN sea-GEN beach-PTV To the seashore. 25 According to the Academy Grammar of the Russian language, the ordinals and the cardinal odin, ‘one’, decline as adjectives, whereas the cardinals million, ‘million’, milliard, and ‘billion’ decline as nouns (AG80: 1366). However, the cardinals are traditionally characterised as a particular part of speech that is determined by their morphological categories and by the commonality of their syntactic functions. In the nominative and accusative cases, the cardinals govern that inanimate nouns are in the genetive: dva stol-a, ‘two tables-GEN’. In other cases (which are referred to as the oblique cases). the cardinals agree with nouns: dvu-h stol-ov, ‘two-GEN table-GEN (ibid.). 78 5 Collaborative productions 06 L: 07 R: → 08 L: → 09 R: Joo. PRT Yeah. Hyvä: ilma kaksikymmentä: kuusi: (.) kaksikymmentä (.) Good weather twenty six (.) twenty (.) kahdeksan, eight Good weather twenty-six twenty-eight, Astet[-ta. Degrees-PTV Degrees. [Astet-ta kesä-llä ja talvi o:n kahdeksantoista. Degree-PTV summer-ADE and winter be.3SG eighteen Degrees in the summer and in the winter eighteen. In the second example, N and E are discussing the linguistic situation of Tajikistan, or more specifically, the situation of the universities in Tajikistan. N attempts to determine whether it is possible to study all subjects both in Russian and in Tajik. (13) TI 26. <21:40> 01 N: 02 E: 03 N: 04 E: 05 N: → 06 E: → 07 N: 08 E: Èt-o (.) è vy: [s-- smože-te skaza-t’, This-NEU (-) VOC 2PL c-- can-2PL say-INF That erm can you say, [V ↓èto-m god-u, PRE this-PREP year-PREP In this year, »kaki-e predmet-y ili >vs-e vs-e< predmet-y tak what-PL subject-PL or all-PL all-PL subject-PL so naprimer v nacional’n-om univ(h)e- vy by-l-i for.example PRE national-PREP unive2PL be-PST-PL ↑da v gosudarstvenn-om [(.) nacional’n-om. yes PRE state-PREP (.) national-PREP What subjects or all subjects, for example, in the national university, you were in the state national. [Da da. Yes, yes. .h Tam vs-e vs-e èt-i [(.) predmet-y .h There all-PL all-PL this-PL (.) subject-PL There all these subjects [Predmet-y, Subject-PL Subjects, »možno (.) izuč-- uča-t’-sja i: po-tadžik-sk-i i can (.) st-study-INF-PRT both PRE-Tajik-ADJ-ADV and po-rus-sk-i [ili >tol’ko< nekotory-e. PRE-Russian-ADJ-ADV or only some-PL can studied both in Tajiks and in Russian or only some. [Da:. Yes. 79 5 Collaborative productions 09 10 N: 11 E: 12 N: Net net, NEG NEG No, no, [Vs-e? All-PL All? [Predmet-y da. Subject-PL yes Subjects, yes. Aga. Aha. At the beginning of this example, speaker N begins to formulate her question on language possibilities, but she first has to check whether her interlocutor has studied at the national state university. After I’s confirmation, N poses her question again in line 5 and because she has previously mentioned the subject predmety, ‘(school) subjects’, speaker E has an opportunity to complete the compound TCU. N subsequently emphasises her question by repeating the pronoun vse, ‘all’. In the above-mentioned examples, the expectations for the upcoming completion were established prior to it. In the first example, the topic, weather, was mentioned at the beginning of the turn. In the second example, speaker N began to ask a question, but interrupted in order to clarify something first. As she had already pronounced the word predmety, ‘subjects’, in line 3, the speaker was able to use it as a final component. There are certain dissimilarities that separate adjectival phrases from noun phrases. Specific types of words tend to induce collaborative productions. For example, the intensifiers paljon, ‘lot’, and tosi, ‘very’, are often used in Finnish collaborative productions (VISK § 615, see, also Hopper 2011: 23). The participants in the next excerpt are discussing the Dead Sea. In line 2, J begins to describe the Dead Sea by commenting that it is very salty, paljon suolaa, ‘a lot of salt’. While the structure that J uses in line 2 is not very Finnish, and speaker A begins to complete J’s utterance: (14) IC 9. <01:01:50> 01 S: 02 J: → Ole-tte-ko ↑kuul-leet (.) mikä on Kuollut>meri< Be-2PL-Q hear-PRT (.) what be.3SG Dead:Sea minkä↑lainen meri se on. what:kind:of sea 3SG be.3SG Have you heard what kind of sea is Dead Sea. Se on paljon [suol-- suola, 3sg be.3SG lot sal-- salt It is a lot of salt. 80 5 Collaborative productions 03 A: → 04 L: [Suo-- suola-a, Sal-- salt-PTV Salt. ºSuola-a.º Salt-PTV Salt. The use of these intensifiers can also make them recognisable, and they can be used in conversation once their recognisability has been established. In the next fragment, V is telling about his hometown in Siberia. Because the town is located in the interior of Russia, the climate is continental. In line 3, V begins to define the summer temperature and in line 4, B and L complete the clause with the adjective kuuma, ‘hot’. After the first completion, V continues to describe the winter temperature. In line 7, B produces the final component by pronouncing the intensifier tosi, ‘very’, in his own speech. (15) FC 67., 68. <16:57 > 01 V: 02 L: 03 V: → 04 B: → 05 L: 06 V: → 07 B: → Siellä on (2) ilmasto: (1) ilmasto mannerilmasto. There be.3SG (2) climate (1) climate continental:climate There is climate climate continental climate. Mannerilmasto. Continental:climate. Kesä tosi [(.) ↑kuuma. Summer very hot. [Kuuma. Hot. [Kuuma. Hot. Talvi: [(.) tosi kylmä. Winter very cold. [Tosi kylmä. Very cold. The Russian material contains few instances of comparative constructions. Traditionally, they are seen as a part of adverbial phrases, and the type of connection they represent is referred to as primykanie, which literally means ‘joining without declining’ (AG-80 § 1881, 1830). In the following example, N and E are discussing the payment policy of Tajikistan. The participants claim that the salaries are low and that the government does not try to solve this problem with sufficient determination. However, in the year that the conversation took place, the increment in salary had been five percent. In line 4, E continues and begins to explain that doctors have a different type of salary policy: U vračej desjat’ procent, ‘doctors have ten 81 5 Collaborative productions percent’. She subsequently pauses for a short time and speaker N then adds the comparative form vyše, ‘higher’. Originally, speaker E had had another word in mind, bol’še, ‘more’, but she agrees with the offered completion and repeats it.26 (16) TI 33. <1:42> 01 N: 02 03 E: 04 → 05 N: → 06 A ↑vaš-ø prezident naprimer (.) on ne: k But 2PL.POSS-MASC president for.example (.) 3SG NEG PRE naprimer ne skaza-l-ø vot èt-u problem-u:, for.example NEG say-PST-MASC VOC this-ACC problem-ACC But your president, for example, did not talk about this problem. On ne [ne govor-it on ob èt-om. 3SG NEG NEG talk-3SG 3SG PRE this-PREP He does not talk about this. [Na sčet èto-go govor-it na èto-t god PRE count 3SG-GEN talk-3SG PRE this-MASC year uže (.) .h pjat’ procent-ov uže zarplat-u already (.) .h five percent-PL.GEN already pay-ACC u nas b- pjat’ procent-y il-- da pjat’ procent. .h PRE 1PL.GEN b- five percent-PL or yes five percent .h About these talks in this year, already five percent of pay we have five percent. U vrač-ej desjat’ procent [(.) bol’-še vy-še PRE doctor-PL.GEN ten percent (.) big-COMP high-COMP da. yes Doctors have ten percent higher, yes. [Vy-še. High-COMP Higher. Aga horoš-o. Aga good-ADV Aha, okay. In line 4, after speaker E has produced the preliminary component of the co-construction, she pauses for a short time. This could be interpreted as a sign indicating that she is encountering some problems forming her utterances. Traditionally, collaborative productions are claimed to also include word-searches (Lerner 2004a: 261). Structurally, word searches 26 With regard to language use, in this particular example, the grammatical form of the noun after a cardinal number varies. In Russian, the cardinal odin-odna-odno-odni, ‘one.MASC-FEM-NEU-PL’, requires that the noun is in the nominative case either in the singular or in the plural. Cardinals from two to four take the genitive case in the singular, and cardinals from five to twenty take the genitive case in the plural. (AG-80 § 1371.) In line 3, the speaker first uses the correct form of the noun after the cardinal number pjat’, ‘five’: pjat’ procent-ov, ‘five percent-PL.GEN’. Later in that same line, she uses the nominative case in the plural after the cardinal of pjat’ procent-y, ‘five percent-PL.NOM’. The cases used after the cardinals constitute one of the areas that can undergo changes in the Russian lingua franca interaction. 82 5 Collaborative productions enable the recipients to participate, but usually only the searched word is produced. The placement of the contribution can vary, and it can affect how the contribution is produced. (Ibid. 262.) If the offered word is delayed, the current speaker is provided with an opportunity to find the correct word by him- or herself and the offered word is produced as a guess. If the recipient wants to solve the search immediately at the beginning of the search sequence, he or she does not usually frame the contribution as a guess. (Ibid.) In this particular case, the final component by speaker N is produced without any specific guess markers. 5.2.3 Subject NP and the predicate verb An interesting example of participants co-constructing is the subject NP and the finite verb. In both Finnish and Russian, the NP in the nominative case triggers agreement in the finite verb. The data presented by Helasvuo (2004: 1332) contained no excerpts in which the preliminary part contains a subject NP and the final part a verb and object, or a copula and the predicate nominal. Linear organisation– word order–is particularly relevant with regard to co-constructions. Although word order is often characterised as being relatively free in both Finnish and Russian. However, a clear preference for subject – verb ordering occurs in Finnish spoken discourse data. When the subject is first-or second- person, the percentage is greater than 95%. (Helasvuo 2001.) Helasvuo (2003) has also demonstrated “that it is more common to have an intonation unit boundary between the verb and the object than between the subject and the verb.” This indicates that subject and verb are more likely to be grouped together than the verb and the object, which is often assumed to be the way constituents are grouped. (See also Helasvuo 2004: 1332.) In Russian grammar, the structures of these types of co-constructions can be analysed in terms of structure schemas (AG-80 § 1908–1914). The structural core of a clause consists of few words. The co-construction of the subject NP + V equals those instances when there are two components that form a structure schema (AG-80 § 1911). For example, a schema of N 1 - Vf means that there is a noun in the nominative case (N1) and a finite verb (Vf), as in the phrase mal’chik spit, ‘a boy sleeps’. When agreement phenomenon is taken into account, the subject NP defines the gender, number and person of the verb. The following two examples are identical in that their completion consists of a finite verb in the imperfect and the subject pronoun in the first83 5 Collaborative productions person singular. In the first fragment, M and S exchange weekend experiences and in the second, J and G talk about travelling in Russia. (17) FC 105. <12:00> 01 M: → 02 S: → 03 M: 04 B: 05 M: 06 B: Minä, 1SG I, Nuku-i-n. Sleep-PST-1SG Slept. Ei mitään. hehe Nothing. Te on mitä nukku-u: kaksikymmentä tun↓ti-a. 2PL be.3SG what sleep-3SG twenty hour-PTV You is what sleep for twenty hours. Ky-- kaks--, Te-- twen--, Kakskyt e: neljä. Twenty-four. M initiates his preliminary component by pronouncing the first-person pronoun mina, ʻI’. Speaker S completes it with the final component, nuku-in, ‘sleep-PST-1SG. Even though a change in speakership occurs, the projection remains the same. Similar examples also occur in the Russian data. The participants of the next excerpt, J and G, co-construct a subject NP + finite verb construction. When using the above- mentioned concept of structural schemas, this particular example represents the N1 - Vf schema. (18) IS 57. <25:45> 01 J: 02 G: 03 → 04 J: → 05 G: >Da da da< ty kuda ezdi-l-ø v Rossi-i. Yes yes yes 2SG where travel-PST-MASC PRE Russia-PREP Yes, yes, yes, where did you travel in Russia. Nu ja: často byva-ju v Mosk↑v-e no krome ja eščë Well 1SG often be-1SG PRE Moscow-PREP but except 1SG yet (.) ja zaby-l-ø nazvanie mest-a. (.) 1SG forget-PST-MASC name place-GEN Well, I often go to Moscow, but except to Moscow I forgot the name of the place. Nu v: Ivanovsk-oj oblast-i ja, Well PRE Ivanovo-PREP.FEM area-PREP 1SG Well, in the Ivanovo area I, Putešestvova-l-ø. Travel-PST-MASC have travelled. Nemnožko da. A little, yes. 84 5 Collaborative productions 06 J: 07 G: 08 J: 09 G: V Peterburg-e? PRE St.Petersburg-PREP In St Petersburg? Net e↑ščë (.) net no hoč-u. NEG yet NEG but want-1SG No, not yet, but I want to. I teper’ v Finljandi-ju. And now PRE Finland-ACC And now to Finland. ↑Da konečno v Finljandi-ju tože. Yes of course PRE Finland-ACC also Yes, of course, to Finland, too. At the beginning of the excerpt, J asks G about the places he has been to. G responds that he often visits Moscow, but does not remember the names of the other places. In line 3, G begins his preliminary component by defining that the Ivanovo area is the place where he has travelled. J pronounces the final component and adds the verb putešestvova-l-ø, ‘travelPST-SG’ (cf. Andreeva 2007: 39). J has prefaced his completion already at the beginning of the example so that he can do it easily. The two examples above are not overly complex. The first one includes an intransitive verb and in the second, G has already produced the locative adjunct before proceeding to the subject of the clause. Therefore, the completion of the clause is projected at the beginning of G’s turn. In the Finnish example, the colloquial personal pronoun mä, ‘I’, was followed by a finite verb in the imperfect in the first-person singular. Due to differences in the morphosyntactic categories in Finnish and Russian, the connection in the Russian example is not obvious. In the imperfect, the Russian morphology does not differentiate between persons, but between the singular and plural. Therefore, the form of the finite verb is accurate and because the speaker of the first component used ja, the ‘I’ pronoun. The predicative is in the masculine form because the person that they are talking about is a man. Example 19 features speakers N and E constructing a clause that consists of a subject vse, ‘all’, and of a finite verb, izučali,‘studied’, which is in the imperfect. This fragment has the finite verb in the plural. The collaborative production is rather unsuccessful due to some overlapping, and the speaker of the preliminary component does not indicate explicitly how she reacts to the completion. 27 27 With regard to language use, one of the tendencies in lingua franca interaction is the use of semantically close words. In this particular example, speaker I uses two constructions that are related to studying. In line 4, she begins her utterance by cutting-off the word my iz-, ‘we stu-ʼ, acknowledging that the word choice was not perhaps accurate. She then produces the construction of učili vse v škole, ‘my studied all at school’. 85 5 Collaborative productions (19) TI 31. <6:45> 01 N: → 02 E: → 03 04 E: 05 Čtoby ↑vs- [(.) daže v kišlak-ah (.) mog-l-i, That all-PL (.) even PRE kišlak-PREP.PL (.) can-PST-PL That everyone even in villages could, [Izuča-l-i rus-- da: da: est’. Study-PST-PL Rus-- yes yes be.3SG. Studied Rus- yes, yes, is. (1) È: a togda kogda my iz-- è: (.) uči-l-i vs-e VOC and then when 1PL stu-- VOC (.) study-PST-PL all-PL v ↑škol-e, .h PRE school-PREP And when we studied all at school, tam po-mo-emu roditel-i ne: ne razreša-l-i čtoby There PRE-1SG-DAT parent-PL NEG NEG allow-PST-PL that e-go dočka i:li e-go .h é ↑syn (.) 3SG.MASC-GEN daughter or 3SG.MASC-GEN .h VOC son (.) izuča-t’ v universitet-ah. study-INF PRE university-PL.PREP There, in my opinion, parents did not allow (that) his daughter or his son (would) study at universities. One of the interesting features of Finnish grammar is the use of a specific (auxiliary) negative verb when person is expressed in the negative verb. For example, e-n ‘I-NEG.1SG’, e-t ‘you-NEG.2SG’ etc. The difference between the present and past tenses is marked in the main verb. (VISK § 108.) In the following example, the participants co-construct subject-finite verb construction in which the finite verb is negative. Participants are discussing a study group that they have founded in order to facilitate their Finnish studies. L suggests that perhaps they could also take Muhammad with them. In line 1, M explains that they have told him about the group. Then, in line 4, L completes the S+V construction by pronouncing the verb in negation. The negative verb ei, ‘NEG.3SG’ indicates the person and the form of the main verb halua, ‘want+PRES’. (20) FC 97. <03:26> 01 L: Otta-kaa ↑Muhammed mukaan Muhammed. Take-IMP.2PL Muhammed with Muhammed Take Muhammed with you Muhammed. In line 5, she returns< to the original construction and produces izučat’ v universitetah, ‘to study at universities’, although the verb izučat’ can be used in constructions such as izučat’ biologiju, ‘to study biology’, that is ‘to study a subject’. The semantically appropriate construction would be učit’sja v universitetah, ‘to study at universities’. This might be explained by the fact that the collaborative production in lines 1–2 consists of the verb izučat’ and of the cut-off word rus-, ‘Russian’. Thus, the structure has been introduced to the conversation before and this makes it accessible. 86 5 Collaborative productions 02 B: 03 M: → 04 L: → Jos halua-n (-) If want-1SG (-) If I want Minä sano-n häne:-lle e: hän, 1SG say-1SG 3SG-ALL VOC 3SG I say to him he, Ei halua. NEG.3SG want. does not want to. This example differs slightly from the two previous examples. In those examples, the subject was expressed by the first-person pronoun and even though the change of speakership occurred, the predicate remained in the first-person form. These collaborative productions are not exceedingly complex. The constructions simply consist of a noun phrase and a finite verb. Nonetheless, those types of co-constructions make the anticipatory completion possible. The first two examples demonstrated that the final component grammatically continued the projection of the preliminary component. In other words, although the opportunity of speakership occurred, the coparticipant produced a predicate verb that conjugated with the subject that was expressed by a personal pronoun. In the last two examples, the projection was in effect, but as the subject was expressed by a quantor pronoun (vse, ‘all’) or a personal pronoun (hän, ‘he’), the final component could not have possessed any other form than the one produced. 5.2.4 Objects, other complements required by predicate verbs, and predicatives Let us now turn to concentrate on the participants who construct a predication. In the first example, the participants are discussing their breakfast habits. T is telling about her husband, Raimo, and about their children. In lines 1–7, she talks about Raimo eating porridge and drinking some tea with honey. In line 8, T begins to recount that their children eat only sweet yoghurt. In line 9, speaker A adds the object, mehua, ‘juice’, to the construction that was initiated by speaker T. (21) MV 50. <44:08> 01 T: 02 T: >Raimo< syö perusteellise-sti. Raimo eat.3SG thorough-ADV Raimo eats thoroughly. .hh Kaurapuuro-a (.) .hh Oatmeal:porrige-PTV .hh Oatmeal porridge 87 5 Collaborative productions 03 A: 04 T: 05 T: 06 07 A: T: 08 T: → 09 A: → 10 T: Niin. Yes. S-- pää-llä ö: juusto-a si-tä m ↑fetajuusto-a. S-- PRT VOC cheese-PTV 3SG-PTV VOC feta:cheese-PTV S- top of it cheese that feta cheese. Joo sitten pää-llä tee-tä missä on hu↓naja-a. Yeah then PRT tea-PTV where be.3SG honey-PTV Yeah, then tea with with honey. (-) J(h)oo elikkä todella vaikka-pa hän on ↑suomalainen. Yeah well really although-CLT 3SG be.3SG Finn Yeah, well really although he is a Finn. Ja >sitten meidä-n lapse-t< ne syö vain makea And then 1PL-GEN child-PL 3PL eat.3SG only sweet jugurtti-a ja vain to-ta juo-vat, (.) yoghurt-PTV and only well drink-3PL And then our children, they eat only sweet yoghurt and just well, drink, Mehu-a. Juice-PTV Juice. Mehu-a. Juice-PTV Juice. In some cases, the final component can slightly overlap the preliminary component. However, Sacks, Schegloff & Jefferson (1974: 708) defined a smooth transition (from one turn to a next), and this covers those transitions with no gap and no overlap and the transitions characterised by a slight gap or overlap. The second example contains two collaborative productions. Let us now turn our attention to the first collaboration. In this example, the participants are discussing the delicacies of Finnish cuisine: (22) MV 43., 44. <01:00:10> 01 S: 02 → 03 A: → >Ni ni< valmista-a-k-s se sii-- teke-e-k-s Well well prepape-3SG-Q-CLI 3SG it-- make-3SG-Q-CLI se sii-t jauhelihakastikkee-n. 3SG it-ELA mincemeat:sauce-GEN Well, does he prepare it does he makes mincemeat sauce of it? Vai teke-e-k-s se sii-t jotai .hh [(.) Or make-3SG-Q-CLI 3SG it-ELA something Or does he make something of it [Muruke. Meatloaf. ((17 LINES REMOVED)) 88 5 Collaborative productions 21 S: 22 T: 23 S: → 24 T: → 25 S: [Ja sit se, And then 3SG And then it, [Joo. Yeah. »>laitta-a sii-hen< jotain [.hh put-3SG it-ILL something puts there something [päälle. PRT on top of it. vai jotai päälle. Or, something on top of it. The subject that is discussed in this excerpt is mincemeat, which o the participants claim is a very typical food in Finland. At the beginning (lines 1–2), the current speaker, S, suggests that perhaps immigrants are accustomed to preparing some sort of mincemeat sauce that is similar to what Finns would prepare themselves. In line 3, speaker A proposes that another possible dish could be meat loaf. As we have already discussed in terms of the conversational structure, jointly produced constructions are located in different positions. These positions are short pauses (micro-pauses less than 0.2 seconds) and inhalations. Jointly produced constructions are usually produced so that they do not create a disturbance in the ongoing conversation. Furthermore, lexical markers are possible, for example, in Finnish jotain, and in Russian chto-to, ‘something’. As we can see, the completion is placed near the end of the unit. Syntactically, this addition is enabled by the disjunctive conjunction vai, ‘or’, which is pronounced in the beginning of the turn. In addition to the lexical NPs, the infinitive phrases can also work as objects (VISK § 499). The following example consists of a verb haluta, ‘to want’, which requires an infinitive phrase as an object. This example features B who is sketching a plan for how the participants of the Finnish course would spend their final day together. In lines 1–2, B and L jointly produce the structure kuka haluaa luistella, ‘who wants to skate’. B accepts the infinitive phrase by repeating it: (23) FC 92. <07:02> 01 B: → Per↑jantai-na (1) kaikki teke-e ruoka ja syö-mme Friday-ESS (1) everyone prepare-3SG food and eat-1PL täällä ja menn-ää ↑ulos ja: kuka: halua-a, here and go-4 out and who want-3SG On Friday, everybody prepares food and we eat here and we go out and who wants to, 89 5 Collaborative productions 02 L: → 03 B: Luistel-la. Skate-INF to skate. Luistel-la. Skate-inf to skate. In some examples, government phenomenon can also trigger the construction of grammatically appropriate words or the participants can negotiate the correctness. This is understandable because the construction has otherwise been executed, but it needs an object to be complete. In these cases, the grammar can be considered to be real-time phenomenon (Hopper 1987: 141), which means that the syntactic structures unfold in real time (Auer & Pfänder 2011: 3, Hopper 2011: 27). Clauses can also emerge as syntactic units and these can likewise be achieved through the participants’ mutual activity (Helasvuo 2001b: 33–35). In the example that follows, M is telling his interlocutors about football, which is his hobby. In line 1, he produces the word ihmise-t, ‘people-PL’, and produces the grammatically incorrect form pelaa, ‘play-3SG’. He then pauses and produces the right form, pelaavat, ‘play-3PL’. A two-second pause then occurs before R adds the word jalkapallo-a, ‘football-PRT’, in line 3. After the offered final component, M selects the last word of the compound noun and declines it in a different case, namely in the genitive: pallo-n, ‘ball-GEN’. At this point, speaker B wants to express his opinion regarding the correct declension and begins to elaborate on his own version. He first produces the form pallo, ‘ball-NOM’, and then he forms the postpositional phrase pallon kanssa, ‘with a ball’ (literally ‘ball-GEN with’). Even though R offered the grammatically correct form in line 3, M accepts the form pallon kanssa in line 6 by repeating it. (24) FC 111. <08:25> 01 M: 02 03 R: 04 M: 05 B: Nyt ihmiset pela-a (.) pelaa-vat Now people-PL play-3SG (.) play-3PL Now people play (2) Jalkapallo-a. Football-PRT. Football. Pallo-n. Ball-GEN. Football. Pallo: p(h)allo-n kanssa. Ball ball-GEN with A ball with a ball. 90 5 Collaborative productions 06 M: <Pallo-n kanssa.> Ball-GEN with With a ball. As in the previous example, the participants are also able to co- construct different types of structures in which the predicate verb requires that the complement be in a specific case. In the following example, M begins his utterance with jos kysymme, ‘if we ask’, which is subsequently completed by speaker L by saying Naukolta, ‘from Nauko’. Syntactically, the final component fits the preliminary component. M accepts this complement by repeating it. (25) FC 65. <19:35> 01 M: → 02 L: → 03 M: J:os jos kysy-mme, If if ask-1PL, If if we ask, Nauko-lta. Nauko-ABL From Nauko. Nauko-lta. Nauko-ABL From Nauko. There are also similar cases in the Russian data. In the following example, speaker E is telling about her daughter. They live in Tajikistan, but E wants her daughter to learn Russian. In line 2, speaker N overlaps with speaker E and begins her utterance of čtoby ona sami govorila, ‘that she herself would speak’, which is then completed by speaker E in line 3, porusski, ‘in Russian’. In this example, speaker E has already described the whole situation in her first utterance. Speaker N then reformulates speaker E’s utterance and speaker E adds the adverb po-russki. Speaker E accepts the suitability by repeating that word. 28 (26) TI 30. <7:57> 01 E: Tol’ko u ne-e četyre mesjac-a po-èt-omu kogda Only PRE 3SG.FEM-GEN four month-GEN PRE-it-DAT when u ne-e ↑bud-et god (.) ja postavi-l-a PRE 3SG.FEM-GEN will.FUT-3SG year (.) 1SG make-PST-FEM čto-by ↑on [<on-a> tože (.) rus-sk-ij jazyk. that-CLI 3SG.MASC 3SG-FEM also (.) Russian-ADJ-MASC language 28 With regard to language use, the use of third-personal pronouns fluctuates in this particular example. In Russian, the third-personal pronouns indicate the gender of the referent. In line 1, she first uses the personal pronoun on, ‘he’, then she produces the correct pronoun, ona, ‘she’. Moreover, the gender category in Russian is one of the categories that can undergo changes in Russian lingua franca interaction. 91 5 Collaborative productions 02 N: → 03 E: → 04 H: Only she is four months old that is why when she turns one year I decided that he she also (.) Russian language. [Čto-by è: čto-by on-a >sam-i< That-CLI VOC that-CLI 3SG-FEM self-PL govori-l-a by [po-rus-sk-i. talk-PST-FEM CLI PRE-Russian-ADJ-ADV That that she herself would speak in Russian. [Po-rus-sk-i. PRE-Russian-ADJ-ADV In Russian. Aga horoš-o. Aha good-ADV Aha okay. The following example is somewhat similar to the previous excerpt. Speaker N is the same person as in the previous example, but the interlocutor in this excerpt is different. The participants are discussing the different areas of Tajikistan and the languages that are spoken in those areas. N states that she has noticed that in some areas, the Tajik and Uzbek people use either Tajik or Uzbek, but not Russian. (27) TI 24. <22:21> 01 N: 02 A: 03 N: 04 → 05 06 A: → .h I mnogi-e tadžik-i >govor-jat po<-uzbek-sk-i tam. .h And many-PL Tajik-PL speak-PL3 PRE-Uzbek-ADJ-ADV there And many of the Tajik people speak in Uzbek there. [Da. Yes. [Tak čto ja ma:lo slyša-l-a rus-sk-ij tam. So 1SG little hear-PRET-FEM Russian-ADJ-MASC there So I heard very little Russian there. Tam .h uz↑bek-i tad↑žik-i vmeste govor-jat ili There Uzbek-PL Tajik-PL together speak-PL3 either po-tadžik-sk-i i[li po-uzbek-sk-i. PRE-Tajik-ADJ-ADV or PRE-Uzbek-ADJ-ADV There the Uzbek people and the Tajik people converse either in Tajik or in Uzbek. [po-uzbek-sk-i da. PRE-Uzbek-ADJ-ADV yes. Da: u nas rus-sk-ij malo (.) v gorod-e Yes PRE 1PL.GEN Russian-ADJ-MASC little (.) PRE town-PREP Dušanbe po-↑mo-emu oni (1) tadžik-i vs-ë rav↑n-o Dušanbe PRE-1SG-DAT 3PL (1) Tajik-PL all-NEU same-NEU oni razgovariva-jut po-rus-sk-i. 3PL talk-3PL PRE-Russian-ADJ-ADV Yes, we have very few Russians in the town of Dušanbe. I think that the Tajik however talk in Russian. Although Tajik is the official language of Tajikistan, the country has a relatively significant Uzbek-speaking minority. Russian serves as an 92 5 Collaborative productions interethnic language although, as we see in this example, other languages can be used as well. N states that many of the Tajik people actually know Uzbek (line 1) and when a Tajik and an Uzbek are speaking to each other, they may use either Tajik or Uzbek (line 4). At the end of the line 4, N is beginning to offer these two languages, but she does not reach the end of her turn before the interviewee adds the second language (po-uzbekski, ‘in Uzbek’; line 5). Syntactically, N’s turn is constructed by using coordinating double conjunctions (ili-ili, ‘either-or’), which join two equal clauses. Thus, by interpreting the structure of the turn, the interviewee can complete it. On the clause level, participants can also co-construct different types of characterising clauses, and these typically function as assessments (Helasvuo 2004: 1323–1325). The preliminary component consists of a noun phrase and copula, and the final component consists of either a noun phrase or an adjective phrase. The next excerpt is a textbook example of characterising clauses 29. The participants are discussing the final day of their Finnish course and what they could do after they receive their diplomas at a ceremony. In line 8, B begins his turn with koska tämä on, ‘because this is’, and R completes it with the adjective phrase viimeinen päivä, ‘the last day’. B repeats the phrase and accepts it as suitable. (28) FC 94. <06:22> 01 B: 02 L: 03 B: 04 R: 05 B: 06 L: Sitten maanantai-na (.) >ja: ja ja< jälkeen, Then Monday-ESS (.) and and and after Then on Monday and after, Todistus-t-en jao-n jälkeen. Diploma-PL-GEN presentation-GEN after After giving out the diplomas. Todistus-t-en >mene-mme< kaikki jos halu-a men-nä Diploma-PL-GEN go-1PL all if want-3SG go-INF ravintola-an jos halua-mme mennä-mme restaurant-ILL if want-1PL go-1PL After getting our diplomas, we go all if want to go to a restaurant if we want to go Kahvi-lle. Coffee-ALL For a coffee. ↑Elokuvateatteri-in. Movie:theatre-ILL To a movie theatre. Niin. Yes. 29 There are actually two co-constructions (also in lines 3-4), but I will analyse the first one in the next subsection 5.2.2.4. 93 5 Collaborative productions 07 B: 08 → 09 R: → 10 B: Ja baari-in juo kahvi-a puhut-aan. And bar-ILL drink.3SG coffee-PTV talk-4 And to a bar, drink coffee, let’s talk. Kosk(h)a tämä: on on, Because this be.3SG be.3SG Because this is, Viimeinen päivä. Last day. The last day. Viimeinen päivä. Last day. The last day. At the end of B’s preliminary component, the finite verb on, ‘is’, is reduplicated. This could also be a sign of some types of problems in formulating utterances. It is also important to note that even though they are discussing the next week, B uses the pronoun tämä, ‘this’. However, the use of the pronoun does affect the outcome and R produces the final component viimeinen päivä, ‘final day’. The participants in the following example are discussing different reality television programmes and they think that those programmes divide television viewers into two separate categories. There are those viewers who like them while there are others who cannot stand watching them. Before this particular excerpt, speaker S has stated that a certain group of people watches television only for fun and that they love these reality television programmes and to them, television is only bread and circuses, ‘panem et circences’. M tries to defend this group and argues that even though S might think that those programmes are rubbish, others might appreciate this as a good programme. However, in line 2, S utters the word tositeeveetä, ‘reality television’, and makes it clear that as far as he is concerned, both rubbish and reality television are one and the same. (29) MV 40. <01:33:50> 01 M: → 02 S: → 03 M: Su-n miele-stä ↑roska-a hän-en miele-stä-än 2SG-GEN mind-ELA rubbish-PTV 3SG-GEN mind-ELA-PX se on [hyvä ohjelma.] 3SG be.3SG good programme You think it is rubbish he thinks it is a good programme. [Tositeevee-tä.] Reality:television-PTV Reality television. Eri: mielip(h)ite-i-tä on. Different opinion-PL-PTV be.3SG There are different opinions. 94 5 Collaborative productions Finnish and Russian languages differ with respect to the verb ‘be’ in the present tense, but similar co-constructions can be found in both languages. The Finnish copula verb is expressed in all tenses, but in Russian, the copula is zero-marked in the present tense. In Finnish, the copula shows tense, person and the number of the subject. In Russian, the verb byt’, ‘to be’, is usually zero-marked in the present tense. Therefore, a copulative clause consists of a subject and a predicate nominal that is indicated by a nominal part. Structurally, the next three examples are similar. In the first two examples, there is a pronoun èto, ‘it’, as well as the short forms of the adjective nevežlivo, ‘impolite’, which functions as an adverb, as well as estestvenno, ‘natural’, which functions as a predicative. In the first example, speaker H displays agreement with the agreement-token da, ‘yes’ (Szczepek 2000b: 26). (30) UM 41. <01:40:43> 01 N: 02 ?: 03 N: 04 05 N: → 06 A: → Da ↑èt-o naverno ne zna-ju (-) nu: naš-ej Yes 3SG-NEU probably NEG know-1SG (-) well our-GEN kul’tur-y da. culture-GEN yes Yes it is probably I do not know well of our culture yes. Da. Yes. DA. Yes. (3) Ne govor-im ne govor-im tak @ja:@ da èt-o ne NEG talk-1PL NEG talk-1PL so 1SG yes 3SG-NEU NEG zna-ju èto [(.) know-1SG 3SG We do not talk we do not talk like that I yes it is I do not know it is [Ne↓vežliv-o. Impolite-NEU Impolite. (31) DM 54. <01:31> 01 H: 02 03 H: On-a nič-ego ne ume-l-a krome kovr-y 3SG-FEM nothing-GEN GEN can-PST-FEM except carpet-PL dela-t’. do-INF She could not make anything els but carpets. ((laughter)) Kovr-y dela-t’ zna-l-a on-a (.) tabasaransk-ie Carpet-PL do-INF know-PST-FEM 3SG-FEM Tabasaran-PL kovr-y. carpet-PL Carpets she knew how to do Tabasaran carpets. 95 5 Collaborative productions 04 M: 05 H: → 06 M: → ↑I: nosk-i vjaza-t’ ↑čulk-i vjaza-t’. And sock-PL knit-INF stocking-PL knit-INF And to knit socks and stockings. Nu: <èt-o> èt-o (.) Well 3SG-NEU 3SG-NEU (.) Well it it È↓t-o estestvenn-o. 3SG-NEU natural-NEU It is natural. The example that follows has a subject NP proiznošenie, ‘pronunciation’, and a predicative consisting of an adjective phrase sovsem drugoe, ‘completely different’: (32) UM 44. <01:27:18> 01 N: 02 ?: 03 N: 04 ?: 05 N: → 06 М: → Da i e↑ščë tam n: nekotory-m meša-et tot ↓fakt čto Yes and yet there s- some-DAT bother-3SG that fact that tot è tot šved-sk-ij ja↓zyk kotoro-that VOC that Swedish-ADJ-MASC language whic->kotoro-mu ↑uča-t’-sja< v Finljandi-i škol’nik-i: i which-DAT teach-INF-PRT PRE Finland-PREP pupil-PL and gimnazist-y (.) èt-o ↑fin-sk-ij variant student-PL 3SG-NEU Finnish-ADJ-MASC variant šved-sk-ogo jazyk-a. Swedish-ADJ-GEN.MASC language-GEN Yes and some are bothered by that fact that that the Swedish language that is taught in Finland at schools to pupils and sixth formers is the Finnish variant of the Swedish language. Ugu. Uhhuh. Kotor-yj (.) è: nu: uže dovol’no <mnogo> Which-MASC (.) VOC well already quite lot otliča-et-sja ot to-go švedsk-ogo jazyk-↑a differ-3SG-PRT PRE that-GEN Swedish-GEN language-GEN kotor-yj na kotor-om govor-↑jat v Šveci-i. which-MASC PRE which-PREP.MASC speak-3PL PRE Sweden-PREP Which well already pretty much differs from that Swedish language which in which they speak in Sweden. Ugu. Uhhuh. To est’ (.) proiznošenie uže [(.) drug-oe. It be.3SG pronunciation already (.) different-NEU That is the pronunciation is already different. [>Sovsem< drug-oe. Completely different-NEU Completely different. The participants in the Finnish examples appropriately used the verb ‘be’, but in the Russian examples, the participants did not even attempt to use any 96 5 Collaborative productions type of the be-verb. While the structures of Finnish and Russian differ in this respect, the participants have the same means of co-constructing utterances. It is possible that participants co-construct the syntactic structures that are more complex. Those structures consist of two parts and the first, or the preliminary component, is a full lexical noun phrase that one speaker produces. In contrast, the second part, or the final component, is a clause, comprising a co-indexical pronoun with the NP that is produced by another speaker. (VISK § 1018, Pekarek Doehler 2011.) This phenomenon can be referred to as left-dislocation, or it can be described as a ‘split to the left’. However, these characteristics only apply when describing the written forms of language. When analysing spoken language, these categories can be somewhat undemonstrative. (See, e.g. Auer 2009: 4 for challenging the terminology used in the research of spoken language.) Finnish grammar (VISK § 1018) emphasizes the turn organisational context of those constructions and uses the term ‘split’ to refer to the beginning because they are placed at the beginning of the turn. Pekarek Doehler (2011) has analysed the left-dislocated constructions in French conversations and divides the sequential structures of these constructions into three parts. The first part is a first speaker who syntactically breaks off or hesitates in formulating his or her turn. The second part consists of the second speaker producing an anticipatory completion that is syntactically designed to fit the first speaker’s utterance. Third, the first speaker can express either an acceptance or a refusal. Interactionally, by way of the final component in a range of actions can be accomplished, such as providing help, displaying knowledge or enacting alignment. (Pekarek Doehler 2011: 55, 59.) Pekarek Doehler argues that an NP does not necessarily strictly command the possible format of the anticipatory completion. Doehler states that this indicates that both grammatical projection and its interpretation are context sensitive and are heavily dependent on the sequential environment. It is also possible that other-completion occurs “beyond what can be interpreted as the preliminary component.” (Ibid. 56–57.) Example 33 features N, who initiates her turn by inquiring from E whether their schoolbooks and lectures were all in Russian. She produces two lexical NP’s at the beginning of her turn: vaši vse èti učebniki, ‘all of your schoolbooks’, and lekcii,‘lectures’. In line 2, E starts the final component simultaneously with N’s utterance and confirms that all their lectures and books were in Russian: 97 5 Collaborative productions (33) TI 13. <15:31> 01 N: → 02 E: → 03 N: .h Vaš-i ↑vs-e èt-i učebnik-i .h i: lekci-i .h 2PL.POSS-PL all-PL this-PL schoolbook-PL .h and lecturePL ↑vs-e [(.) by-l-i na ↓rus-sk-om. all-PL (.) be-PST-PL PRE Russian-ADJ-PREP Your all these schoolbooks and lectures all were in Russian. [Vs-e by-l-i ↑po rus-sk-im All-PL be-PST-PL PRE Russian-ADJ-INST.MASC jazyk-e. language-PREP All were in the Russian language. Aga. Aha. Although Pekarek Doehler’s article discusses French conversations, the same observations also appear to be valid for this Russian example. According to Pekarek Doehler (2011: 56–57), these types of NPs in French normatively project several potential follow-ups. One of them is a full clause yielding a left-dislocation or hanging topic, which also occurs in this example. Nonetheless this constitutes only one example. In order to analyse phenomenon more precisely, it requires more research. 5.2.5 The semantic motivation that triggers collaborative productions: the Finnish verb mennä, ‘to go’ + direction I will address one interesting case of Finnish that resembles government instances, although they are considered to be grammatically different. In Finnish conversations, there are four cases consisting of a verb mennä, ‘to go’. This type of verb requires that its complement possesses a certain direction, not a particular case. In Finnish, this direction involves the illative, allative, or the complement can be an adverbial. Grammatically, those are not seen as a part of a government phenomenon (VISK § 1225). In the first example, the participants are conversing after their Finnish class. At the beginning of this excerpt, speaker M tells that they will go study afterwards. In line 5, L checks that she has understood the utterance. In lines 7–8, speakers M and R collaboratively produce the structure tänään me menemme kirjastoon, ‘today we go to the library’: (34) FC 96. <04:06> 01 M: 02 R: Tänään me: mene-mme: ö: (2) opiskele-maan. Today 1PL go-1PL VOC (2) study-INF Today we go study. A:. VOC 98 5 Collaborative productions 03 M: 04 05 L 06 R: 07 M: → 08 R: → 09 M: 10 11 S: M: 12 L 13 R: 14 M: Ö:, VOC (2) Te mene-tte o↑piskele-maan. 2PL go-2PL study-INF You go study. Ö:. VOC Tänään me mene-mme, Today 1PL go-1PL Today we go, Kir↓jasto-on. Library-ILL To the library. Kirjasto-on. Library-ILL To the library. (-) Kuka: halua-a, Who want-3SG Who wants to, Mi-tä te mene-tte ↑opiskele-maan. What-PTV 2PL go-2PL study-INF What are you going to study. Puhe↓kieli. Spoken:language. Puhekieli. Spoken:language. The formation of the joint utterance is made possible by the semantic motivation. The verb mennä, ‘to go’, requires a complement that indicates direction. Indicating direction is anticipatory, and participants are therefore able to produce constructions together. In the first example and in the next one, the complements are expressed by full NPs. In the next case, three participants contribute to constructing the conversation. Speakers B and L begin the sequence in lines 1–2. In line 3, B begins to describe what the study group would do on Monday after they have received their diplomas: jos haluamme mennämme, ‘if we want we go’, which is then completed by speaker R in line 4: kahville which is literally ‘to coffee’. However, this was not what the original speaker had in mind (cf. Sczcepek 2000b: 21). In line 5, B rejects the offer made by R and adds the complement he had been envisioning, elokuvateatteriin, ‘to a movie theatre’: (35) FC 93. <06:22> 01 B: Sitten maanantai-na (.) >ja: ja ja< Then Monday-ESS (.) and and and Then on Monday and after, 99 jälkeen, after 5 Collaborative productions 02 L: 03 B: 04 R: 05 B: 06 L: Todistus-t-en jao-n jälkeen. Diploma-PL-GEN presentation-GEN after After giving out the diplomas. Todistus-t-en >mene-mme< kaikki jos halu-a men-nä Diploma-PL-GEN go-1PL all if want-3SG go-INF ravintola-an jos halua-mme mennä-mme restaurant-ILL if want-1PL go-1PL After the diplomas we all go if want to go to a restaurant if we want to go Kahvi-lle. Coffee-ALL For a coffee. ↑Elokuvateatteri-in. Movie:theatre-ILL To a movie theatre. Niin. Yes. The original speaker rejects the final component of the collaborative production, and the original speaker does this by the power invested with him by the overall structure of co-constructions. However, this demonstrates that constructing collaborative productions is not merely a simple task. When the form of the complement is determined by government phenomena, the case is determined in advance. For example, if the verb is kysyä, ‘ask’, it requires that the complement be in the elative case: kysyä Nauko-lta. When the verb requires that its complement possesses a certain direction, there are many possibilities. One can mennä kahville ‘go to coffee’, or ‘go swimming’, mennä uimaan In other words, the complement can be a noun phrase, an adposition phrase, or an infinitive phrase (VISK § 1225). It is also possible that similar constructions occur over a within a short time span. This is because once the common ground is established, it can be used again. The following example contains two constructions that have the verb mennä. In the first excerpt, the final component consists of an adverb ylös, ‘up’, and in the second excerpt, the final component is another adverb, ulos, ‘outside’. (36) FC 58. <01:04:10> 01 B: → 02 L → Miksi: (1) miksi: (.) tuli (.) tulipalo mene-e Why (.) why (.) fire (.) fire go-3SG [(.) ylhää-llä. (.) high-ADE Why fire fire goes up. [>Ylös< ylös. Up up. 100 5 Collaborative productions 03 B: 04 05 L 06 B: 07 R: 08 M: 09 R: → Ylös ei: ylös ylhää-llä kaikki kerrokse-n koska se on Up NEG up high-ADE every floor-GEN because 3SG be.3SG ensimmäinen kerros. first floor Up not up high every floor because it is the first floor. (1) Mhy. Uhhuh. Ja jos on <tuli tuli> (2) And if be.3SG fire fire (2) And if it is fire fire Mutta savu ↑voi men-nä ulos. But smoke can.3SG go-INF out But the smoke can go out. Men-nä ulos. Go-INF out Go out. (---) 10 L: → No mutta nor↑maalise-sti e au-- jos minä katso-n normal-ADV VOC he-- if 1SG watch-1SG tule-en automaattise-sti minä: ↑halua-n men-nä [ulos. fire-ILL automatical-ADV 1SG want-1SG go-INF out Well but normally if I look at fire automatically I want to go out. [Ulos joo. Out yeah. Well but Even though the structures of this subsection were not exceedingly complex, it demonstrates that in addition to syntactic criteria, participants are also able to observe semantic categories. As a semantic category, a frequent category that can be used in collaborative productions is direction. 5.3 Participants co-construct a complex sentence Next, I will present examples of participants co-constructing a complex sentence. The Finnish and Russian data contain only a few of these cases. Structurally, these complex sentences approximate those compound turnconstructional units that consist of if-then structures (Lerner 1991: 442). As in the previous examples, the important aspect for the analysis of these cases is the two-part utterance format. When taking into account complex sentences, structurally they consist of two or more simple sentences that are conjoined by conjunction that is usually situated in the preliminary part. There are only few cases of complex sentences in the data that include a main clause and a subordinate clause. Let us consider the following example in which the subordinate clause functions as an object: 101 5 Collaborative productions (37) FC 98. <03:26> 01 L: 02 B: 03 M: → 04 L: → 05 M: 06 L 07 M: → 08 09 M: 10 L → 11 M: Otta-kaa ↑Muhammed mukaan Muhammed. Take-IMP.2PL Muhammed with Muhammed Take Muhammed with you Muhammed. Jos halua-n (-) If want-1SG (-) If I want Minä sano-n häne:-lle e: hän, 1SG say-1SG 3SG-ALL VOC 3SG I say to him he, Ei halua. NEG.3SG want. does not want to. Ei: mutta ei minä e-n sa↑no-nu e-n NEG but NEG 1SG NEG-1SG say-PCP NEG-1SG Now but now I did not say did not say. <E-n sano-nut.> NEG-1SG say-PCP I did not say. Hän ei sano-nut mutta ↑minä kysy-i-n 3SG NEG.3SG say-PCP but 1SG ask-PST-1SG minä: (.) halua-n 1SG (.) want-1SG He did not say, but I asked him I want (1) Ö: VOC Että hän tule-e. That 3SG come-3SG That he comes. ↑Että hän tule-e. That 3SG come-3SG That he comes. sano-nut. say-PCP häne-ltä e: 3SG-ABL VOC The final component in these cases is more complex in nature because it consists of more than only one word, but the speaker who produced the preliminary component has the opportunity to evaluate the suggested final component and to continue speaking. In this example, speaker M repeats the final component and approves its suitability. When analysing co-ordinate complex sentences, different types of conjunctions can function as markers that enable collaborative productions. In the first example, the complex sentence is constructed by using an adversative conjunction a, which can be translated as ‘but’ in this particular case. However, this conjunction is not as contrastive as the conjunction no. Instead, the conjunction a shows that the contents of the co-ordinate sentences are somewhat opposing, but not fully contradictory. 102 5 Collaborative productions The participants in this excerpt are discussing that education was free in the old days and therefore everyone had an opportunity to study and to receive an education. Today, however, not everyone has the opportunity to educate themselves because education is expensive. In line 16, speaker E begins to describe the current situation as den’gi est’ a, ‘there is money but’, which is then completed by speaker N as horošo ne učatsja, ‘(they) do not study well’. The conjunction is situated at the end of the preliminary component. While this makes the completion possible, it requires another main clause to complete it. (38) TI 6. <13:10> 01 N: 02 E: 03 N: 04 E: 05 N: 06 E: 07 N: 08 E: 09 N: 10 E: 11 N: 12 E: 13 N: 14 E: Oni govor-jat čto kogda ↑my (-) besplatn-o by-l-o. They talk-3PL that when 1PL (-) free-NEU be-PST-NEU They say that when we were it was free. ↑A:. VOC .h A sejčas [(.) govor-jat <mnogo sto-it.> .h But now (.) talk-3PL lot cost-3SG But now they say it costs a lot. [sčas (-) dollar [(.) (-) dollar? Now (-) dollar (.) (-) dollar Now a dollar? [Aga. Aha. ((Tajik language)) [dvadcat’ tri četyre ↑dollar-a. twenty three four dollar-GEN twenty-three, four- dollars. [Ugu. Aga. Uhhuh. Aha. Očen’ dešev-o. Very low-NEU Very low. Da. Yes. .h Čut’-čut’ (-) bogat-yj >bogat-yj uč-it-sja.< .h little-little (-) rich-MASC rich-MASC study-3SG-PRT Very little. Rich people study. Ugu da: èt-o žalk-o. Uhhuh yes 3SG-NEU pity-NEU Uhhuh, yes, it is a pity. Da. Yes. .h (-) vse-m by-l-o voz↑možn-o. .h (-) all-DAT be-PST-NEU possible-NEU Everyone had the opportunity. Da tam (.) .h ↑bedn-yj uč-it-sja ot↓ličn-o. Yes there (.) .h poor-MASC study-3SG-PRT excellent-ADV Yes,there poor study excellent. 103 5 Collaborative productions 15 N: 16 E: → 17 N: → 18 E: 19 N: Da. Yes. A den’g-i ne to o .h t-- bo↑gatstv-e (-) den’g-i But money-PL NEG that PRE .h t-- wealth-PREP (-) money-PL ↑est’ ºa [h-, be.3SG but wBut money is not the issue. From wealth, there is money but w-, [Horoš-o ne uč-at-sja. Well-ADV NEG study-3PL-PRT They do not study well. ((Tajik)) ne horoš-o uč-at-sja oni. NEG well-ADV study-3PL-PRT 3PL Not well, they do study. Da ºda.º Yes yes. As we have observed in the previous subsections (especially in subsection 5.2.1 concerning the coordinated words and phrases), the recognisability of structures can also enable collaborative productions. The same observation is valid for co-ordinate complex sentences. The following example shares some characteristics with the list constructions (for example, see Jefferson 1990, Koivisto 2011: 96–109, Lerner 1994). According to Jefferson (1990: 81–82), a list can be constructed by more than one speaker and a linking term (‘and’, ‘or’) can be used to achieve a collaborative construction. The participants of the following excerpt are discussing languages that are used for educational purposes in Tajikistan. The official language is Tajik, but since the Russian language was in such a significant position in Tajikistan during the Soviet era, some politicians are advocating raising its status again. To accomplish this, there are Russian-language days at schools, which means that the language spoken in classes during these days is Russian. In line 1, N begins to enumerate possible subjects: u škol’nikov vo vtornik est’ i biologija i geografija I, ‘the students have on Tuesday biology and geography and’. Speaker N uses the conjunction I, ‘and’, in order to form a list, therefore creating recognisability that can be detected by the coparticipant. In line 3, the co-participant, E, produces the recapitulation vse predmety razgovarivaet tol’ko po-russki, ‘all subjects are spoken only in Russian’. (39) TI 28. <16:16> 01 N: .h Aga: tak ja duma-l-a. .h Aha that 1SG think-PST-FEM Aha, that is what I thought. 104 5 Collaborative productions 02 → 03 E: → 04 N: 05 Značit >u u< škol’nik-ov vo vtornik est’ i So PRE PRE pupil-GEN.PL PRE Tuesday be.3SG both biologija i: geografija i, biology and geography and So the pupils have on Tuesday both biology and geography and, Vs-e predmet-y razgovariva-et ↑tol’ko russki. All-PL subject-PL discuss-3SG only Russian All the subjects are taught only in Russian. Aga a vy (.) èt-o v ↑dejstvitel’nost-i ra↑bota-et Aha but 2PL 3SG-NEU PRE reality-PREP work-3SG sejčas, (.) now Aha, but does it work in reality now, Vy skaza-l-i čto ne ↑vs-e učitel-i govor-jat 2PL say-PST-PL that NEG all-PL teacher-PL speak-3PL po-rus-sk-i horoš-o. PRE-Russian-ADJ-ADV good-ADV. You said that not all teachers can speak Russian well. In the previous examples, the conjunction occurred in the preliminary component. When an utterance ends in a conjunction, the recipient can address the situation in different ways. On the one hand, as reported in previous research, when the preceding turn contains a conjunction at the end, the recipient does not usually orient to it as being incomplete (Koivisto 2011: 58, Walker 2012: 149). That is, the recipient treats the previous turn as informationally complete (Walker 2012: 149). On the other hand, a conjunction at the end of the preliminary component can project continuation (for example, Koivisto 2011: 96–109 for list constructions). Example 40 features speakers G and J discussing their experiences in a small Russian town. G has moved to Russia in order to get a degree. He is originally from the Ivory Coast and has previously mentioned on the tape that because he looks different from an average Russian, he has faced many problems in Russia. In this excerpt, G is talking about his Russian friends who are friendly towards him at the university but who do not invite him to their homes. In lines 2–3, he expresses that he sometimes gets the feeling that they would like to invite him, but they are afraid of the reactions of their parents and friends. He creates a complex sentence with the contrastive conjunction no, ‘but’. He then continues that he has never been invited anywhere by a Russian although the idea might have crossed their minds. A contrast has been developed between what is wanted and what might occur. In line 6, G initiates the utterance ty znaeš’ čhto on hočet no, ‘you know that 105 5 Collaborative productions he wants to’, and J completes the complex sentence with on ne možet, ‘he cannot’. The same contrastive structure is identified by speaker J. (40) IS 61. <09:48> 01 G: 02 03 04 05 06 → 07 J: → 08 G: Počemu prinës-ø-ø tak-oj ↓čelovek. Why bring-PST-MASC such-MASC human being Why did you bring that kind of man. Inogda ty ↑tože duma-eš’ čto oni ↑hot-jat, Sometimes 2SG also think-2SG that 3PL want-3PL Sometimes you can also think that they want to, No oni očen’ bo-jat-sja čto bud-ut govori-t’ But 3PL very be.scared-3PL-PRT what will.FUT-3PL say-INF >roditel-i ili ljud-i< vokrug znakom-ye vot tak-ie. parent-PL or people-PL around acquaintance-PL PRT such-PL But they are very scared what will say parents or people around or acquaintances well that kind of. Nikogda net mne rus-sk-ij čelovek >skaza-l-ø< Never NEG 1SG.DAT Russian-ADJ-MASC man say-PST-MASC podj-ëm kuda-to ili. go-IMP.1PL somewhere or. A Russian man has never said to me “let’s go somewhere”. On mož-et byt’ i ↑duma-et, 3SG perhaps and think-3SG He perhaps thinks about it, Ty zna-eš’ čto on hoč-et no, 2SG know-2SG that 3SG want-3SG but You know that he wants to but, On ne mož-et. 3SG NEG can-3SG He cannot. Da: on ne mož-et dela-t’. Yes 3SG NEG can-3SG do-INF Yes he cannot do it. In this particular case, the interpretation of the conjunction no, ‘but’, is purely contrastive. In line 6, speaker G talks about “wanting,” and wanting is usually something that is done out of pure will. This type of use can be also called the concessive, meaning that it is not expected (see, for example, VISK § 1139, AG-80 § 3043–3057 for concessive relations). On the other hand, speaker G has prefaced his upcoming utterance at the beginning of the excerpt and has projected the form of the final component. He has explained that although he has Russian friends, they do not invite him to their home. In line 6, he initiates his turn: ty znaeš’čto on hočet no, ʻyou know that he wants to but’, which is then completed by speaker J: on ne možet, ‘he cannot’. Regarding this particular case, the conjunction no, ‘but’, has not lost its contrastive power. In English, there seems to be a tendency towards 106 5 Collaborative productions the turn-initial conjunction but becoming a turn-initial discourse particle (Mulder & Thompson 2008). Mulder and Thompson call this the “grammaticization of the ‘final but’” (ibid. 194). 5.4 Summary At the beginning of chapter 5, I referred to the previous research on lingua franca interaction and to the fact that candidate completions were absent in scholars results, such as in the data presented by Firth (1996: 246). Although the data analysed in the present study differs significantly from Firth’s data on telephone calls, it is interesting that there are many collaborative productions on many different levels. There are different types of linguistic structures which enable collaborative productions. Participants can co-construct list constructions with the help of coordinating conjunctions. If the conjunction is not used in a list construction, it can still enable the construction of collaborative production even though it is not common (Koivisto 2011: 58, Walker 2012: 149). The use of intensifiers in the preliminary part creates a projection that can be delivered in the final part with adjectives. In government cases, there is also a possibility that participants can negotiate the correct form of the noun phrase. In both the Finnish and Russian data, there were cases that consisted of a noun phrase functioning as a subject and a predicate verb. Structurally, these examples were not exceedingly complex. Two of them consisted of a personal pronoun and a predicate verb. One consisted of a proper noun and a predicate in negative form. These findings differ from the results reported by Helasvuo (2004). In her research, Helasvuo reported no instances that consisted of a subject and a predicate. Furthermore Helasvuo concluded (ibid. 1332), that it is more likely for the subject and verb to be grouped together than the verb and object. With respect to the findings of the present study, this claim has been reopened for discussion. The recognisability of structures and actions can enable collaborative productions. Moreover, semantic features can also trigger collaboration. In the Finnish examples, the participants were able to co-construct structures such as the verb mennä, ‘to go’, and the directional complement. Recognisability also means that participants can recycle the same construction and produce similar collaborative productions. In addition, the order of the elements can be reversed because the recognisability has already been established. 107 5 Collaborative productions As was evident in the examples, different types of collaborative productions can all occur in a short extract. While this means that the extract must be analysed thoroughly and meticulously, it also shows that the conversational participants are continuously following the conversation. 108 6 Word searches 6.1 Previous research on word searches This chapter is devoted to another type of constructions that participants coconstruct in conversation. These constructions differ somewhat from the collaborative productions because in these cases, the first speaker encounters some difficulties in formulating his or her speech. As a response, the recipient has an opportunity to help his or her co-participant. In this chapter, I concentrate on those conversation excerpts that contain participant problems in finding a correct or appropriate word, or when the speaker does not know how to inflect words. These word searches sometimes share many features of co-constructions. To understand word searches therefore requires a thorough analysis of turn-organisational, sequential and syntactic structures of the conversational excerpts. While the term ‘word search’ may be too obscure to capture the whole phenomenon for describing sequences, it tells us something very essential about these sequences. The interactants are searching for something and sometimes the item being sought is more than a word, as it can also be a longer syntactic unit. (Kurhila 2006: 91.) Goodwin & Goodwin (1986: 56) emphasize that a successful word search is not merely a distinct task. Indeed, achieving an outcome “is relevant to the talk in progress, since it is precisely some type of outcome that will permit the interrupted talk to move to word completion once again” (ibid.). However, Goodwin & Goodwin (ibid.) criticise the use of the term ‘word search’, but they concede that this term might be “a useful gloss for the phenomena being investigated, it clearly does not describe what is being sought with adequate accuracy.” I will adopt the term word search as a general conception and will define the search sequences more adequately when analysing them. As I established earlier, my aim is to focus on those word-search sequences that involve both the self and other, which can also be described by the terms ‘current speaker’ and ‘recipient’. Word searches are common in all everyday conversations, and thus these are not characteristic only of NNS interaction. Researchers have also analysed the word-search strategies of aphasic speakers (Helasvuo et al. 2004, Laakso & Klippi 1999). Only during the last decade or so, aphasic word search has been studied from more interactionally-oriented point of view (for example, see Helasvuo et al. 2004: 2). 109 6 Word searches If we describe word searches from the perspective of repair organization, word search can be defined as representing a self-initiation of repair (Kurhila 2006: 91). More specifically, word search is a specific type of selfinitiation of repair. Self-initiations within the trouble-source turn are often marked with non-lexical perturbations, such as cut-offs or sound stretches (Kurhila 2006: 91, Schegloff et al. 1977). That was likewise the case in example 1. The self-initiated word search was marked by non-lexical perturbations and this led to other-repair. As Kurhila (2006: 91) suggests, “hesitancy-marked turns were found to be one environment preceding otherrepair.” The difference between cut-offs and pauses can also be made based on the position in the turn. Generally speaking, a cut-off is usually postpositioned because it initiates a repair on an element that has already been produced in the turn. A pause is usually prepositioned because it is a substitute for a next-due element, that is, it initiates a repair on an element that comes after it. Syntactically, a cut-off is disjunctive because it interrupts the syntactic projection that was formed in the sentence-so-far. Pausing may delay the syntactic projection, but nonetheless it carries it forward. (Lerner 1996: 242– 243, Schegloff 1979: 273.) This chapter will examine the way actual word searches are initiated, solved, and evaluated (table 6). First, I will introduce actual search instances and demonstrate how the problematic cases arise in the conversation. Second, I will analyse how the search is solved. Third, the initiator of the word search can evaluate the produced word and its appropriateness. This type of general formula of word searches assures a consistent analytical procedure, but it also enables u7 to consider those examples that are constructed differently. 110 6 Word searches Table 6. Word searches. Word searches Problem Finnish (37) Pauses, hesitations Russian (14) 8 9 29 5 24 12 11 2 27 6 3 5 4 3 2 2 Verbalisation, use of other languages Resolution Direct offer, one word Prolonged search sequences Reaction Repetition, acceptance Direct continuation Co-participant continues Unsolved cases As the figures above show, word-search sequences consist of three steps: initiation, resolution and reaction. It is also possible that the sequence remains unsolved. 6.2 How speakers indicate that there is a problem? One of the means that the speakers have to indicate that they are encountering a problem in turn formation is to pause their turn. Other ways are, for example, separate search sounds (‘mm’, ‘uh’, and ‘er’). At a later stage, speakers may also make the search more explicit with questions like ‘What is it’ (Schegloff et al. 1977: 363, 367). 111 6 Word searches When analysing word searches, I will follow Kurhila’s (2006: 91) suggestion to classify word searches as belonging to repair organisation. The organisation of repair can be understood as a hyperonym for the different procedures through which interactants address their problems in producing, hearing or understanding talk (Schegloff et al. 1977: 361, Sorjonen 1997: 111). Repair sequences consist of two parts, of a trouble-source turn (the repairable) and the repair turn that may be preceded by a repair initiation. The repairable may have some specific features (for example, errors or mistakes in speech production) but that is not necessary. (Schegloff et al. 1977: 363.) To appeal for assistance, interactants can also use other non-lexical actions. One of the main actions is gaze (see Goodwin 1987: 117). For instance, recipients characteristically gaze toward the speaker during word searches. Goodwin & Goodwin (1986) described a range of nonverbal practices used by both current speakers and recipients to co-construct the activity of searching for a word. Speakers typically withdraw their gaze from the recipient when the search begins, but may also show the facial expression of a “thinking face.” Recipients interpret this withdrawal as a sign that even though the participants is not speaking, the speaker has not relinquished the floor, is engaged in searching for a word, and does not to be interrupted. However, if the current speaker gazes at the recipient during a search in progress, this is interpreted as an invitation to participate in the search. Gaze shifts are therefore a systematic component of word searches in face-to-face interaction. Furthermore, the speakers in Kurhila’s data (2006: 96) direct their search to the recipient by shifting their gaze to him/her when appealing for assistance. Other nonverbal actions, such as hand gestures, often occur in connection with word searches. While searching for words, speakers often wave or whirl their hands (Goodwin & Goodwin 1986) or, they could also raise their palm or finger to indicate their turn holding (Laakso 1997: 170-171). As I was not the sole data collector for the data collection presented here, some interviews were recorded, but not videotape. This means that I was not able to analyse gaze or gestures consistently throughout the material. However, I have transcribed gazes, gestures, and other nonverbal elements as precisely as possible. In some of the examples, the current speaker attempts to involve the interlocutor in determining the correct item by using gaze or other gestures. I will discuss the role of these features more thoroughly with the relevant examples. 112 6 Word searches 6.2.1 Pauses and hesitations that indicate a word search When facing problems in turn-formation, the current speaker has multiple means of expressing it. This chapter will present an analysis of the examples including the pauses, speech perturbations and other hesitation markers that indicate a word search. In those cases, the turn is interrupted and the point of interruption is marked by a variety of ways that signal the initiation of selfrepair. The point of interruption is placed in the mid-course of a turnconstructional unit. (Goodwin & Goodwin 1986: 55, 60.) In the first example, the participants are discussing immigrants in Finland. S begins to tell about an ambassador in Mexico. In line 4, she is displaying difficulties in finding a correct word. She pauses before producing the question mikä se on suomeksi, ‘what is it in Finnish’. In line 4, she defines her search and describes a certain group of immigrants. She defines the group with the help of nationalities and mentions the Somali and Chilean immigrants. In line 5, another participant fills in the missing word and pronounces the English equivalent of refugees. It is interesting that speaker S attempts to find the appropriate word by pronouncing those refugee groups that are essential in the Finnish context. (41) WN 1. <35:42> 01 S: 02 S: 03 S: Minä voi-n selittä ↑suome-ksi vähän tä-stä 1SG can-1SG explain.INF Finnish-TRANS little this-ABL naise-sta joka on suurlähe-- suurlähettiläs Meksiko-ssa. woman-ABL who be.3SG ambassa-- ambassador Mexico-INE I can explain in Finnish of this woman who is an ambassador in Mexico. Hän ↑kysy-y mi-tä se on (.) tärkeä paperi 3SG.FEM ask-3SG what-PTV 3SG be.3SG (.) important paper <maahanmuuttaj-i-en> e massmedia-n kanssa. immigrant-PL-GEN VOC mass media-GEN with. She asks what is is an important paper of immigrants with mass media. Ja tämä nainen sa↓no-o että (2) Suome-ssa anta-a (.) And this woman say-3SG that (2) Finland-INE give-3SG (.) >maahanmuuttaj-i-en< noin kaksikymmentä viisi vuot-ta sitten immigrant-PL-GEn about twenty five year-PTV then kun se alka-- tämä (.) mikä se on when 3SG begin-- this (.) what 3SG be.3SG suom-- e- suome-ksi, Finn-- VOC Finnish-TRANS And this woman says that in Finland they give the immigrants about twenty-five years then when it begin this what is it in Finnish, 113 6 Word searches 04 05 ?: → 06 G: → 07 S: E-n ole varma tämä sano-j-a kun esimerki-ksi tule-e NEG-1SG be sure this word-PL-PTV when example-TRANS come3SG ↑somalialais-i-a ↑chileläis-i-ä koska se on, Somali-PL-PTV Chilean-PL-PTV because 3SG be.3SG I am not sure about the word when for example come Somali or Chilean because it is, Refugee, Refugee, Yeah ↑pako[lais-i-a. Yeah refugee-PL-PTV Yeah refugees. [Pakolais-i-a e-n ole kaikki <sano-a Refugee-PL-PTV NEG-1SG be all say-INF sano-j-a> suome-ksi. word-PL-PTV Finnish-TRANS Refugees I do not have all say words in Finnish. In line 6, G translates the English word into Finnish and produces it in the partitive case. S approves the suggested word and admits that she does not know all the words in Finnish. From the perspective of the whole search sequence, it begins with a pause that is followed by a verbal initiation and when listing different groups, the current speaker is trying to find the correct word. For this reason, the sequence is rather complicated. However, the actual word offer involves only one word. First, the word is in English and is then translated into Finnish. In this second example, M displays difficulties in finding the right word (line 9). She hesitates in formulating her turn. T helps her out (line 10) and produces the correct form. M has pronounced and declined the correct word in line 7, but due to a loud voice and overlapping with speaker S, she displays uncertainty about the noun choice she has made. In line 9, she begins with the cut-off word vaih-, then utters another cut-off word, vaihtoehto ʔ (although, this could be interpreted as the nominative case of the word, it is pronounced with a glottal stop at the end. Glottal stops or cutoff words are sounds that signal problems in clause formation. [Schegloff et al. 1977]). Finally, she pronounces the word vaihtoa, ‘change’, which has the same beginning as the compound noun vaihtoeht, ‘change’. (42) MV 38. <01:34:41> 01 M: 02 S: Kyllä se on se on kiinnostava asia. Yes 3SG be.3SG 3SG be.3SG interesting thing Yes it is a very interesting thing. Se on roska-a, 3SG be.3SG rubbish-PTV It is rubbish, 114 6 Word searches 03 M: 04 S: 05 T: 06 S: 07 M: 08 S: 09 M: → 10 T: → 11 M: Mutta ihmiset katso-vat si-tä. But people watch-3PL 3SG-PTV But people watch it. Ihmiset ↑halua-vat si-tä. People want-3PL 3SG-PTV People want it. Miksi, Why, [LEIPÄ-Ä JA] sirkushuvi-a. Bread-PTV and circus-PTV Bread and circuses. ((from Latin: panem et circenses)) [Kaksi vaihto]ehto-a, Two alternative-PTV Two alternatives, NiiYeaEi ole vaih-- vaihtoehtoʔ vaihto-a, NEG bg.3SG alt-- alternative change-PTV There is no alt- alternative- change, (1) Vaihtoehto-a. ((tavuttaen)) Alternative-PTV (There is no) alternative. ((syllabifying)) Vaihtoehto-a. Alternative-PTV Alternative. M accepts T’s offer by repeating it. Word searches can be described as those situations in which the speakers display hesitancy in their utteranceformulation and they indicate their difficulty in finding the relevant linguistic items to convey the meaning so that it would be understood by other speakers. Thus, the progressivity of the turn (and, hence, the conversation) is halted due to speaker’s problems in formulating (understandable) talk. These conditions of mutual understanding are not met before the participants have found the relevant linguistic means to produce intelligible talk. (Kurhila 2006: 91.) In the following example, speaker V attempts to formulate an utterance that contains a negation in the imperfect tense. This negation consists of a negative verb ei and a participle. However, the speaker is not sure about the participle form of the verb ymmärtää, ‘understand’. First, he produces the negative verb at the beginning of his turn. He subsequently hesitates and interrupts his turn. After the hesitation and interruption, he cut-offs the word ymmär- and then produces a possible form, ymmärnyt, which contains the beginning of the verb and the participle suffix -nyt. Despite the form being incomplete, the co-participant, speaker L, can now formulate the correct 115 6 Word searches form of the verb and then correct V’s utterance. V repeats the word and then formulates another utterance. That utterance does not have a hesitation marker, question or reformulation, but L corrects the form Suomessa, ‘in Finland’, into the grammatically appropriate form, Suomeen, ‘to Finland’. Nonetheless, V does not accept this correction and repeats his own version, Suomessa. (43) FC 79. <01:11:59> 01 V: → 02 L: → 03 V: 04 L: 05 V: E-n: ö: (3) ymmär-- ymmär-nyt. NEG-1SG VOC (3) under-- under-PCP I did not under- undernd. Ymmärtä-nyt. Understand-PCP. Understood. Ymmärtä-nyt ö: (2) milloin tule-e ↓Suome-ssa. Understand-PCP VOC (2) when come-3SG Finland-INE Understood when comes in Finland. Suome-en. Finland-ILL To Finland. Suome-ssa. Finland-INE In Finland. Thus, there are two corrections in this example. The first is initiated by the current speaker in line 1. The recipient produces the correct form in her utterance and this is accepted by the search initiator. The second correction is initiated by the recipient and it is not accepted by the current speaker. This type of rejection can be analysed in terms of the repair organisation and the distinctions that are made by it. As mentioned before, Schegloff, Jefferson and Sacks (1977: 363) made a distinction between correction and repair. Correction is something one does in order to replace an erroneous form by one that is correct. In this sense, correction represents a sub-group of repair, which is a more general term. Concerning typical word-search sequences, there are not erroneous forms, so producing the right form is not a correction. If this particular example is taken into account, it can be argued that the first example was a self-initiation of a correction that led to an othercorrection. Since the current speaker had initiated the search of a grammatically appropriate form and the correct form was introduced, he accepted it and repeated it. The second case was not self-initiated, but it was other-initiated and other-corrected. Since the current speaker did not 116 6 Word searches experience difficulties in producing the grammatically inappropriate form, he did not think it was necessary to change the original form. When examining the grammatical corrections that occur in the conversation between natives and non-natives, Kurhila (2006:21, 31) discovered that only the native speakers correct, but they do not make them consistently. In most of the cases, the native speaker allows the error to pass because “the current speaker is always an authorative source of knowledge with respect to her own thoughts and experiences” (Kurhila 2006: 37). This difference between the participants’ handling of grammatical issues can be explained by how the different participants orient to the conversation. The non-native speaker prefers to resolve the grammatical issues before moving on. For the native speakers, the most important part is the interactional aspect. They want to accomplish the exchange of turns so that it leads to mutual understanding. (Ibid. 132–133.) However, if the conversational topic of the conversation is more familiar to a non-native speaker, or if it is a part of his or her expertise, the non-native speaker becomes a specialist (Orletti 2000: 283–284). A non-native speaker can also perceive himself or herself as being competent, such as in work-related matters, but incompetent as a language user (Vickers 2010: 125). In non-native conversations, the situation is somewhat different. No one can be said to have primary access to the knowledge of the spoken lingua franca. However, this does not imply that there would not be grammatical corrections. The correct grammatical form can be accomplished through joint activity. The participants usually also know who is the most competent in the spoken lingua franca and request assistance. When there are speakers with a more rudimentary knowledge of the spoken lingua franca, they usually ignore the errors and move on. On the other hand, when observing fluency and how non-native speakers are perceived, the lack of an audience or an actual interlocutor may cause perturbations and other signs that can be interpreted as low-fluency (Ejzenberg 2000: 304). In some of the cases, government phenomenon can also trigger the construction of grammatically appropriate words or the participants can negotiate the correctness of the utterances. This is understandable because the construction is otherwise executed but it needs an object to be complete. In other words, a speaker can begin a construction without actually knowing how it can be executed. The recipient is therefore expected to produce the completion. (Szczepek 2000b: 25.) 117 6 Word searches In the example that follows, M is telling his interlocutors about football, which is his hobby. In line 1, he utters the word ihmise-t, ‘people-PL’, then produces the form pela: with the lengthened final vowel, and subsequently pauses and produces the right form pelaavat, ‘play-3PL’. Prosodically, the above-mentioned elicited information can be accomplished specifically by “lengthening on one or more of the vowels in the incomplete intonation unit” (ibid. 25). A two-second pause the occurs before R adds the word jalkapallo-a, ‘football-PRT’, in line 3. After the offered final component, M takes the last word of the compound noun and declines it into a different case, namely in the genitive: pallo-n, ‘ball-GEN’. After that, speaker B wants to express his opinion about the correct declension and begins to elaborate his own version. First, he produces the form pallo, ‘ball-NOM’, and then produces the postpositional phrase pallon kanssa, ‘with a ball’ (literally ‘ball-GEN with’). Although R offered the grammatically appropriate form in line 3, M accepts the form pallon kanssa in line 6 by repeating it. (44) FC 111. <08:25> 01 M: 02 03 R: 04 M: 05 B: 06 M: Nyt ihmiset pela-a (.) pelaa-vat Now people-PL play-3SG (.) play-3PL Now people play (2) Jalkapallo-a. Football-PRT. Football. Pallo-n. Ball-GEN. Football. Pallo: p(h)allo-n kanssa. Ball ball-GEN with A ball with a ball. <Pallo-n kanssa.> Ball-GEN with With a ball. It is important to note that only the participants in the Finnish data- the current speaker and the recipient- are both involved in grammatical searches. When considering the grammatical structures that occur in both languages, it is evident that both languages have the same type of overall structure. Traditionally thinking, both are Subject + Verb + Object (SVO) languages and they both possess a moderate number of morphological categories in declining nouns or in verbal morphology. However, the 118 6 Word searches presence or absence of grammatical searches exists. Although my goal is not to offer external explanations for the instances that are presented in actual conversations, the difference is significantly significant to be taken into account. From a cross-linguistic perspective, the difference is also interesting (cf. Couper-Kuhlen & Selting 2001 on interactional linguistics). The difference can perhaps be explained in terms of the different environments in which languages had been learnt. As for the Finnish language, the lingua franca speakers are usually adults who have learnt Finnish in different language courses, which could imply that they orient to grammatical issues. These speakers have usually also moved to Finland which means that Finnish had not been a part of their life previously. The lingua franca speakers of the Russian language have had different histories. In a certain sense, the Russian material is versatile because it consists of both those who have learnt Russian in their adult life and those who have always had Russian in their lives. In other words, the differences between these two languages are affected by the historical circumstances. Furthermore, while the lingua franca speakers of Russian may have had a more native-speaker relation to the language, they do not value grammatical appropriateness as being as important as the lingua franca speakers of the Finnish language. The concept of fluency can be used to explain the difference between these two groups of language users. On the one hand, fluency is something that seems to be easily described. If somebody is referred as a fluent speaker of a foreign language, he or she is seen as using language as a near-native. (Hüttner 2009: 274). Nevertheless, the methods used to measure or define fluency differ from each other. Furthermore, the concept of fluency is normally understood as an individual phenomenon. According to the classification proposed by Fillmore (1979 [2000]: 51), the most important aspect of fluency is the ability to fill time with talk. However, this ability means that the discourse of fluent speakers is coherent and reasoned and it is not filled with semantically empty material. The third and fourth type of fluencies concern the creative language use in a wide range of contexts. This means that a fluent speaker is able to communicate appropriately and express his or her ideas in many ways. Fillmore (1979 [2000]) emphasised that fluent speakers speak in “semantically dense” (emphasis original) sentences. Thus, speakers who are highly fluent can also use redundancy and repetition to avoid potential breakdowns (Ejzenberg 2000: 299). On the other hand, the non-fluent 119 6 Word searches speakers in that study also often confronted problems in speech formation, but when they formulated a usable phrase, they produced it many times (ibid. 305–306). In short, fluent speakers incorporated more formulaic chunks in their repertoire and they used them as fillers when they were planning how to move forward (see also Wood 2006: 29, 30). This orientation to fluency involves analysing fluency not just as an individual phenomenon, but as being more dependent on interaction. However, when analysing the concept of fluency from an interactive perspective, the categories become more vague. The raters of Hüttner’s study (2009) emphasized that all the speakers’ involvement, and therefore it was considered to be ‘one conversation’. The raters also observed that the speakers were “having a good time” (ibid. 288–289) and in their create a sense of unity, the participants also used laughter. 6.2.2 The use of the spoken lingua franca One possibility is that the current speaker verbalizes his or her search. In this subchapter, these verbalized cases include the use of the spoken lingua franca. There are many different ways of verbally initiating a word search. For example, Helasvuo, Laakso and Sorjonen (2004: 7) reported four basic construction types that occur in their aphasic data. These include a missing element that can be a part of a noun phrase (semmonen/tämmönen/toi + X, ‘a kind of/this kind of/that’ + X) or it can occur in a characterising construction (se on + X, ‘it is + X’), in a locative construction (siellä on + X ,‘there is + X’), or it can be a part of a question (mikä se on? – X ,‘what is it? – X’). In NP constructions, a pronominal premodifier is produced. In Finnish, these modifiers such as a tai/tuo, ‘that’, se, ‘that’, and semmonen, ‘this kind of’. In Russian, čto èto takoe ‘what is this’ (Daragan 2003). These pronominal modifiers are often combined with other search elements such as pauses or search sounds (the Finnish ää, öö and creak; the Russian èè, ‘nonphonological length’) and these indicate a word search that should be completed with the head noun (Helasvuo et al. 2004:9, Podlesskaja & Kibrik 2009: 178–179). What then follows from these pronominal premodifiers and perturbations is that the syntax of the word-search sequences is often quite fragmented (Helasvuo et al. 2004: 1). The speaker begins with one syntactic construction, but when she or he is unable to find an appropriate word, the construction remains incomplete and a new one is begun. The pronominal modifier usually specifies that the missing element is a noun (Helasvuo et al. 2004: 9). The next example features participants who 120 6 Word searches are attending an editorial meeting and discussing possible ways of conducting an interview with people who are participating in a public event. The event is called Maailma kylässä, ‘World Village Festival’, which is held annually in Helsinki. This festival is renowned for its international atmosphere and the high number of immigrants who participate in it. One possibility that is suggested is that they would ask different people what they eat for breakfast. M shares her idea of setting up a candid-camera situation. She initiates her search in line 3 by producing a question with an interrogative pronoun mikä se on, ‘what is it’ (Kurhila 2006: 100). In line 4, T responds to her question with the word kopi, ‘booth’. M the repeats the proposed word with a rising intonation in line 5. (45) MV 51. <26:50> 01 M: 02 03 M: → 04 T: → 05 M: 06 A: → 07 T: 08 M: 09 10 11 T: 12 M: On-ko se hauska (.) jos me tee-mme (.) >jotenkin pila.< Be.3SG-Q 3SG funny (.) if 1PL do-1PL (.) somehow prank Is it funny if we make somehow a prank. (1) Kame:ra keske-llä .hh se on me rakenta-mme jotakin Camera middle-ADE .hh 3SG be.3SG 1PL build-1PL something mis-- se on (.) >mikä se on.> wher-- 3SG be.3SG what it be.3SG. A camera in the middle it is we build something wher- is it what is it. Koppi. Booth. A booth. ↑Koppi. Booth. A booth. Laatikko. Box. A box. Tai laatikko. Or box. Or a box. Keske-llä. Middle-ADE. In the middle. Laita-mme siihen °kamera-n ja suoraan,° Put-PL3 there camera-GEN and straight We put there a camera and straight, (3) Ekraani-lla jos ↑löyty-y, Screen-ADE if find-3SG On a screen if available, Ekraani-lla tai jotain, Screen-ADE or something On a screen or something, 121 6 Word searches For Speaker M, who initiated the word search, the proposed word is sufficient. Speakers A and M elaborate on the search and suggest another word, laatikko, ‘box’, but M continues her speech by describing the surroundings for the whole candid camera idea. In example 46, speaker A is telling about the procedures they had to follow with his wife when she was pregnant. In lines 5–6, he recounts that he asked for permission to film an ultra-sound procedure, but that he was not granted it. At this particular point, no reference had been made to the person who turned down his request. In line 7, speaker K suggests that it was a doctor, which is first accepted by the original speaker, but then at the end of line 8, he begins to hesitate that perhaps it was some other person. In line 9, T suggests that it was a nurse. The suggestion is accepted by the original speaker. It can be said that the participants are jointly attempting to find the right word. Although Speaker K’s suggestion was incorrect, it belongs to the same semantic field as the appropriate word (on semantic contiguity, for example, see Kurhila 2006: 116 and subsection 6.3.2). (46) MV 56. <25:37> 01 A: 02 M: 03 A: 04 ?: 05 A: 06 A: 07 K: 08 A: → 09 T: → Mä ker-- kerran ot-i-n kame:ra (.), 1SG on-- once take-PST-1SG camera (.) I once took a camera, Joo. Yeah. Se ol-i ↑ultraääni. 3SG be-PST.3SG ultra:sound. It was an ultrasound. Joo. Yeah. Mä hal-- halus-i-n kuva-ta mutta sano-i .hh 1SG wa-- want-PST-1SG shoot-INF but say-3SG .hh ei saa kuva-ta. NEG.3SG allow shoot-INF. I wanted to shoot, but they said that it is not allowed. Kys-- kysy-i-n miksi se ei: Ask-- ask-PST-1SG why 3SG NEG.3SG I asked why it is not Lääkä↑ri:. Doctor. Se: ol-i ((caughs)) (.) kyllä ol-i lääkäri tai se 3SG be-PST.3SG (.) yes be-PST.3SG doctor or 3SG mi-tä se ol-i, what-PTV 3SG be-PST.3SG It was yes it was a doctor or it what it was, Sairaanhoitaja. Nurse A nurse. 122 6 Word searches 10 A: Sairaanhoitaja. Nurse A nurse. Although the search in the previous example indicated that the missing element was a noun, there are other cases that arise as well. My data contain examples of a speaker searching for an infinitive form of a verb as a part of the structure. The participants of the following excerpt are the same as in the previous search, with M proposing another idea of showing pictures to the interviewees and asking them to describe what they see. (47) MV 42. <01:09:11> 01 M: 02 03 → 04 T: → 05 M: Mu-n m-- minu-n miele-stä se: tarko-- se pääasia 1SG-GEN m-- 1SG-GEN mind-ELA 3SG mea-3SG main:thing ei vain mi-tä ↑hän sano-o. NEG only what-PTV he say-3SG I think in my opinion it mean- it is a main thing not just only what he says. Me <tunne-mme me> .hh tiedä-mme jotakin mi-tä hän 1PL know-1PL 1PL .hh know-1PL something what-PTV 3SG voi sano-a. can.3SG say-INF We know we know something what he may say. Mu-n .hh miele-stä pitä otta-a ↑semmonen kuva 1SG-GEN .hh mind-ELA have.to.3SG take-INF such picture ei tarvi ö: nu tut-- ö: mikä se on, NEG.3SG need VOC PRT res-- VOC what 3SG be.3SG I think we have to take that kind of picture which need not to what is it, Kään[tä-ä. Translate-INF to translate. [Kääntä-ä. Translate-INF to translate. It is also possible to formulate words that somehow resemble the actual words in Finnish or in Russian. However, those cases are rather uncommon. By actual words, I refer to those words that have Finnish or Russian stems and suffixes, but they are not really words that exist (for more on what is called foreignizing, see Færch & Kasper 1983a: 47). With reference to Finnish and Russian, those words can be called Fennicized or Russified words (for example, see Kurhila 2006: 111–16 for Fennicized words). In the following example, N wants to know whether her interlocutor has siblings. She invents a word that bears some resemblance to a Finnish word. In line 1, she produces the word siskoruksia (in the singular siskorus, which could be 123 6 Word searches interpreted as consisting of one noun sisko, ‘sister’, and the suffix –rus, which is familiar from the noun sisarus, ‘sibling’. That is also the word that L produces in line 2 and N subsequently repeats it in line 3. (48) FC 71. <01:35:35> 01 N: → 02 L: → 03 N: ↑Liina on-ko: sinu-lla ö ↑siskoruks-i-a. ((laughter)) Liina be.3SG-Q 2SG-ADE VOC sibiling-PL-PTV Liina do you have sibilings. Sisaruks-i-a. Sibling-PL-PTV Siblings. Sisaruks-i-a ↑joo. Sibling-PL-PTV yeah. Siblings yeah. 04 L: Yks minu-lla on yksi sisko. One 1SG-ADE be.3SG one sister I have one sister. Other cases are also possible. In the example that follows (46), the participants work together to define the meaning of the word sarjakuva, which means a ‘comic strip’. For animated films, there is another word, namely piirroselokuva or animaatioelokuva. In line 2, E begins to define the word and tries to determine whether a comic strip is something that could be watched on television. After a pause, C elaborates on the search by guessing that sarjakuva could resemble a cartoon in English. Following his own suggestion, C continues by stating that a sarjakuva can occur in a newspaper. E’s response to this is ai jaa, which could be interpreted as a sign that E responds to the offer as if it were news to her. (Sorjonen 1999: 175–176, VISK § 799.) After the evaluation, J continues the actual search and produces an interesting word. She takes the word ‘cartoon’ and fennicizes it in line 9. Speaker C pronounces the word in English in line 4, but in this case, J uses the word and adjusts it to the Finnish system. At the end of the word, there is a final vowel –i, which is also used in borrowed words from foreign languages and is adjusted to the Finnish system. (Kurhila 2006:111–117.) (49) CC 23. <01:19/2> 01 02 E: Sarja↓kuv-i-ssa. Cartoon-PL-INE In cartoons. (2) 124 6 Word searches 03 E: → 05 06 C: → 07 08 C: 09 J: 10 E: 11 J: → 12 E: 13 C: 14 J: 15 C: 16 J: 17 C: 18 J: 19 C: On-ko tämä niin kuin teevee-ssä <tai mikä on Be.3SG-Q this so like television-INE or what be.3SG sarjakuva.> cartoon Is this like in TV or what is a cartoon. (1) On-ko sarjakuva kuin niin ö: mt car↑toon. Be.3SG-Q cartoon like so VOC VOC cartoon Is sarjakuva like a cartoon. (1) On-ko: esimerki-ksi ö sanomalehde-ssä, Be.3SG-Q example-TRANS VOC newspaper-INE Is for example in a newspaper, Mhy. Uhhuh. °Ai jaa.° Oh yeah. Kartuuni ↑joo. Kartuuni yeah. Okei. Okay. >On-ko se< on-ko tämä (1) sarjakuva. Be.3SG-Q 3SG be.3SG-Q this (1) cartoon Is it is this a cartoon. Sarja-- sarjakuv-i-ssa ↑joo. Cart-- cartoon-PL-INE y eah Cart- in cartoons yeah. ↑Sarjakuva on ↓lehde-ssä. Cartoon be.3SG newspaper-INE A cartoon is in a newspaper. °Mhy.° Uhhuh. Ei-k-s ↓niin. NEG.3SG-Q-CLI so. Isn’t it. Minä ymmärs-i-n joo. 1SG understand-PST-1SG yeah I understood yeah. Joo. Yeah. In this excerpt, the speakers experience uncertainty regarding the Finnish word, concerning whether a comic strip can be seen on television. At this point, the English equivalent is offered, which is then reformulated into a possible Finnish word. After this elaboration, C shows a newspaper and asks whether it contains a comic strip. In line 13, he states that a comic strip is in the newspaper and asks for a confirmation in line 15 by saying eiks niin, ‘right?’ J confirms this assumption in line 16. Speaker E, who initiated the word search, does not participate on the final evaluation process. 125 6 Word searches The next example (50) in Russian somehow resembles the previous case. Three speakers are participating in this excerpt. Speaker H is talking about his grandfather who used to make oxcarts for the Aghul people in the Caucasus. In line 3, he states that because there no roads, there were also no large motor vehicles. He attempts to create another name for the carts and produces the question in Russian of nu èto kak èto nazyvaetsja, ‘well how do you called it’, in line 4. In line 5, speaker M produces the correct word, telegi, ‘carts’. However, P produces a word that is possibly Russian in line 6, pelegi. The initiator of the word search approves of the word that is offered by M and repeats it. (50) DM 50. <13:27> 01 H: 02 03 04 → 05 M: → 06 P: 07 H: U ↑nas deduska by-l-ø Abd- >e-go zva-l-i< PRE 1PL.GEN grandfather be-PST-MASC Abd- 3SG-GEN call-PSTPL Abdul Aziz. Abdul Aziz We had a grandfather he was called Abdul Aziz. On poni↑ma-ete èto-t dela-l-ø èto-t He understand-2PL 3SG-MASC do-PST-MASC 3SG-MASC arb-y ↑dlja agul’c-ev. oxcart-PL PRE Agul-PL.GEN He you understand did those oxcarts to Agul people. U ↑ni-h dorog-i ne by-l-o da tak-ie bol’š-ie PRE 3PL-GEN road-PL NEG be-PST-NEU PRT such-PL big-PL samohodn-y. motorized:vehichle-PL They did not have roads yes those big motorized vehicles. Nu èt-o >kak kak< èt-o nazyva-et-sja. Well 3SG-NEU how how 3SG-NEU call-3SG-REFL Well how it is called. Teleg-i. Cart-PL. Carts. Pe↓leg-i da èt-i, Peleg-PL yes it-PL Pelegi yes they. Teleg-i dela-l-ø. Cart-PL do-PST-NEU Carts did. 6.2.3 Loan word examples To solve the search sequence, the participants in the previous examples did not change the language being used even though it is possible to use languages other than the spoken lingua franca (for example, see Kurhila 126 6 Word searches 2006: 105). For these types of examples, some elements of Finnish or Russian might occur while the search sequence might be initiated in Finnish or Russian but with the searched word being pronounced in another language. The searched word is therefore present in the conversation but in the ‘wrong’ language, and the current speaker wants to make him- or herself clear and to find the appropriate word. In communication involving foreign languages, there is always the possibility of switching from L2 to either L1 or to another foreign language. Language changes may involve varying stretches of discourse from a single word to complete turns. (Færch & Kasper 1983a: 46.) Using other languages also constitutes a specific type of ‘communication strategies’, which are also known as achievement strategies (ibid. 36–37, 45–53). An achievement strategy refers to learners attempting to solve problems in communication by expanding their communicative resources rather than by reducing their communicative goal. Achievement strategies may also be called compensatory strategies because they aim at solving problems in the planning phase that are caused by insufficient linguistic resources. (Ibid. 46.) Communication strategies may be defined in a variety of different ways, but the main classification is based on how the roles of the interactants of a conversation are seen to be formed. On the one hand, the term can refer to the mutual attempt by two interlocutors to agree on a meaning in situations where requisite meaning structures do not seem to be shared (Tarone 1983: 65). On the other hand, communication strategies can also be defined as potentially conscious plans for solving a problem in reaching a particular communicative goal (Færch & Kasper 1983b: 212). Færch and Kasper (1983a: 50–51) emphasise the role of the participant who has problems in reaching a communicative goal. While they admit that problems in interaction can be solved by mutual efforts, it is the interactant facing the problems who has the opportunity to decide whether to use a specific psycholinguistic strategy. He or she can also signal the problem to the other interlocutor and can attempt to solve the problem in collaboration. Færch & Kasper (1983a: 36–37) divide communication strategies into two major types, reduction strategies and achievement strategies. Reduction strategies, which are governed by avoidance behaviour, include both formal reduction strategies and functional reduction strategies. Formal reduction strategies refer to situations in which a speaker wants to avoid producing non-fluent or incorrect utterances and decides instead to communicate by 127 6 Word searches means of a ‘reduced’ system, focussing on stable rules and items that have become reasonably well automatised (ibid. 38–43). Functional reduction strategies, in contrast, refer to cases in which learners experience problems in either the planning phase (due to insufficient linguistic resources) or in the execution phase (retrieval problems), and their behaviour is based on avoidance rather than achievement. By adopting a functional reduction strategy the learner ‘reduces’ his or her communicative goal in order to avoid the problem. (Ibid. 43–44.) In L2-research, communication strategies are usually seen as a tool that L2-speakers use to solve problems in speaking (Færch & Kasper 1983a: 31– 35, 1983b). The theory focusses on the decisions by individual speakers, leaving the interactional aspect of these strategies more or less unanalysed (cf. Kurhila 2006: 94, Tarone 1983). In the example that follows (51), speaker B encounters problems in line 1. He cannot remember the Finnish equivalent for the word ‘cousin’. He initiates his search by verbalising that he cannot remember the word in Finnish. He produces the correct word in English and L then translates it into Finnish as serkku, ‘cousin’. In line 3, B confirms the appropriateness of the word and repeats it. (51) FC 104. <13:15> 01 B: → 02 R: 03 L → 04 B: 05 B: Ja sam- sunnuntai-na >minä minä< mene-n ↑Herttonieme-en And VOC Sunday-ESS 1SG 1SG go-1SG Herttoniemi-ILL koska häne-llä ↑opettaja mikä se on because 3SG-ADE teacher what 3SG be.3SG minä ei: muista ö cousin. 1SG NEG remember VOC cousin. And on Sunday I will go to Herttoniemi because he has teacher what is it I do not remember cousin. No mutta opettaja nyt on ↑tauko. Well but teacher now be.3SG break Well but teacher now is a break. Serkku. Cousin. A cousin. Serkku hän ensimmäinen serku:-t se on iso juhla: Cousin 3SG first cousin-PL 3SG be.3SG big party juhla: >joulujuhla.< party Christmas:party He is a first cousin. It is a big party party Christmas party. Kaikki: on punaise-en (.) paita: ja pipo:. All be.3SG red-ILL (.) blouse and cap Everyone is in red a blouse and a cap. 128 6 Word searches The mixture of different languages can also be indicative of word searches. In the next example, speaker R attempts to form a Finnish compound noun with the English word medium and the Finnish word aika, ‘time’, meaning the average time it takes for an ambulance to arrive at the scene of an accident. In line 2, speaker L produces the actual Finnish word keskiaika, ‘average time’ 30. (52) FC 62. <31:14> 01 R: → 02 L: → 03 B: 04 L: 05 R: 06 L: No tämä on medi-- tämä on >medium ö: kesken< Well this be.3SG medi-- this be.3SG medium VOC uncompelete ↑ik-- aika. VOC time Well this is medium uncomplete time. Keskiaika. Average:time. Noin. About. Kaksi, Two, Kaksi↓kymmentä minuutti-a. Twenty minute-PTV Twenty minutes. Joo joo joo. Yeah yeah yeah. Another possibility is to use a foreign language that not all the participants are familiar with. The following example features speaker M telling the others about an old Soviet film in which a Soviet spy gives birth in Germany. M and T both speak the same native language, Russian. This example consists of a two-word search sequence. The first search is selfinitiated and self-repaired by the current speaker. In line 3, she produces the initiation kak skazat’ kričat’, ‘how to say kričat’, accelerating her speech. She finds the appropriate word on her own and produces it in a load voice, HUUtaa, ‘scream’. In line 8, she is having difficulties in finding an appropriate Finnish word and as a consequence, she initiates the search in Russian. First, she begins to produce the Russian word špio-, which means spy. Then she decides to use another Russian word razvedčik, ‘scout’. The completion of that search consists of two parts. First, T begins with the word valokuvaa-, ‘photograp-’, but then she suggests another word, vakooja, meaning ‘spy’. The two words begin with same two sounds. S produces yet 30 The Finnish word keskiaika could also be translated into English as ‘the Middle Ages’, but that interpretation is not possible in this context. 129 6 Word searches another word, vakoilija, which is synonymous with vakooja. M then repeats both words and confirms their acceptability. (53) MV 57. <23:56> 01 M: 02 03 M: 04 05 06 07 08 → 09 T: 10 S: 11 M: → 12 13 Jok-- joka tapaukse-ssa synnytys ja nainen ö: Eve-- every case-INE birth and woman VOC In in every case a birth and a woman (1) >kak skazat’ kričat’< >How to say kričat’< HUUta-a (.) oma kiele-llä Scream-INF (.) own language-ADE Screams in her own language Minä muista-n ↑elävä kuva saksa-n ö: aika-na 1SG remember-1SG live film German-GEN VOC time-ESS sotilaa-n aika soldier-GEN time I remember a film in the German times .h he ajattel-i-vat että hän on ↑saksalainen .h 3PL think-PST-3PL that 3SG be.3SG German They thought that she is German. Mutta ö hän on ö: venäläinen But VOC 3SG be.3SG VOC Russian But she is Russian. Mikä se on špio-- ö: ö razvedčik What 3SG be.3SG sp-VOC VOC scout. What is it špio- razvedčik. Valokuvaa-- voi ei kun ö [vakooja] Photograp-- oh NEG but VOC spy A photograp- oh no I mean spy. [vakoilija] [spy vakooja Spy. Vakoilija mutta kun hän synty-i lapsi Spy but when 3SG bore-PST.3SG child Vakoilija but when she bore a child Hän sano-o .hh äiti venäjä-n kiele-llä ma:ma: 3SG say-3SG .hh mother Russian-GEN language-ADE mother s/he ((the child)) says in Russian mu:m. ((laughter)) As I mentioned previously while introducing the material, the Finnish data consist of both Russian native speakers and those who have learnt Russian at school. However, usually only the Russian native speakers use Russian in their word searches and they predominantly use it when they can perform the search with other native speakers (for example, Mondada 2004: 31–34). The following example constitutes one of the rare cases of a native 130 6 Word searches and non-native Russian speaker using Russian when conversing in Finnish. J and L are native Russian speakers, and A’s mother tongue is Lithuanian. (54) CC 31. <17:01/2> 01 J: → 02 L: → 03 J: 04 A: → 05 06 07 A: 08 L: A: Čto tak-oj huhu ili šušu ili čto tak-oj ↑spletnja. What such-MASC rumour or šušu or what such-MASC rumour What is that rumour or šušu or what is rumour. Hu↑hu-j-a. Rumour-PL-PTV. Rumours. >Huhu huhu.< Rumour rumour. Huhu on spletnja on huhu. Huhu be.3SG spletnja be.3SG huhu Huhu is spletnja is rumour. Juratik. (1) Huhu on ↑toinen sana. Huhu be.3SG another word. Huhu is another word. Juoru-j-a ↑ja ↓huhu-j-a. Gossip-PL-PTV and rumours-PL-PTV Gossips and rumours. In this example, J initiates her search in Russian by stating čto takoj, ‘what’s such’, and then continues in Finnish by saying huhu, ‘rumour’. She completes her search initiation by also producing the same question in Russian: čto takoi spletnja, ‘what’s such gossip’. In Russian, she uses the word spletnja, which means ‘gossip’; the correct Russian equivalent for the word huhu would be sluh. Semantically, these words are rather close. In line 4, speaker A translates the word huhu into Russian as spletnja. She then addresses speaker J, Juratik, who initiated the word search, and requests confirmation. After a pause, A begins to hesitate and rethinks her offer in line 7 as huhu on toinen sana, ‘a gossip is another word’. In line 8, another native Russian speaker, L, confirms that the words are almost synonyms. In the next example, speaker S produces the word that she does not recall in Finnish, soveltaa, ‘apply’. The infinitive form of the Finnish verb is connected to the Russian finite verb of načinaet, ‘begins’, without pauses or other hesitations. After producing the missing word in Finnish, speaker S clarifies that she does not know the word in Russian. In line 3, an unknown speaker begins by offering the word podnjat’, ‘raise’. The person initiating that search wants to check the suggested word and produces the word kak, ‘how’, in line 4. Since the suggested word was not correct, speaker K produces instead the beginning of the word prikla-, ‘app-’, in line 5. This is 131 6 Word searches sufficient for the original speaker, who first cuts off the appropriate word and then pronounces the whole word. For her purposes, the incomplete infinite form made the search complete. (55) UM 39. <02:02:58> 01 S: → 02 03 ?: 04 S: 05 K: → 06 S: A potom é ja uči-l-as’ v: (-) universitet-e But then VOC 1SG study-PST-FEM PRE (-) university-PREP psihologi-i, psychology-GEN But then I studied psychology at the university, No >pedagog učitel’< pedagog če-mu by on ni But pedagog teacher pedagog what-DAT PRT 3SG NEG u↑či-l-ø-sja vsë ravn-o načina-et soveltaa study-PST-MASC-PRT all same-ADV begin-3SG soveltaa vot >kak ne zna-ju kak po-rus-sk-i skaza-t’,< well how NEG know-1SG how PRE-Russian-ADJ-ADV say-INF But a pedadog and a teacher no matter whatever he teaches all the same begins to soveltaa well I do know not how to say it in Russian, Podnja-t’. Raise-INF To raise. Kak. How. Prikla-App->Prik--< prikladyva-t’ ↑vs-ë vsë čto iz App-apply-INF all-NEU all-NEU what PRE kako-go-to drug-oj oblast-i sfer-y what-GEN-CLI another-GEN.FEM area-GEN sphere-GEN è nahod-it da prikladyva-t’ v svo-↓ej sfer-↓e VOC find-3SG yes apply-INF PRE own-PREP.FEM sphere-PREP prepodavani-ja. teaching-GEN. Apply all from another sphere and area and find yes apply to his or her own teaching sphere. When stepping outside the spoken lingua franca, the initiator of the wordsearch sequence needs to have prior knowledge of the other speakers’ language skills. Among friends and peers in informal settings, this type of strategy can be effective and suitable, but among co-workers or in institutional settings (in meetings, etc.), it requires careful consideration. On the one hand, the resources of other languages are made available by using them in conversation. Thus, a loan word search can speed up the search sequence. On the other hand, loan words may exclude some of the participants from the conversation. However, in my data, this is not the case. Participants use other languages besides the spoken lingua franca because 132 6 Word searches they know that the other participants are multilingual speakers and therefore the search sequence can be resolved effectively. The traditional argument is that bilingual or multilingual speakers are able to use two or more languages during the same turn. This conversational code switching can be defined as juxtaposing speech passages that belong to two different grammatical systems or subsystems within the same verbal exchange. 31 Most frequently an alternation takes the form of two subsequent sentences, as when a speaker uses a second language either to reiterate his or her message, or to reply to someone else’s statement. (Gumperz 1982: 59.) This code switching can be accomplished through joint activity, and code switching itself does not necessarily imply that the bilingual or multilingual speaker is not sufficiently competent in the grammatical system of the language being used (Gumperz 1982: 64, see also Færch & Kasper 1983a: 46, Jenkins 2011: 928, Thomason 2001: 129-136). Cogo (2009: 268) points out the following: In conclusion, contrary to the assumption that code-switching is evidence of linguistic deficiency in bilingual or multilingual speakers, the data analysis suggests that code-switching is used as an additional resource to achieve particular conversational goals in interactions with other intercultural speakers. Some researchers envision language alternation as being a hyperonym that covers both code switching and transfer (for example, Auer 1988). Transfer refers to those cases in which the language alternation is connected to a particular conversational structure such as a word, or a sentence. Code switching refers to the language alternation that is connected to a particular point in conversation (Auer 1988: 192). Auer argues that (ibid. 192–193) the participants of a bilingual conversation face two types of tasks, and Auer refers to these as ‘problems’. First, if two or more languages are being used in conversation, then the participants need to decide which language to speak. Second, in order to cope with the problems that are caused by the actual conversation, such as turn-taking, participants must to be able to solve them. Toward this end, the use of two or more languages may assist in conquering these problems. In general, this language alternation can be part of the participants’ cultural and social identity. As a consequence, code switching is not used to compensate for some linguistic deficiency; on the contrary, code switching can be a strategy that signals solidarity and membership. In these cases, the 31 In these cases, the speaker is considered to have access to both the systems of the subsystems. If the speaker uses language that is not considered to be a part of his or her repertoire, the term ‘language crossing’ is used for these instances (Rampton 1998: 291). 133 6 Word searches community is multilingual so that the participants ‘can code switch into a third language which is none of the speakers’ mother tongues’. (Cogo 2009: 269.) In the following excerpt, O is recounting that she eats porridge and that she likes it to be salted (lines 2–3), but after that she encounters problems in describing the type of porridge that her husband prefers. M helps her in line 5 and suggests the word makea, ‘sweet’. In line 7, T poses a clarifying question concerning whether the porridge they are talking about is oatmeal porridge, and repeats the question in line 8. At this point, to facilitate the word search, O uses a Russian word, ovsjanka, ‘oatmeal porridge’, to describe the porridge. T translates that word into Finnish and O subsequently repeats the translation (cf. Kurhila 2006: 113). All the participants in this excerpt are Russian. For them, using Russian language is a way to expedite the search and to return to the actual conversation being held in Finnish. (56) MV 48. <45:22> 01 O: 02 03 04 M: 05 O: 06 T: 07 08 O: → 09 T: → 10 O: Minä (.) syö-n puuro-a, 1SG eat-1SG porrige-PTV I eat porridge, Mutta minä: ehh pidä-n ö ↑suolapuuro-sta But 1SG VOC like-1SG VOC salt:porrige-ELA But I h like eh salted porridge. A: mi↑nu-n miehe-ni pidä ehh VOC 1SG-GEN husband-PX like.3SG VOC My husband like eh Ma↓kea. Sweet. Makea joo. Sweet yeah. Mutta ↑kaurapuuro, But oatmeal:porridge. Kaurapuuro tai mikä ↑puuro, Oatmeal:porridge or what porridge, °Ov↓sjanka.° Ovsjanka ((oatmeal porridge)) Ovsjanka ↑kaurapuuro. Ovsjanka oatmeal:porridge. Kaur↓apuuro. Oatmeal:porridge. As we have seen in the previous example, loan words are not regarded as the final resolution of a search; rather, they are considered to be a means of finding the resolution (ibid. 111). These strategies of interlingual transfer 134 6 Word searches result from a combination of different linguistic features from the IL and the L1 or from other languages (Færch & Kasper 1983a: 46). The difference between loan words and foreignised words is determined by their usage. For loan words, speakers utilises the resources from other languages. They know the word in the other language and they present it as a possible resolution to the problem. The ‘foreignised’ word search sequences involve speakers trying to resolve the search combining resources from the languages that are present in the interaction. (cf. Kurhila 2006: 112–113). Both of these strategies are dynamic. Kurhila (ibid. 116) also observes that producing foreignised (for example, Fennicised) words requires a relatively good knowledge of the structure of the language being used. 6.3 Actual resolving of the search 6.3.1 Direct offer, one word At the focus of this study are the type of word-search sequences that are initiated and resolved by two (or more) speakers. In terms of repair organisation, those sequences consist of the self and the other. The sought after word is usually offered and after that offer, the initiator of the search can evaluate its appropriateness. It is also possible that the search sequence is prolonged. In the previous chapter, I examined how word-search sequences are initiated, and made the interesting observation that whether the initiation is done nonverbally (by pausing, etc.) or with words (using the spoken lingua franca or other languages), the resolution is usually just one word. After the sought word is produced, the search initiator can evaluate the word and move on. When the sequential context of the search sequence is considered, it can be argued that the search often occurs near the end of the unit. If this observation is then associated with collaborative productions, this provides a place for the co-participant to complete the unit in progress. (Lerner 1996: 261–263.) However, the turn prior to the completion is often fragmented. The course of the turn is affected by pauses, hesitations and other markers of uncertainty in turn-formation. In the first example, E is searching for a word and accompanies her turn with two pauses and a hesitation. The first pause is a micropause and the second is longer. After the second pause and a hesitation sound, J offers the word pelätä, ‘be scared’. 135 6 Word searches (57) CC 15. <35:35/1> 01 E: 02 J: 03 E: 04 J: 05 C: Sit↓ten (.) hän halua-a (2) ö:, Then (.) 3SG want-3SG (2) VOC Then he wants to, ↑Pelä-tä, Be:afraid:of-INF Be afraid of. Joo, Yeah, Hän halua-a pelä-ä, 3SG want-3SG be:afraid:of-INF He wants to be afraid, °Joo joo.° Yeah yeah. Syntactically, E has formulated her turn so that it indicates to J what the searched word is. The verb haluta, ‘to want’, requires an object in the form of an infinitive phrase (called the A-infinitive; VISK § 469). E accepts the suggested word with the response particle joo, ‘yeah’. If the word that is being searched for is produced by resorting to using another language, usually the offer is only one word. This means that the next speaker translates that word into Finnish. In the following example, the participants are talking about telling jokes and how difficult it is. In lines 1– 2, J begins to describe her own attitude towards jokes and explains that she usually only listens because she cannot tell jokes herself. She subsequently continues that perhaps the reason for not knowing how to tell jokes is that one needs to possess some special quality. In line 4, she produces the utterance minä luulen se tarvitse, ‘I think it needs’, which could be interpreted as the current speaker expressing a personal belief of what one needs to have in order to be a good joke teller (cf. Baumgarten & House 2010: 1191–1192). After expressing her belief, she pauses for two seconds and begins to search for the correct word in English by saying ‘how to say it English I don’t know’. The English utterance is produced in a quiet voice, indicating that it is oriented towards the speaker herself (for instance, see Baumgarten & House 2010: 1196). After another long pause, she finds the possible English word as ‘experience maybe’. In line 8, E translates the English word into the Finnish word kokemus. J’s turn in line 4 has a specific syntactic structure in formation that she alone completes in line 9 when she uses the partitive case of the suggested word and then produces the whole construction once more as se tarvitsee kokemusta, ‘it needs experience’. 136 6 Word searches (58) CC 16. <36:37/1> 01 J: 02 03 C: 04 J: → 05 06 07 08 09 J: E: → J: Mutta ↑itse minä e-n osaa kerto ↓vitse-j-ä. But myself 1SG NEG-1SG know tell.INF joke-PL-PTV But I myself I do not how to tell jokes. Minä on aina vain kuuntele (-). 1SG be.3SG always only listen (-) I always just listen. Joo. Yeah. Kun se on (.) minä luule-n se tarvitse (2) e: When 3SG be.3SG (.) 1SG think-1SG 3SG need.3SG (2) VOC how to say it in ºEnglish I don’t know.º When it is I think it needs how to say it in ºEnglish I don’t know.º (3) Experience maybe. (2) Kokemus. Experience. ↑Kokemus-ta se tarvitse-e kokemus-ta. Experience-PTV 3SG need-3SG experience-PTV Experience it needs experience. The search initiation in these examples was complex either because it consisted of multiple pauses and hesitations, or because it was produced in a language other than the spoken lingua franca. However, the actual offer was merely one word and it was offered without any specific guess markers. As the first example indicates, the form of the search initiation can predict the form of the searched word. If the word search is initiated by either using another language or by guessing, the participants may discontinue the current discussion and create a side sequence, meaning that the word search becomes the focus of the interaction (Jefferson 1987: 88, Kurhila 2006: 39). This type of side sequence can be called exposed correction (Jefferson 1987). A conclusion is needed in order for the conversation to move forward. The pattern [A: X, B: Y, A: Y] is found when the correction is accepted, and [A: X, B: Y, A: X] if it is rejected. (Jefferson 1987: 88.) It is also possible that the correction is embedded, which means that it is not the interactional business that the participants are engaged in (ibid. 95). Search can also be divided into immediate and delayed contributions (Lerner 1996: 262). Lerner (ibid.) continues as follows: Many word search candidates are held off or delayed, giving the current speaker an opportunity to self-repair and when a candidate is finally produced it is regularly designed as a confirmable – ‘try-marked’ – guess. In contrast, recipient contributions initiated at the beginning of a search – i.e. immediately after the onset of the search – are ordinarily not produced as guesses. 137 6 Word searches 6.3.2 Prolonged search sequences Although word searches are usually resolved quickly and with only one word, it is also possible for the search to prolong. In this chapter, I attempt to analyse why this occurs. One might think that the prolonging of a word search is due to insufficient knowledge of the language being used, and perhaps in some cases this is true. When non-native speakers attempt to resolve a search concerning grammatical issues, the possibility exists that the search will be prolonged. Some participants may have better knowledge of inflection or case endings but it is also possible that all participants are in equal position. In the following example, C begins his search for the correct word form by introducing two possible forms that have come to his mind. Later in the sequence, it becomes obvious that he is searching the participle of the verb kuulla, ‘to hear’. This verb resembles another verb kuulua which can be translated into English with several words such as ‘be heard,’ be audible,’ or ‘to belong’. The two verbs, kuulla and kuulua, are inflected differently but the difference is not very drastic. The participle of the verb kuulla is kuullut and the participle of kuulua is kuulunut. After C has produced both alternatives, E and J continue the search sequence by confirming both of them in lines 2 and 3. J checks the form by producing the English equivalent of ‘I have heard’. In line 4, E repeats her own offer, but after hearing the English translation, C accepts J’s suggestion. The repairs or corrections focus on the morphological or morphophonological level of the utterance. As the participants are nonnative speakers, it is possible that although their pronunciations may differ, none of them is willing to correct the speech of the others on the phonetic level. (Cf., e.g. Brouwer 2004: 93.) (59) CC 32. <17:09/2> 01 C: 02 → 03 E: → Joo minä ((coughs)) (.) ole-n kuul-lut (.) ↑kuulu-nut Yeah 1SG (.) be-1SG hear-PCP (.) belong-PCP kuu--, hea-Yeah I have heard belonged, >On-ko se> on-ko se: ole-n kuul-lut vai ↑ole-n Be.3SG-Q 3SG be.3SG-Q 3SG be-1SG hear-PCP or be-1SG kuulu-nut tai, belong-PCP or Is it is it I have heard or I have belonged or, Ole-n kuulu-nut. Be-1SG belong-PCP. I have belonged. 138 6 Word searches 04 J: → 05 E: 06 C: → 07 J: 08 C: Ole-n kuul-lut (.) I have heard. Be-1SG hear-PCP I have heard. I have heard ((same in English)). Ole-n ↓kuulu-nut. Be-1SG belong-PCP. I have belonged. O↑kei (.) .hh ole-n kuul-lut .hh yks(h)i joka on Okei (.) .hh be-1SG hear-PCP .hh one which be.3SG (.) sama (1) ei ole ai- aivan sama mutta (.) sama, (.) same (1) NEG.3SG be ex- exactly same but (.) same Okay I have heard one which is the same not exactly same but the same. °Mhy.° Uhhuh. Aika: sama (.) kuin tämä Pikku-Kalle-vitsi. Pretty same like this Pikku-Kalle-joke. Word-search sequences can be prolonged when more than one coparticipant is involved in the process. The involvement of two or more coparticipants can result in long sequences and the following example illustrates this, as it is the longest word search sequence—one minute--- that occurs in the data. The participants of this excerpt are discussing their breakfasts. Speaker A is describing that he eats bread with some fromage frais, feta, margarine, and honey. The reason he has this particular breakfast is because has eaten these foods for breakfast since he was a child in Iran, and this is how his mother prepared breakfast. However, there is an alternative to the honey. In lines 7 and 9, he initiates a search. After line 9, there is a long pause. In line 11, he produces the imperative of the Finnish word, sano, ‘say.2SG.IMP’, which is directed to the whole group (although it is in the singular). In other words, while this indicates that he wants someone else to take the turn and produce the missing word, he does not specify who should take the turn (cf. Sacks et al. 1974: 704). Speaker A continues this search by explaining that it is something that is made of fruit. T repeats the explanation and J offers the first suggestion, marmelaadi, ‘marmalade’. In line 17, T offers another possible dish made of berries, kiisseli, ‘fruit soup’. The initiator does not comment on these suggestions, but clarifies the search by stating that it something sweet (line 22). In line 23, T clarifies the search further by posing the question aamulla, ‘in the morning’. The search sequence goes on to side-track in line 25 when speaker M shows a bowl of candy to the initiator and this leads to laughter. In line 38, speaker A produces the following description for the missing word: mansikka joka keitetään, ‘strawberry that is preserved’. Finally, following 139 6 Word searches this description, the correct word is found-- hillo, ‘jam’. This correct word is greeted with a rising intonation and a repeat in line 40. (60) MV 47. <46:22> 01 A: 02 03 S: 04 A: 05 T: 06 A: 07 A: → 08 T: 09 A: → 10 11 12 A: → 13 14 T: 15 J: 16 S: 17 T: 18 S: 19 T: 20 ?: Minä laita-n (.) bon↑jour feta se-n pää-llä (.) floora-a? 1SG put-1SG (.) bonjour feta 3SG-GEN PRT (.) flora-PTV I put bonjour feta on top of it flora margarine? Ja (.) ö: (.) mikä (1) hunaja. And (.) VOC (.) what (1) honey And what honey. Huna[ja.] Honey. [Ja] he ihmettele-vät. And 3PL wonder-3PL And they wonder. Joo [jo-] Yeah yea [Kos]ka lapse-sta asti äiti on teh-nyt kotona, Because child-ELA since mother be.3SG do-PCP at.home Because since childhood mother has made (it) at home, Eli hunaja ↑tai (.) toinen mikä tehd-ään. So honey or (.) another what do-4 So honey or other that is made. Mhy, Uhhuh. Mikä se-n nimi on. What 3SG-GEN name be.3SG What is the name of it. (2) Sano. Say.2SG.IMP Say. Hedelmä-stä keitet-ään, Fruit-ELA make-4 It is made of fruit, (1) Hedelmä[-stä keitet-ään. Fruit-ELA make-4 Made of fruit. [Marme[laadi. Marmalade. [Marmelaadi. Marmalade. Kiisseli. Fruit soup. Marme↑laadi [ei.] Marmalade no. [Ei.] No. ºEi,º No, 140 6 Word searches 21 22 A: 23 T: 24 A: 25 M: 26 T: 27 A: 28 M: 29 T: 30 31 T: 32 A: 33 T: 34 35 T: 36 A: 37 S: 38 A: 39 T: → 40 A: 41 T: 42 A: [ºÖ:º] VOC [Mikä] se on makea syödä. What 3SG be.3SG sweet eat What is it sweet to eat. Aamu-lla? Morning-ADE In the morning? Aamu-l[la.] Morning-ADE In the morning. [Tämä?] ((shows a bowl of candy)) This? [Hedelm-i-stä?] Fruit-PL-ELA Of fruit? [he he he] Hedelm-i-stä [ei ei ei he he] ((laughter)) fruit-PL-ELA no no no he he he of fruit no no no he he [he he he] he he he [Ei ei ei,] no no no, ((laughter)) Puuro-a. Porridge-PTV Porridge. Ei: [ei. No: no. [Siis aamul-So morningSo in the morn((laughter)) Kiisseli. Fruit soup. Mansikka joka, Strawberry that, Kiisseli. Fruit soup. Ei mansikka joka keitet-ään, No strawberry that preserve-4 No strawberry that is preserved, Hillo-a. Jam-PTV Jam. ↑Hillo. Jam. Aivan. Exactly. Ilman ↑hillo-a (.) minä e-n syö aamiais-ta. Without jam-PTV (.) 1SG NEG-1SG eat breakfast-PTV Without jam I do not eat breakfast. 141 6 Word searches This example suggests that regardless of how long it takes, or how sidetracked the search might become, the participants are willing to find the correct word (cf. Egbert et al. 2004: 193–196). At the beginning of the example, A describes his breakfast and recalls how his mother would cook him breakfast when he was a child (lines 1-6). In lines 2 and 3, a word search occurs that is self-initiated and solved. The actual search begins in lines 7 and 9 when A attempts to recall what his mother used to serve besides honey. This search is accompanied by a long pause (2 seconds) and he subsequently states sano, ‘say’, pointing to speaker S, but the word-search sequence is not that easily solved. The search continues with explanations that the word in mind is something prepared by cooking fruit. Then other participants begin to suggest different types of fruit preserves (marmalade, fool; lines 15-18). In line 23, T wants to check again that they are still talking about breakfast. Until this point, even though the search is not yet solved, it has remained rather structured. In line 25, M begins a sequence that confuses the conversation for a short time. During this sequence, M takes a candy bowl that is on the table and shows it to the others, proposing it as the resolution of the search. T continues the search in line 26 and asks whether the preserve is made of fruit. Simultaneously, M and A are still laughing at M’s proposal and due to this, T receives a response to the question that is neither an affirmative nor negative. In line 31, T proposes that they might be talking about porridge which leads to a digression from the search. A redirects the topic to the search by saying that the preserve they are discussing is in fact strawberry. Finally, T produces the correct word (hilloa, ‘jam’; line 39), which is also accepted by A. Although this example can be interpreted as not being highly fluent in terms of individual fluency, it can be evaluated instead based on the conversational structure. From the point of view of the sequential organisation, it can be stated that other than the sequence that was initiated by M in line 26, virtually none of the turns overlap. The search somewhat resembles a game where the participants give turns to each other. This observation is in line with the assumption that although there are many repairs, this does not necessarily imply that the speaker is to be assessed as being non-fluent (Riggenbach 1991: 433). In this particular case, the participants are also involved in other matters than resolving a word search. For instance, they joke about the search and laugh together. (cf. Hüttner 2009: 288–289 regarding ‘having a good time’ in fluency studies.) 142 6 Word searches In the next example, E initiates a word search for the English noun, ‘clown’. Both C and J suggest a Finnish equivalent noun: C’s offers the word pelle in line 3 and J’s proposes klovni in line 4. In comparison to the previous example, klovni is an actual Finnish word, whereas kartuuni is not, even though they are both similarly formed. However, E does not accept either of these suggestions. In line 5, she requests clarification. C subsequently emphasises his own suggestion by extending it to an entire phrase: Se on pelle, ‘it is a clown’. J offers e again her own translation of klovni in line 7. In lines 9 and 11, C repeats his offer for a third and fourth time. After these repetitions, J begins to hesitate in line 12 and states en minä tiedä, ‘I do not know’.32 After J’s hesitation and withdrawal, E describes the context of the word use in lines 14–17. She explains that there might, for example, be a group of people where one person constantly tells jokes. This type of behaviour makes him or her look like a clown. In line 17, E uses the word offered by C pelle, ‘clown’, and in doing so, confirms the suitability of the word. (61) CC 29. <14:00/2> 01 E: → 02 03 C: → 04 J: → 05 E: 06 C: 07 J: 08 E: 09 C: Mi-tä on clown. What-PTV be.3SG clown What is a clown. (1) Pelle. Clown. Klovni. Clown. Mikä? What? Se ↑on pelle. 3SG be.3SG clown It is a clown. Klovni. Clown. Pe-Cl-↓Pelle. Clown. 32 Baumgarten & House (2010) studied the English expression ‘I don’t know’ in both native and non-native conversations and discovered that in L1 discourse, this expression is oriented towards the speaker and the recipient, whereas in lingua franca discourse, it is oriented towards the speaker only. In other words, in L1 discourse, the speaker who produces the expression can demonstrate that he or she is aware that there might a potential disagreement between speakers and therefore he or she invites the recipient to participate and to produce dissenting contributions. (ibid. 1196). The excerpt under analysis is somewhat antithetical to those findings. In this particular excerpt, speaker J has stated in lines 4 and 7 that the correct word is klovni, ‘clown’, but then she begins to hesitate in line 12. She acknowledges that her statement might have been incorrect and she therefore retracts it. 143 6 Word searches 10 E: 11 C: 12 J: 13 C: 14 E: 15 16 17 18 J: 19 E: ↑Pelle. Clown. Pelle on clown. Pelle be.3SG clown. Pelle is a clown. Ei: minä e-n tiedä ehkä minä e-n NEG 1SG NEG-1SG know perhaps 1SG NEG-1SG 1SG °minä e-n tiedä.° NEG-1SG know No I do not know perhaps I do not I do not know. Joo. Yeah. Että esimerki-ksi m: joskus on (.) ↑class clown. That example-TRANS VOC sometimes be.3SG (.) class clown. That for example sometimes (it) is a class clown. Esimerki-ksi että .hh jos on poru-- porukka ja sitten Example-TRANS that .hh if be.3SG gro-- group and then yksi (.) ihm-- ihmise-n siellä (.) vain (1) m: one (.) hum—human:being-GEN there (.) only (1) VOC <kerto-o vitse-j-ä.> tell-3SG joke-PL-PTV For example that if there is a ga- a gang and then there is one hum- human being there who just tells jokes. Koko aika. All time All the time. Ja ↓sitten hän on kuin pelle. And then 3SG be.3SG like clown And then he is like a clown. °Mhy.° Uhhuh. Ja jos (.) jos joskus (.) se ei: ole (1) ö: mukava-a And if (.) if sometimes (.) 3SG NEG be (1) VOC nice-PTV (.) kaiki-lle. (.) all-ALL And if sometimes it is not nice for all. As mentioned previously, when two or more speakers participate in a search or defining process, this might lead to a prolonging of the sequence. In addition, the fact that the speakers are non-native might also affect the process. For instance, a participant may know grammar better or have more knowledge of the language used than the other participants. On the other hand, when searching for lexical items, there might not be a speaker who knows ‘more words’ then the others and as a consequence, the search prolongs and provides an opportunity for more speakers to participate. It is also worth noting that when there are two or more suggestions or possible resolutions for the search, they are treated as offers, not as final resolutions. 144 6 Word searches The initiations of the search sequence were not complex. They were initiated verbally and this could have led to a direct offer. Syntactically and structurally, the search sequences were such that they required only one noun as an answer. Although the initiation is not complex, the search sequence may be prolonged for different reasons. One of the reasons can be a situation when two or more participants contribute to the search process by producing two or more possible variants. This type of communication strategy is referred to as semantic contiguity (Kurhila 2006: 116, Yule & Tarone 1997: 20). If a speaker uses the strategy of semantic contiguity, he or she attempts to identify the unknown referent by producing words such as synonyms, metonyms, or antonyms. After the identification, the participants can further negotiate about the actual referent (Kurhila 2006: 116). In the next example, the participants are attending an editorial meeting and they are discussing the different styles of conducting interviews in outdoor conditions. The main issue is what would be the correct way to obtain interviewees and what types of questions would be appropriate to ask. One possibility could be to ask them about their breakfast habits. In line 2, M produces a possible variant of a word she has in mind, aihe, ‘subject’. T confirms that word in line 3. After T’s affirmation, M produces another possible variant of the word she is looking for osa, ‘part’. Next, S produces yet another word, teema, ‘theme’ in line 5. The participants have now established a semantic field that includes synonyms. Finally, M uses the word osa in the sense of ‘a part of something’. (62) MV 45. <58:44> 01 M: 02 → 03 T: → 04 M: → 05 S: → .hh Jos lopu loppu-j-en lopu-ksi se on viimeinen .hh If end end-PL-GEN end-TRANS 3SG be.3SG final pätkä. part If at the end it is the final part. Jos >me teem-me noin< (.) pitä-ä järjestä-ä tää If 1PL do-1PL about (.) have-3SG organize-INF this jotenkin jos (.) .hh me ota-mme (.) somehow if (.) .hh 1PL take-1PL (.) yks ↑aihe da ºaiheº, one subject da subject If we do so we have to somehow if we take one subject da subject, Ai[he,] Subject, Osa? Part? Teema. Theme. 145 6 Word searches 06 M: Osa ä ö osa ruoa-sta jo↑ka vain (.) Part VOC VOC part food-ELA which just (.) vietnami↑laise-lla ↑tai kiinalaise-lla ↑tai somalilaise-lla Vietnamese-ADE or Chinese-ADE or Somali-ADE ↑tai venäläise-llä mi-stä vaan se °Irani-ssa.° or Russian-ADE what-ELA only 3SG Iran-INE Part of food that is only in Vietnamese or Chinese or Somali or Russian or in Iran. Thus, even though the search is focussed on one word, several factors can contribute to the search sequence taking longer to resolve. The first is that a speaker’s insufficient knowledge of a language might lead to participants negotiating for the correct grammatical form. Second, when searching for lexical items, the use of other languages can misdirect the search and can even prolong it. Third, the search could be prolonged by when the speakers propose synonyms or other semantically contiguous referents. On the other hand, by listing constructions, the interlocutors can negotiate what is actually being searched for. As Lerner (1993: 26–27) observes: “list construction can provide the resources to respond to a proffered completion without explicitly rejecting it, yet without accepting it.” 6.4 Reaction to the resolution The final stage of a successful search sequence concerns how the resolution is evaluated. Analysing the previous examples, I have also discussed the reactions of the search initiator, but I will now concentrate on those reactions more thoroughly. 6.4.1 The original speaker repeats or otherwise accepts the offered word When I analysed the collaborative productions, I demonstrated that the original speaker has the right to continue speaking after the final component of the collaborative productions has been produced. That same rule also applies to word searches and the structure of the search sequence also makes this possible. Word-search sequences are typically designed so that speaker usually conducts the search near the end of the unit, and produces only the next word that is being searched for, and this creates a place for the coparticipant to complete the unit in progress. Following the search, the primary speaker may continue his or her turn. (Lerner 1996: 261–263.) The speaker who initiated the search usually repeats the offered word. In the first example, B begins to recount that her girlfriend’s father has a dog and that he goes to the forest to do something, but then B does not recall the 146 6 Word searches actual verb that is used to describe the activity that the father engages in. In line 4, C and J produce both the accurate Finnish word, metsästää, ‘to hunt’. (63) FC 102. <15:00> 01 B: → 02 R: 03 B: → 04 L → 05 B: Mutta hän isä mene-e >mikä se on< am-- ampun-, But 3SG father go-3SG what 3SG be.3SG sh-- shooBut he father goes what is it sh- shoo-, Tule-e. Come-3SG Comes. Ja: tarvitse-e ↑koira-a. And need-3SG dog-PTV And needs a dog. Metsästä-ä. Hunt-INF To hunt. Metsästä-ä ↑kyllä. Hunt-INF yes To hunt, yes. This example differs slightly from the typical word-search sequences because these sequences usually lack a lexical noun, but the missing word in this example is a verb. The following example also involves a missing verb, but the reaction to the offer is a verbal acceptance rather than a repeat. In line 1, E begins to hesitate and does not produce the infinitive form. In line 2, J helps out and E accepts the word in line 3 by stating joo, ‘yeah’. After E’s acceptance, J repeats the whole sentence. (64) CC 15. <35:35/1> 01 E: 02 J: 03 E: 04 J: 05 C: Sit↓ten (.) hän halua-a (2) ö:, Then (.) 3SG want-3SG (2) VOC Then he wants to, ↑Pelä-tä, Be:afraid:of-INF Be afraid of. Joo, Yeah, Hän halua-a pelä-ä, 3SG want-3SG be:afraid:of-INF He wants to be afraid, °Joo joo.° Yeah, yeah. As was evident in the first example, the initiator of the word search can accept the offered word and subsequently incorporate it as part of his or her own utterance. The initiation is most often formed so that it tells the coparticipant what type of structure is being searched for. In addition, the 147 6 Word searches placement of the search at the end of a unit helps the co-participant to participate. In the following example, J states that not everyone is able to tell jokes and perhaps there is some sort of quality that can contribute to joke telling. However, she does not remember the word in Finnish and initiates a word search in English in line 4. Following a long pause, she produces an English word that could capture the meaning that she intends. In line 8, E offers the Finnish word kokemus, ‘experience’ in the nominative case. J takes that word and declines it in the partitive case, which is needed for the whole construction. After declining the noun, she produces the entire phrase of se tarvitsee kokemusta, ‘he needs experience’. (Brouwer et al. 2004: 85.) (65) CC 16. <36:37/1> 01 J: 02 03 C: 04 J: → 05 06 07 08 09 J: E: → J: Mutta ↑itse minä e-n osaa kerto ↓vitse-j-ä. But myself 1SG NEG-1SG know tell.INF joke-PL-PTV But I myself I do not how to tell jokes. Minä on aina vain kuuntele (-). 1SG be.3SG always only listen (-) I always just listen. Joo. Yeah. Kun se on (.) minä luule-n se tarvitse (2) e: When 3SG be.3SG (.) 1SG think-1SG 3SG need.3SG (2) VOC how to say it in ºEnglish I don’t know.º When it is I think it needs, how to say it in ºEnglish, I don’t know.º (3) Experience maybe. (2) Kokemus. Experience. ↑Kokemus-ta se tarvitse-e kokemus-ta. Experience-PTV 3SG need-3SG experience-PTV Experience it needs experience. A search can also be resolved incorrectly. Nevertheless, these incorrect resolutions do not lead to disorder. For example, in the next order, J is speaking about a friend of hers who does not understand jokes. In line 6, J begins to search for a word. She cuts-off the word san-, then re-introduces the phrase joskus, ‘sometimes’, and inhales. She also produces a hesitation sound and pauses before producing the missing word in English, ‘behaviour’. In line 7, C tries to translate that word into Finnish and first produces käy-, then käyttäy-, and he subsequently verbalizes his offer by using the structure mikä se on käyttäy-, ‘what is it käyttäy-‘, and further elaborates his own offer of joku käyttäy-, ‘something käyttäy-’. 148 6 Word searches Subsequently there is a pause before J accepts C’s offer. That offer is a beginning of the word that adds to it the suffix -minen and forms the Finnish word, käyttäminen ‘using’. This form can have two interpretations, as it can be either a Fennicized word, or it can be an actual word that can be translated into Finnish as ‘use’. If we support the second position, that käyttäminen is ‘use’, it is evident that this is not the same word as behaviour, which would be translated into Finnish as käyttäytyminen. However, both words share the same sounds. In addition, J’s ability to formulate a Finnish word that has a suffix -minen requires her to have a rather good knowledge of Finnish (Kurhila 2006: 116). (66) CC 26., 27. <05:17/2> 01 J: 02 C: 03 J: 04 E: 05 J: 06 → 07 C: → 08 09 J: 10 C: 11 J: Minu-lla on ↑yksi kaveri (.) 1SG-ADE be.3SG one friend I have one friend °Mhy.° Uhhuh. Joka (.) joka ↑ei ymmärtä vitse-j-ä. Who (.) who NEG.3SG understand joke-PL-PTV Who does not understand jokes. Mhy. Uhhuh. Mt mutta se ö: kestä pi:tkä aika aika kun minä VOC but 3SG VOC take.3SG long time time when 1SG ymmärs-i-n se-n. understand-PST-1SG 3SG-GEN But it took a long time before I understood it. <Joskus minä san-- sano-i-n> joskus .hh hän-en ö: Sometimes 1SG sa-- say-PST-1PL sometimes .hh 3SG-GEN VOC (1) behaviour, (1) behaviour ((taps the table with her nails)) Sometimes I said, sometimes his behaviour, Käy-- ↑käyttäy-- >mikä se on< käyttäy-- joku käyttäy-Beh-- behavi-what 3SG be.3SG behavio-- some behavio— Beh- behavi- what is it. Behavio- something behavio(1) Hän ei: (.) hän-en (.) käyttäminen, 3SG NEG.3SG (.) 3SG-GEN (.) using He is his using, Joo. Yeah. Käyttäminen (.) ol-i ↑no (1) tosi hauska ja minä Using (.) be-PST.3SG well (1) very funny and 1SG aina (.) sano-i-n vitse-j-ä hän-en näköinen always (.) say-PST-1SG joke-PL-PTV 3SG-GEN look:a:like on myös hauska. be.3SG also funny. Using was very funny I always told jokes his look-alike is also funny. 149 6 Word searches The data contain few cases of rejection or overtly displayed disbelief. In these cases, the initiator can express his or her disbelief in the third turn (Kurhila 2006: 43, Schegloff 1992: 1304–1317). In the next example, speaker M wants to know how they can make sound more audible to the audience. She inquires in line 1 tarvitseko ö tämä, ‘do we need this’, and then pauses. A adds the word kaiuttimet, ‘loudspeakers’. M and A subsequently create a side sequence in lines 3–4 when M asks for a clarification by using mikä, ‘what’, and A repeats the word kaiuttimet, ‘loudspeakers’. In line 5, M exhibits uncertainty over the word search and initiates a third-position repair. She repeats the offered word, pauses, and produces voi olla, ‘might be’, in a quiet voice. After an unclear response that was produced by speaker A, M explains further that she is referring to the technical solutions for the sound to be audible. M’s utterance in line 5, voi olla, can be interpreted as her not completely accepting the offered word. In other words, loudspeakers can be a part of the solution, but she also has other issues on her mind. (67) MV 55. <29:45> 01 M: → 02 A: → 03 M: → 04 A: 05 M: → 06 A: 07 M: A: miten <ääni ö kuule-mme> (.) tarvits-e-ko ö tämä, VOC how voice VOC hear-1PL (.) need-3SG-Q VOC this And how do we hear voice do we need this, Kaiuttime-t. Loudspeaker-PL Loudspeakers. Mikä? What? Kaiuttime-t. Loudspeaker-PL Loudspeakers. Kaiuttime-t (.) ºvoi olla.º Loudspeaker-PL (.) might.3SG be Loudspeakers might be. (Mhy.) Uhhuh. Se kuule-e ääni mi-tä me tee-mme teknise-sti. 3SG hear-3SG voice what-PTV 1PL do-1PL technical-ADV It hears voice what we do technically. This type of third turn that occurs after the repair creates a chain between the repairable and how the repair is evaluated. The co-participant can show that he or she does not understand the self-initiated search. Thus, the first speaker (self) can repair the problematic understanding. This can constitute a third-position repair (Schegloff 1992: 1301). Furthermore, this can be interpreted as forming a part of those procedures that the participants use to 150 6 Word searches maintain intersubjectivity, which is mutual understanding. When intersubjectivity faces a risk of breakdown, the participants have the means to remedy of the situation. Schegloff (ibid. 1338) emphasizes the procedural nature of those actions that the participants can take. However, there is no need to specify prematurely what the trouble might be, or what may have caused it. In short, the participants can manage the problems locally. 6.4.2 Original speaker continues directly The initiator of the word search can also continue his or her speech without any specific reference to the offered word. In these cases, the speaker accepts the suggested word but does not repeat it or otherwise evaluate it. Later, the current speaker can integrate the suggested word into his or her own turn. In the first example, speaker J is talking about immigrants who engage in voluntary work. In line 2, she begins to hesitate in her word choice. (68) PM 7. <54:26> 01 J: 02 → 03 D: → 04 J: He ovat (.) yhdessä että se on 3PL be.3PL (.) together that 3SG be.3SG vähän sitten ero kun su-lla ↑on jo (.) little then difference when 2SG-ADE be.3SG already (.) jotain ja ole-t heidä-n kanssa, something and be-2SG 3PL-GEN with They are together and there is a little difference when you already have something and you are with them, Ehkä ↑tee-t vapaaehtoistyö-stä (1) kuin yht-- oma-ssa Perhaps do-2SG voluntary:work-ELA (1) like coo-- own-INE yhteiskunna-- ei (-) mutta [(.) >piire-i-ssä societ-NEG (-) but (.) circle-PL-INE piire-i-ssä.< circle-PL-INE Perhaps you do voluntary work like in own societ- no but in circles in circles. [Yhteistyö-ssä. Cooperation-INE In cooperation. Että e:-t (.) sinä e-t mene mihinkään mui-hin kun That NEG-2SG (.) 2SG NEG-2SG go anywhere else-ILL when sinu-lla on jo, 2SG-ADE be.3SG already That you do not go anywhere else when you already have, As mentioned previously, the speaker who initiated the word search can also integrate the offered word into his or her own turn and through that, demonstrate that the offer has been accepted. In the following example, E’s 151 6 Word searches offer occurs in a form that is needed for the construction. The modal verb voida, ‘to be able’, requires an object that is in the A-infinitive (which is called the dictionary form; VISK § 469). (69) CC 17. <37:55/1> 01 J: → 02 E: → 03 J: 04 E: 05 J: Se ↑on minä luule-n jos ö: (.) ↑kertomus vitse-j-ä 3SG be.3SG 1SG think-1SG if VOC (.) story joke-PL-PTV on myös tosi tärkeä kuitenkin tilanne on be.3SG also very important however situation be.3SG vaikea ja se on .hh e: kuin sinä voi-t e: difficult and 3SG be.3SG .hh VOC like 2SG can-2SG VOC It is I think if a story and jokes are also very important, however, a situation is difficult and it is you can Men-nä. Go-INF To go. M-- että jou sinä voi käyttä huumori ja men-nä pois tä-stä M-- that VOC 2SG can use.INF humour and go-INF away thisELA >tilannee-sta< vaikea-sta tilannee-sta. situation-ELA difficul-ELA situation-ELA That you can use humour and get away from this difficult situation. °Mhy.° Uhhuh. Se on tosi tärkeä m- joskus n- on (1) 3SG be.3SG very important VOC sometimes VOC be.3SG (1) sem↑mois-i-a tilanne-i-ta. such-PL-PTV situation-PL-PTV It is very important sometimes there are such situations. (70) TI 21. <4:54> 01 N: → 02 E: → 03 N: Èt-i vy ↑tože zna-ete labora-- [èt-i laboratori-ju This-PL 2PL also know-2PL labora-- this-PL laboratory-ACC i vs-e tak-ie vešč-i. and all-PL such-PL thing-PL These you also know labora-, these laboratories and all other things. [.h laboratori-i konečno. .h laboratory-PL of course. Laboratories, of course. A kogda v Rossi-ju v Rossi-i by-l-i tam But when PRE Russia-ACC PRE Russia-PREP be-PST-PL there >mnogo tadžik-ov< vy tak poznakomi-l-i-s’. many Tajik-PL.GEN 2PL that get:acquainted-PST-PL-REFL A when you were in Russia, you got acquainted with many Tajik there. 152 6 Word searches As is evident in example 69, if the suggested word occurs in the correct form, it can be accepted without specific reference to its suitability. 6.4.3 Either the co-participant continues or the continuation is longer As mentioned earlier, the word-search sequence is typically designed so that only the sought after word is produced and the initiator of the word search may then continue speaking. It is also possible that the speaker who offered the word might produce a longer unit or continue with his or her talk. In the next example, speaker N is struggling with a specific Russian concept. This is because it is customary in Russian academic speech to construct compound nouns with one part of an adjective and one part of a noun. For example, filfak: filologičeskij fakul’tet, would be the ‘Faculty of Philology’. In this particular example under discussion, the participants are speaking about the educational system of Tajikistan and especially about the doctoral training there. In line 1, N tries to formulate the appropriate word and initiates the search by stating v ètom med-, ‘in this med-‘. The word med- is cut-off and the utterance ends up having a slightly falling intonation. In line 2, A acknowledges that the idea in N’s utterance was correct and confirms it by using the affirmative particle da, ‘yes’, then she takes the whole construction and recycles it as a part of her own utterance: v medkolledž postupila, ‘I was admitted to the medical institute’. The word medkolledž, ‘medical institute’ consists of the truncated adjective med- (from medicinskij, ‘medical’) and a noun, kolledž (college). This type of truncation is typical of the Russian colloquial speech that is used in educational and vocational spheres (Zemskaja et al. 1981: 126–127). (71) TI 37. <8:32> 01 N: → 02 A: → 03 N: .h A posle posle škol-y (.) posle vy srazu .h but after after school-GEN after 2PL immediately postupi-l-i v ↑èto-m (.) med--, be:admit-PST-PL PRE this-PREP (.) med-But right after school and then you immediately got admitted to this med, Da v medkolledž-ø [postupi-l-a. Yes PRE medical:institut-ACC be:admit-PST-FEM Yes, I was admitted to the medical institute. [Èt-o by-l-o ↑vs-ë 3SG-NEU be-PST-NEU all-NEU po-rus-sk-i. PRE-Russian-ADJ-ADV It was all in Russian. 153 6 Word searches 04 A: .h Èt-o by-l-o da vs-ë, .h 3SG-NEU be-PST-NEU yes all-NEU It was yes all, Even though it is possible that the search sequence is prolonged due to the appropriate word or expression is not being found easily, other instances also occur. The following example illustrates that while the language used is foreign to the participants, this does necessarily restrict their innovative use of the language. In line 1, speaker M initiates the search by first cutting off the word so-, then he cuts off the word surf- and verbalises his search by asking mikä se oli, ‘what was it?’ Speaker R helps him in line 2 and produces the needed Finnish word, surffata, ‘to surf’ in the infinitive form. M accepts the suggested word by repeating it in the correct form, surffas-i-n, ‘surf-PST-1SG’. Speaker S subsequently joins the conversation and produces a word that does not exist in Finnish, which could be analysed analogically as turffail-i-n, ‘turf-PST-1SG’. (72) FC 106. <11:08> 01 M: → 02 R: → 03 M: 04 S: → 05 N: 06 M: 07 R: Minä e-n tei tuota minä so-- surf-- ä: 1SG NEG-1SG VOC well 1SG so-- surf-- VOC mikä se ol-i. what 3SG be-PST.3SG I do not, well, I surf-, what was it? Surffa-- ↑surffa-ta. Surf-surf-INF. To surf. Surf↓fas-i-n. Surf-PST-1SG. I surfed. Turffail-i-n. Turf-PST-1SG I turfed. Surffail-i-n. Surf-PST-1SG I surfed. Surffail-i-n ↑neti-ssä, Surf-PST-1SG Internet-INE I surfed the Internet. ↑Ja mer-ta. And sea-PTV. And the sea. There are few examples of other continuations in the word-search sequences in my data. Usually only the searched word is produced and the co-participant does not adopt a stance in relation to the offered word. The 154 6 Word searches result is that the suitability or the appropriateness of the word is evaluated after the actual word-search sequence. 6.5 Unsolved cases Whenever a search sequence is initiated, there is always a chance that it can fail. However, as previous examples demonstrate, the participants are willing to invest heavily in terms of times and effort to ensure the search sequence is effectively solved. As a consequence, unsolved searches are uncommon. Moreover, one of the reasons for a failure can be the use of other languages than the spoken lingua franca. As has been discussed earlier in this study, when the speakers use other languages, these can affect the ongoing word search in many ways. While it can speed up the sequence, it can also prolong it. The following example features speaker M attempting to find a Finnish equivalent for the word opportunities. First, he initiates the search in line 1 by stating: minä sanon englanniksi opportunity, ‘I say in English opportunity’. Second, he states that he thinks that Helsinki is a good city (line 2) and continues his turn in line 3 with the statement of Helsingissä on paljon opportunities, ‘Helsinki has a lot of opportunities’. He has used the word ‘opportunity’ twice before actually admitting that he does not know the word in Finnish. In line 7, he asks his interlocutor what ‘opportunity’ is in Finnish and indicates that he wants that the interlocutor to take the turn (Sacks et al. 1974: 704). Nonetheless, in line 8, speaker L admits that he does not know the word in Finnish. In line 9, the original initiator of the search produces a dialogic particle mhy, ‘uhhuh’. (73) FC 89. <31:43> 01 M: 02 03 → Minä sano-n ↑englanni-ksi opportunity ja hän sano-o mikä 1SG say-1SG English-TRANS opportunity and 3SG say-3SG what se on (-) ↓Helsingi-stä. 3SG be.3SG (-) Helsinki-ELA I say in English opportunity he says what it is of Helsinki. Minä sano-n (.) e: Helsinki on hyvä kaupungi (.) 1SG say-1SG (.) VOC Helsinki be.3SG good city (.) kaupunki mutta e: city but VOC I say Helsinki is a good city but Ja (1) ↑Helsingi-ssä on (.) paljon opportunities And (1) Helsinki-INE be.3SG (.) lot of opportunities minä en tiedä ↑mikä on opportunity suome-ksi. 1SG NEG-1SG know what be.3SG opportunity Finnish-TRANS And there are a lot of opportunities in Helsinki I do not know what is opportunity in Finnish. 155 6 Word searches 04 05 M: 06 ?: 07 M: → 08 L: → 09 M: (2) Sitten hän >kirjoitta-a< sii-hen (-) Then 3SG write-3SG 3SG-ILL (-) Then he writes to it (↑suome--) Finnis-Sano: ↑sano (.) opportunity. Say.IMPER.2SG say.IMPER.2SG (.) opportunity Say say opportunity. Mutta en ↑tiedä mi-tä se on suome-ksi. But NEG-1SG know what-PTV 3SG be.3SG Finnish-TRANS But I do not what it is in Finnish. °Mhy.° Uhhuh. What is interesting in these cases is that the current speaker is highly determined to find the equivalent so that he compels the recipient to produce the correct word. He wants them to reach a common ground before moving onwards even though he knows the correct word in English. In the previous example, the initiator of the word search did not know the word in Finnish even though he had thought of the correct word in in English. The current speaker specifically wanted to have the appropriate equivalent. However, the co-participants did address the issue under inspection. The interruption of the ongoing development of the talk in progress makes the unavailability of the word relevant. That is, the talk in progress is not “moving toward its projected completion” (Goodwin & Goodwin 1986: 55–56). On the other hand, a halt in the progression of conversation can also threaten the achievement of mutual understanding (Heritage 1996 [1984]: 249–257). However, participants can engage in defence intersubjectivity during conversation. This defence is procedural and locally managed. (Schegloff 1992: 1338, see also Bazzanella & Damiano 1999.) Furthermore, there is no need to count the specific reasons for its occurrences, as it is locally adapted. When analysing the single passages of conversation, it can be implemented in those circumstances that are understandable only to those participants present in that particular conversation. (Schegloff 1992: 1338.) 6.6 Summary Participants have a variety of ways to express that they are having problems forming utterances. The current speaker can pause during his or her turn, produce hesitation sounds, or indicate directly that he or she is missing a word. The co-participant is usually the only speaker to produce sought-after 156 6 Word searches word and the original speaker can subsequently continue his or her turn. The original speaker can accept the offered word by repeating it, by producing an affirmative particle, or by integrating the word into part of the next utterance. If the search sequence prolongs, it is usually because the word has not been determined. Another possibility is that if the search sequence continues longer, the process of determining the correct word becomes the actual action. Although the suitable word is not found immediately, this does not suggest that the participants would begin arguing or behave in a hostile manner. Searching for the appropriate word might also create a sense of participants enjoying themselves. In NNS interaction, no single participant is most. While usually a distinction can be made between whether the search regards a lexical item or a grammatical issue, this distinction may not be relevant. There are some cases in which the grammatical appropriateness surfaces in the conversational so that it causes disturbances in the ongoing conversation. During grammatical searches, the participants can search for the correct form together and decide it in cooperation although in the terms of native Finnish language, the form might be nongrammatical. In other words, participants do not encounter problems with linguistic structures in a sense that they would not know how to decline them, such as a noun, but the problem is that they are not familiar with the noun that they want to decline. Thus, the extensive number of Finnish cases or the moderate number of Russian cases is not necessarily the factor that causes problems. The grammatical searches involving both participants are only found in the Finnish data. Although the Finnish and Russian languages have the same type of overall structure and they both possess a moderate number of morphological categories, only the participants in the Finnish conversations become involved in grammatical searches. This might be due to the circumstances in which the languages had been learnt. Considering Finnish, the lingua franca speakers are usually adults who have learnt Finnish in different language courses, which could mean that they more oriented to grammatical issues. The speakers also have usually emigrated to Finland, suggesting that the Finnish language had not been a part of their lives before. However, the lingua franca speakers of the Russian language have had different histories. The lingua franca speakers of Russian may have a more native-speaker relation to the language. For this reason, they do not consider grammatical appropriateness as important as the lingua franca speakers of the Finnish language (see, e.g. Kurhila 2006: 132–133). 157 6 Word searches As defined in the introduction, the focus of this study is not primarily on the actual language form that is used in conversations. However, some of the possible tendencies in language use can be identified. The possible cases of ungrammatical language use are occasional in the sense that not every speaker shares the same forms, or uses the language similarly. Second, as mentioned earlier, they are merely tendencies, as I have not analysed every single occurrence. In general, these cases may be useful later in planning future research plans. The question of language form becomes increasingly evident when the participants negotiate the correct grammatical forms or other appropriate language use. The offers for the word-search sequences are dominated by either the nominative case or the infinitive form of the verb. These are referred to as the basic or dictionary forms. Three possible explanations can account for this. Firstly, the way the sequence is initiated required only this type of basic form; secondly, the offered word is in the nominative case, thirdly; the verb is in the infinitive. After the offer, the initiator of the sequence has an opportunity to decline the noun in the correct case. For instance, in the Finnish examples, the question mikä se on, ‘what is it’, usually reveals that the searched word is a noun and that it can be produced in the nominative case. With special reference to the Russian language, there are certain categories that might undergo changes during Russian lingua franca interaction. First, the grammatical form of the noun after a cardinal number fluctuates. In example 16, the speaker first used the correct form of the noun after the cardinal number pjat’, ‘five’: pjat’ procent-ov, ‘five percentPL.GEN’. Later in that same excerpt, she used the nominative case in the plural after the cardinal: pjat’ procent-y, ‘five percent-PL.NOM’. A second category that is liable to changes is gender. For example, the third-personal pronouns that usually reflect the referent’s gender might be used differently. When discussing the semantics of the language, one of the tendencies in lingua franca interaction is the use of semantically close words. In example 19, the speaker used two semantically close verbs that were related to studying. From a native speaker’s perspective, the verb use might be inappropriate, but the non-native speakers accept the usage without hesitation. 158 7 Candidate understandings – somewhere in the middle? In the chapter 6, I focused on examples that contained a conversational understanding that surfaced in and through a disruption in the progress of the conversation. Rather than providing the next relevant turn, either the current speaker or the recipient may focus on some elements in the prior speech. This type of disruption makes the unavailability of a word relevant to the conversation. In these cases, the participants orient to establishing mutual understanding so that it disrupts the progress of the conversation. However, this orientation does not always completely disrupt the progress of conversation. The recipient can take multiple actions should he or she experience that something problematic is arising in the prior speaker’s turn. The recipient can either repair the problem, or he or she can take one step forward and formulate how he or she understands the problematic utterance. Those type of instances in which the co-participant articulates his or her interpretation of the previous turns are referred to as candidate understandings (Kurhila 2006: 153–217). In terms of conversational repair, these articulations, or candidate understandings, as I will refer to them in this chapter, can be categorised as other-initiations of repair. In other words, they present a possible understanding of the trouble turn, which is subsequently confirmed or disconfirmed by the speaker of the trouble turn (cf. Schegloff et al. 1977). Thus, the current speaker produces a turn which the recipient interprets which is then evaluated by the first speaker. These interpretations can consist of one lexical word (usually a noun), or it can also be a full clause that contains a finite verb (table 7). Furthermore, the beginning of the unit can contain a response particle. These words and larger units reformulate the previous turns or rephrase it. 159 7 Candidate understandings – somewhere in the middle? Table 7. Candidate understandings. Candidate understandings Finnish Russian Full clauses 4 11 Noun phrases 4 4 Total 8 15 7.1 Full clauses First, I will concentrate on those cases that contain a full clause. Repair initiations are often unfocussed in that they do indicate that the prior speaker’s turn was somehow problematic, but they allow the speaker to decide on where the problem is located. For example, the co-participant can display her difficulty in producing a relevant response by initiating an openclass repair (‘sorry,’ ‘what’, etc., see Drew 1997), by verbalizing his/her difficulty (‘I don’t quite follow.’) or by requesting clarification from the speaker (‘What do you mean?’). In the first example, the participants, N and E, engaged in a discussion on the language situation in Tajikistan. They mention that their workplaces have workers of different nationalities, predominantly Russians, Tajiks and Uzbeks. In lines 1 through 5, the participants are listing those nationalities and N is attempting to determine the language that they use. In line 6, N begins her turn with the particle a, and E subsequently produces a clause that connects all the previous nationalities into one, vse nacional’nosti ‘all nationality-PL’. In line 8, N begins her turn again with the particle a and reformulates the turns that are being produced prior to that and produces the full clause A: a vse govorili russki, ‘oh, all spoke Russian’. (74) TI 2. <27:04> 01 N: 02 E: 03 N: Aga vse-taki na↑čal’nik-i, Aha however chief-PL However chiefs, Aga. Aha. Vs-e ↑rus-sk-ij, All-PL Russian-ADJ-MASC All Russian, 160 7 Candidate understandings – somewhere in the middle? 04 05 E: → 06 N: 07 E: 08 N: → 09 E: 10 N: Aga a ra↑boč-ie by-l-i tadžik-i, Aha but worker-PL be-PST-PL Tajik-PL But the workers were Tajik, Raboč-ie ↑tadžik-i ↑uzbek-i, (---) Workers-PL Tajik-PL Uzbek-PL Workers are Tajik people Uzbek people, A: VOC Vse (-) nacio↑nal’nost-i, All-PL (-) nationality-PL All nationalities, A: a vs-e govori-l-i (.) rus-sk-ij. VOC but all-PL speak-PST-PL (.) Russian-ADJ-MASC But all spoke Russian. Tadžik-sk-ij rus-sk-ij (-), Tajik-ADJ-MASC Russian-ADJ-MASC Tajik Russian, Aga horoš-o. Aha good-ADV Aha okay. These types of reformulations and interpretations of the previous turn(s) can be referred to as paraphrases (Kurhila 2006: 161). The function of these is to recapitulate what has been stated earlier and they usually begin with different types of particles. In addition to particles, different conjunctions can also function as a paraphrase the previous turn(s). To paraphrase a previous turn, the conjunction is placed at the beginning of the paraphrase in the next example. At the beginning of the excerpt, speaker N verifies whether she has understood correctly that her co-participant did not have the opportunity to study foreign languages during her school time. However, in lines 2 and 4, the co-participant corrects this saying that she had studied German, but not that effectively. In line 6, she continues to explain that they currently have a German teacher who is good. In line 8, speaker N continues and produces another main clause, no netu s kem obščat’sja da, ‘but there is no one with whom to talk’. Since it is customary to think that in order to learn foreign languages and become a fluent speaker, one has to have an opportunity to use the language. If a learner does not have this opportunity, his or her language skills may be fragmented. This opinion is again confirmed in lines 11–12: the participants state that it learning a language requires effort and practice. 161 7 Candidate understandings – somewhere in the middle? (75) TI 11. <18:03> 01 N: 02 E: 03 N: 04 E: 05 06 N: 07 E: → 08 N: → 09 E: U vas [ne by-l-o <ne↑mec-k-ij PRE 2PL.GEN NEG be-PST-NEU German-ADJ-MASC ang↑lij-sk-ij,> English-ADJ-MASC You did not have German English, [(-) nemec-k-ij [by-l-o (-) German-ADJ-MASC be-PST-NEU German was. [ne↓mec-k-ij. German-ADJ-MASC German. Nemec-k-ij jazyk by-l-ø. German-ADJ-MASC language was-PST-MASC German language was. .h Nemec-k-ij jazyk vy↑uči-l-ø no v: ne tak .h German-ADJ-MASC language learn-PST-MASC but VOC NEG that horoš-o. good-ADV I learnt German language but not that good. Zde’s ne n:, Here NEG n: Here is not, Est’ ↑od-in nemec-k-ij prepodavatel’ on horoš-ij Be.3SG one-MASC German-ADJ-MASC teacher he good-MASC [on horoš-o zna-et. he good-ADV know-3SG There is one German teacher he is good he knows. [Aga .h no ↑netu s ke-m obšča-t’-sja da. Aha .h but NEG PRE who-INST talk-INF-REFL yes But there is nobody to talk with. Nu da s ke-m, Well yes PRE who-INST Well yes with whom, The conjunction in the previous example was both contrastive and coordinate. However, it is also possible to use conjunctions that are by nature more subordinate. In the following example, C is describing that he is not very fond of written stories that are meant to be hilarious. In line 1, he states that they might have some humoristic element in them, but not that much. In line 3, he begins to recount how he reacts to that genre of written stories by stating, Mä en naura kovasti kun, ‘I don’t laugh loudly when’. In line 4, J participates and produces a clause that connects the different sides of the previous turns and continues the thought that was initiated by Speaker C. However, this case differs from the collaborative productions and word searches because the projection changes. As seen in line 3, C refers to 162 7 Candidate understandings – somewhere in the middle? himself and uses the first-person singular. Subsequently, in line 4, J shifts the perspective and refers to C by using the-second person singular. (76) CC 24. <02:04/2> 01 C: 02 J: 03 C: → 04 J: → 05 C: → Mä ↑luule-n että (2) m: (.) .hh ne voi-vat olla v:ähän 1SG think-1SG that (2) VOC (.) .hh 3PL can-PL3 be a little hh huumorinen mutta (.) mutta ei: (1) kova kovin hh humorous but (.) but NEG very very I think they can be a little humorous but not that much Mhy. Uhhuh. Mä e-n naura ↑kova-sti kun kun, 1SG NEG-1SG laugh hard-ADV when when I do no laugh hard when, >Kun kun< lue-t joo. When when read-2SG yeah. When you read yeah. Kun lue-t joskus ↑on ookoo mutta When read-2SG sometimes be.3SG okay but tavallise-sti ei ole niin niin h(h)auska. usual-AD NEG.3SG be so so funny When you read sometimes is okay but usually is not that funny. I have used the term conjunction for this case, even though the same lexical word kun might also be interpreted as functioning as a specific type of particle. These types of utterance particles and conjunctions are usually the same words, but they exhibit some differences. The utterance particles are used to analyse and observe interaction. These particles indicate how the utterance is connected to the preceding speech. Thus, they connect turns and utterances either in one speaker’s speech, or in multiple speakers’ speech and they are usually placed at the beginning of the turn or the utterance. These particles do not function as indications of syntactic relations. For example, in spoken language, the particle kun is usually used to signal different types of explanations and reasoning, but this does not mean that the utterance launched by it should be interpreted as being syntactically subordinate to the preceding utterance. If the same word kun is analysed as a conjunction, the situation is different. In those cases, the conjunction functions as a sign of subordination. Furthermore, there is a connection between the preceding clause and the clause initiated by the conjunction. This connection might be temporal, causal or contrastive. (VISK § 801 806, 812, 818.) In this particular example, the conjunction is used in both turns. In the preceding turn, it is placed at the end of the turn, and in the following turn, it 163 7 Candidate understandings – somewhere in the middle? occurs at the beginning. By using this conjunction, speaker J demonstrates how she understood the previous turns that were produced by speaker C. She picks up the conjunction, shifts focus, and produces the needed part for C’s utterance to be completed. Let us consider the following example 33: J is telling of her friend who does not understand jokes. However, J did not realise that immediately. A word-search sequence that is initiated in line 6 creates a side sequence until speaker J becomes certain that the Finnish equivalent is käyttäminen ‘using’. Speaker J then produces the particle no and moves forward. J states that she used to tell jokes and she thought that her friend also understood them because her friend appeared to be enjoying his or herself. Then E paraphrases J’s previous turns by saying koska hän ei ymmärrä, ‘because he/she doesn’t understand’. J confirms this paraphrase in line 13. (77) CC 27. <05:17/2> 01 J: 02 C: 03 J: 04 E: 05 J: 06 → 07 C: → 08 09 J: 10 C: 33 Minu-lla on ↑yksi kaveri (.) 1SG-ADE be.3SG one friend I have one friend °Mhy.° Uhhuh. Joka (.) joka ↑ei ymmärtä vitse-j-ä. Who (.) who NEG.3SG understand joke-PL-PTV Who does not understand jokes. Mhy. Uhhuh. Mt mutta se ö: kestä pi:tkä aika aika kun minä VOC but 3SG VOC take.3SG long time time when 1SG ymmärs-i-n se-n. understand-PST-1SG 3SG-GEN But it took a long time before I understood it. <Joskus minä san-- sano-i-n> joskus .hh hän-en ö: Sometimes 1SG sa-- say-PST-1PL sometimes .hh 3SG-GEN VOC (1) behaviour, (1) behaviour ((taps the table with her nails)) Sometimes I said sometimes his behaviour, Käy-- ↑käyttäy-- >mikä se on< käyttäy-- joku käyttäy-Beh-- behavi-what 3SG be.3SG behavio-- some behavio-(1) Hän ei: (.) hän-en (.) käyttäminen, 3SG NEG.3SG (.) 3SG-GEN (.) using He is his using, Joo. Yeah. The word-search sequence that occurs in lines 6–11 was analysed before. 164 7 Candidate understandings – somewhere in the middle? 11 J: 12 E: → 13 J: Käyttäminen (.) ol-i ↑no (1) tosi hauska ja minä Using (.) be-PST.3SG well (1) very funny and 1SG aina (.) sano-i-n vitse-j-ä hän-en näköinen always (.) say-PST-1SG joke-PL-PTV 3SG-GEN look:a:like on myös hauska. be.3SG also funny. Well the using was very funny. And I always told jokes because he is also funny. Koska hän ei ↑ymmärrä. Because 3SG be.3SG understand. Because he does not understand. Joo °ei°. Yeah, no. In this example, E uses a conjunction to recapitulate the preceding conversation, which was side-tracked due to the word-search sequence. This type of interpretation can shift the focus on the previous turns. However, they need not be solely backward-oriented (such as corrections). Since the speaker uses her own words to describe what she thought the other speaker was saying, she often introduces new elements into the conversation. Thus, rather than merely modifying the prior, the articulations take a step forward from the problematic situation by presenting a potential reading of the trouble turn. (Kurhila 2006: 153.) In this particular example, the conversation became side-tracked, but the participants were able to formulate a potential interpretation of what was being said. From the perspective of conversation and what has been evident in the previous examples is that a candidate understanding has a specific place in the conversation and it is followed by a certain action. Typically, “a candidate understanding functions as the first pair part of an adjacency pair, since it projects a confirmation as the next turn” (Kurhila 2006: 155). In the example that follows, participant E is speaking about her ongoing thesis and the theoretical framework. In line I, she describes her work as ispol’zovanie russkogo jazyka na praktike kak finny, ‘the use of Russian language in practice how Finns’. In line 2, M reformulates the previous turns of speaker E and produces the candidate understanding of deti meždu soboj kak, ‘the children together how’. This functions as a first-pair part of an adjacency pair and it projects a confirmation in the next turn. Speaker E produces the second-pair part as a minimal response da, ‘yes’. 165 7 Candidate understandings – somewhere in the middle? (78) UM 43. <01:33:09> 01 E: 02 М: → 03 E: Nu da is↑pol’zovanie rus-sk-ogo jazyk-a Well yes use Russian-ADJ-GEN language-GEN na praktik-e kak ↑finn-y, PRE practice-PREP how Finn-PL Well yes the use of Russian language in practice how Finns, Det-i meždu so↓b-oj kak. Child-PL PRE self-INST how Children with each other how. Da. Yes. While candidate understandings might somehow affect the progressivity of the ongoing conversation, this does not necessarily occur in all cases. Example 79 is a rather long excerpt from a conversation in which the participants are discussing the language policies in different countries. Speaker J has a question concerning the relationship between the Finnish and Carelian languages. She wants know whether the Carelian language is considered to be a dialect of Finnish. At this point, speakers A and M begin to define how those two languages can be divided or connected. In line 9, A describes that there are two different aspects to consider and the first is the Carelian dialect. Then, speaker M wants to clarify the situation and begins his own description in line 10. He states that there are two separate things. There is a Carelian language, which is a part of Finnish language and then there is a separate Carelian language. In line 14, M states that the Carelian language in Finland has almost disappeared. However, the Carelian dialect of Finnish is spoken in Eastern Finland. In line 17, A confirms this and adds that in that part of the country everyone uses this dialect. The speaker therefore demonstrates that he is a knowing participant and that he can make assessments (Pomerantz 1984: 57). (79) UM 40. <01:50:49> 01 J: 02 ?: 03 J: 04 ?: Da i možno od-na želanie >od-no< želanie. Yes and can one-FEM wish one-NEU wish Yes and can one wish. Ugu. Uhhuh. U menja (2) no èt-o ne kasa-et-sja PRE 1SG.GEN (2) but 3SG-NEU NEG concern-3SG-PRT rus-- rus-sk-- rus-sk-ogo jazyk-a. Rus-- Russi-- Russian-ADJ-GEN language-GEN I have but it does not concern Russian language. °Ugu.° Uhhuh. 166 7 Candidate understandings – somewhere in the middle? 05 J: 06 07 08 09 N: 10 М: 11 N: 12 М: 13 J: 14 М: 15 J: 16 М: 17 A: → Nu è: esli ja hote-l-a by iz-- izuča-t’ Well VOC if 1SG want-PST-FEM PRT st-- study-INF ka↑rel’-sk-ij jazyk nu nu a: v: Finljandi-i Carelian-ADJ-MASC language well well VOC PRE Finland-PREP è è nu točn-o °ne zna-ju°. VOC well precisely-ADV NEG know-1SG Well, if I wanted to study the Carelian language well in Finland, well, precisely, I do not know. V Finljandi-i karel’-sk-ij jazyk duma-et kak PRE Finland-PREP Carelian-ADJ-MASC language think-3SG like dialekt >dialekt< ne jazyk. dialect dialect NEG language. In Finland, the Carelian language is considered to be a dialect, not a language. (1) ((simultaneous speech)) Èt-o dve vešč-i èt-o karel’-sk-ij dia↑lekt i 3SG-NEU two thing-PL 3SG-NEU Carelian-ADJ-MASC dialect and It is two thing--it is the Carelian dialect and Est’ karel’-sk-ij jazyk ↑fin-sk-ogo Be.3SG Carelian-ADJ-MASC language Finnish-ADJ-GEN.MASC jazyk-a i est’ ka↓rel’sk-ij jazyk. language-GEN and be.3SG Carelian-ADJ-MASC language There is the Carelian language of the Finnish language and the Carelian language. T-ot ↑i drug-oj est’. One-MASC and another-MASC be.3SG One and another is. T-ot i drug-oj est’. One-MASC and another-MASC be.3SG One and another is. Po-èt-omu ja hoč-u uzna-t’ kak ↑tret-’ego PRE-IT-DAT 1SG want-1SG know-INF how three-GEN.NEU lic-a èt-o položenie karel’-sk-ogo party-GEN this-NEU position Carelian-ADJ-GEN.MASC jazyk-a v Finljandi-i i, language-GEN PRE Finland-PREP and That is why I want to know as a third party this position of the Carelian language in Finland and, V Finljandi-i počti net karel’-sk-ogo jazyk-a. PRE Finland-PREP almost NEG Carelian-ADJ-GEN.MASC languageGEN In Finland, there is almost no Carelian language. °Ugu.° Uhhuh. Est’ dia↑lekt fin-sk-ogo jazyk-a tam Be.3SG dialect Finnish-ADJ-GEN.MASC language-GEN there na vostok-e. PRE east-PREP There is a dialect of the Finnish language in the east. I tam vs-e govor-jat na èt-om dialekt-e. And there all-PL speak-2PL PRE this-PREP dialect-PREP And there everyone speaks in this dialect. 167 7 Candidate understandings – somewhere in the middle? 18 М: °Dialekt-e da° èt-o vse-taki ↑fin-sk-ij jazyk. Dialect-PREP yes 3SG-NEU however Finnish-ADJ-MASC language Dialect, yes it is, however, the Finnish language. The next excerpt is an example of the continuations that comprise a longer unit. The participants are discussing a prominent local man who knew many languages and who was one of the founders of the literary language. In line 4, H states that that man graduated from a madrasa and knew Arabic, Russian, and Azerbaijani. In line 5, M summarises the previous turn by saying vse jazyki on znal mnogo jazyki, ‘All the languages he knew many languages’. However, M’s turn is rather overwhelming in its grandiosity, but it also reflects how much he respects the man. In contrast, H’s second-pair part in line 6 is more subtle, but it also indicates his respect as he states naši dagenstanskie jazyki znal, ‘our Dagestani languages knew’. (80) DM 48. <24:17> 01 М: 02 H: 03 N: 04 H: 05 06 М: → 07 H: Da da ↑hiv-sk-ij on. Yes yes Hiv-ADJ-MASC 3SG Yes he is from HIV. S Hiva tam tam ↑hiv-sk-ij čelovek. PRE Hiv there there Hiv-ADJ-MASC man From HIV there there HIV man. A:. VOC Da: (.) on okonči-l-ø medrese. Yes (.) 3SG graduate-PST-MASC madrasa Yes he graduated from a madrasa. On è: zna-l-ø èto-t a↑rab-sk-ij jazyk on 3SG VOC know-PST-MASC this-MASC Arab-ADJ-MASC language 3SG zna-l-ø èto-t nu: ja ja i govor-ju know-PST-MASC this-MASC well 1SG 1SG and talk-1SG ↑rus-sk-ij jazyk zna-l-ø Russian-ADJ-MASC language know-PST-MASC azerbaj↑džan-sk-ij zna-l-ø, Azerbaidjan-ADJ-MASC know-PST-MASC He knew this Arab language, he knew this well I say he knew the Russian language and knew the Azeri language, Vs-e jazyk-i on zna-l-ø mnogo jazyk-i. All-PL language-PL 3SG know-PST-MASC many language-PL All languages, he knew many languages. Naš-i da- dagenstan-sk-ie ja↓zyk-i zna-l-ø. 1PL.POSS-PL da- Dagestani-ADJ-PL language-PL know-PST-MASC Our Dagestanian languages he knew. In the next two examples, the recipient rephrases what was previously stated and begins his turn with the double affirmative particle da da, ‘yes yes’. 168 7 Candidate understandings – somewhere in the middle? (81) IS 64. <37:01> 01 H: 02 J: → 03 H: No bol’-- bol’šinstvo ↓ljud-i kotor-ye ja zna-ju Well maj-- majority people-PL who-PL 1SG know-1SG >voobšče< otnos-jat-sja k è: k obučeni-ju v in:general relate-3PL-PRT PRE VOC PRE learning-DAT PRE universitet-ø kak by è: formal’nost-i university-ACC how PRT VOC formality-GEN kak-oj-to. something-GEN.FEM-CLI Well the majority of people whom I know in general have an opinion that studying at the university is some kind of formality. Da: da nado čto-to ↑dela-t’. Yes yes have.to something-CLI do-INF Yes, have to do something. Da nado čto-to dela-t’ ja ne mog-u Yes have.to something-CLI do-INF 1SG NEG can-1SG side-t’ doma. sit-INF at.home Yes, have to do something--I cannot just sit at home. (82) IS 65. <13:43> 01 H: 02 03 04 05 J: H: 06 H: 07 J: → 08 H: Daže esli on ne ume-et. Even if 3SG NEG can-3SG Even if he cannot. On tebe skaž-et @da ja ume-ju@. 3SG 2SG.DAT say-3SG yes 1SG can-1SG He will say to you “yes I can”. @Da ja vsë ume-ju@ Yes 1SG everything-NEU can-1SG Yes I can do everything. ((laughter)) I potom on (-) s ošibka u ni-h ↑ah. And then 3SG (-) PRE mistake PRE 3PL-GEN VOC And then he they make a mistake. (1) (-) No on k tebe voobšče ne: sovet (-) But 3SG PRE 2SG.DAT in:general NEG advice ne-prost-o. NEG-simple-ADV Well he you in general the advice is not simple. Da da sam luč-še zna-et. Yes yes self good-COMP know-3SG Yes yes he himself knows better. Ugu. Uhhuh. As we have witnessed in the previous examples, candidate understandings do not necessarily halt the progression of the conversation. Instead, this entails a potential interpretation of the previous turns. It also projects a confirmation as the next relevant turn and therefore functions as a 169 7 Candidate understandings – somewhere in the middle? first-pair part. The continuation can be a response particle, or it can consist of a longer unit. 7.2 Noun phrases An examination of the collaborative productions revealed that the structure of those productions was analysed in terms of preliminary and final components. The preliminary components were produced by the first speaker and the final component by the second speaker. The completion was typically undertaken by producing one word, and the preliminary component projected the form of it. In other words, the offered component was adjusted prosodically, syntactically, and interactionally to the preliminary component. There are also cases somewhat similar to those collaborative productions, but they nevertheless display some differences. In the following example, S is a Spanish woman who is translating a public speech into Finnish. The point of the speech is that there are different immigrant groups in Finland who work in the mass media. At the end of her turn, S describes that it is important that they do that work but she does not get to end of her turn before speaker G gets involved and produces a possible reading of the previous turn. There is a micropause in S’s turn which makes the completion possible. They jointly produce a structure maahanmuuttaja tarvitsee omaa ääni, ‘an immigrant needs his own voice’. However, the completion of ääni is in the nominative case although the previous construction would inquire a completion in partitive case ään-tä, ‘voice-PTV’. Speaker S accepts the completion and she repeats it. (83) WN 2. <34:58> 01 S: 02 03 G: → Mutta nykyään hän ↑sano-o että totta kai että But nowadays 3SG say-3SG that of.course that ulkomaalaise-t teke-e tärkeä-ä (.) ↑tieto e: >työ< foreigner-PL do-3SG important-PTV (.) knowledge VOC work massmedia-n kanssa, massmedia-GEN with But nowadays she says that of course foreigners do important knowledge oh work with massmedia. Koska varma että maahanmuuttaj-i-a ↑tarvitse-e oma-a [(.) Because sure that immigrant-PL-PTV need-3SG own-PRT Well sure that immigrants need their own [Ääni. Voice. 170 7 Candidate understandings – somewhere in the middle? 04 S: Ääni joo sitten Internet on ↑hyvä tie missä me voi-mme Voice yeah then Internet be.3SG good way where 1PL can-1PL laitta-a meidä-n ↑ääni ja sitten me voi-mme löytä-ä put-INF 1PL-GEN voice and then 1PL can-1PL find-INF <erilais-i-a ihmis-i-ä> kene-n kanssa me voi-mme different-PL-PTV people-PL-PTV who-GEN with 1PL can-1PL teh-dä verkko-j-a. do-INF net-PL-PTV Yes voice. Then the Internet is a good way where we can use our voice and then we can find different people with whom we can network. Apart from the completion not being in the correct case, it is not solely backward-oriented, but rather could have been interpreted as adapting to the previous context. They are discussing the immigrants in Finland and during her turn, S has discussed the mass media. In order to attain more positive publicity and succeed in getting their own issues recognized in the public debate, the immigrants need access to those media that could enable it. This projects the word ääni, ‘voice’, which is a metaphor that is used in different contexts. Speaker S then formulates the concept further and states that the Internet would be a good forum for the immigrants to get their voice heard. The Internet would also make it possible for immigrants to network. Example 84 is similar in the sense that the candidate understanding is projected and is not backward-oriented. This example also focusses on the immigrants who live in Finland, but from a different perspective. The participants are discussing closing the seminars of different nongovernmental organisations. In line 3, speaker D states that perhaps it would not be a good idea to pool the resources of different organisations because seminars might become confusing. In the middle of her turn, she expresses her opinion by producing the following negative interrogative: eikö me mennä, ‘don’t we go’, (Heritage 2002). (84) PM 6. <40:55> 01 02 03 D: Se on mu-n oma mielipide. 3SG be.3SG 1SG-GEN own opinion. That is my own opinion. syyt-- yhdistet-ään sii-hen teidä-n oma seminaari prin-- connect-4 3SG-ILL 2PL-GEN own seminar ei-kö me men-nä vähän pois sii-tä meiä-n teema-sta. NEG.3SG-Q 1PL go-INF little away it-ELA 1PL-GEN theme-ELA But if we connect that to your own seminar do not we go a little far away from our theme. 171 7 Candidate understandings – somewhere in the middle? 04 F: 05 D: 06 F: → 07 F: 08 D: → 09 F: Ei: ei (-) koska se on se .h ö <päätavoite> että NEG NEG (-) because 3SG be.3SG 3SG .h VOC main:goal that me-kin puhut-aan vapaaehtoisuut-ta ja maahanmuuttaja että 1PL-CLI talk-4 voluntary-PTV and immigrant that miten se vapaaehtoisuus vo-isi palvel-la (.) tai how 3SG voluntary can-CON.3SG serve-INF (.) or palvel-la vai m(h)iten [sano-a se-n että .h serve-INF or how say-INF 3SG-GEN that .h No no because it is a main goal that we also talk about volunteering and that immigrant that how that volunteering could serve or serve or how to say it that [Joo. Yeah. miten maahanmuuttaja vo-isi (.) ä miten se ↑autta how immigrant can-CON.3SG (.) VOC how 3SG help.3SG maahanmuuttaja kotoutumise-en. immigrant integration-ILL How an immigrant could how it could help an immigrant to integrate. Että: se on se, That 3SG be.3SG 3SG That it is that, Sii-hen ↓toiminto-on. 3SG-ILL action-ILL To that action. Joo kyllä. Yeah yes. In the next example, the participants are discussing the school system of the Tajik schools. Speaker N is inquiring about exams and how they are administered in schools. In line 5, speaker E begins to count the different subjects that the pupils are required to take exams in: istorija, ‘history’, and jazyk, ‘language’. She also produced the word raznye, ‘different’, twice during her turn. Then in line 6, N formulates an umbrella concept that would account for all these different subjects and she produces the phrase obschie znanija kak by, ‘general knowledge sort of’. This utterance functions as a first-pair part and it is confirmed in line 7 as a second-pair part. (85) TI 12. <16:17> 01 H: 02 E: Aga i ↑èto-t èkzamen by-l-ø (.) .h Aha and 3SG-MASC test be-PST-MASC (.) .h [nado by-l-o t-- èto-t ↑pis’menn-yj have.to be-PST-NEU VOC 3SG-MASC written-MASC And this test, it was in a written form? é VOC ékzamen [a:? test VOC [Nas, 1PL.GEN Our, 172 7 Candidate understandings – somewhere in the middle? 03 04 05 E: 06 N: → 07 E: [Pis’menn-yj by-l-ø da vs-ë èt-o na (-) Written-MASC be-PST-MASC yes all-NEU this-NEU PRE (-) èto-t (1) °kak vam skaza-t’°. 3SG-MASC (1) how 2PL.DAT say-INF There was in a written form. All was this how to say it to you. (2) Est’ na pjat-om (-) nu razn-ye èt-i (.) Be.3SG PRE five-ORD.PREP.MASC (-) well different-PL this-PL (.) nu istorija (-) jazyk èt-o nu ra-well history (-) language 3SG-NEU well di-↑razn-ye èt-o by-l-ø, different-PL 3SG-NEU be-PST-MASC In the fifth grade there are well these different well history, language it is well different it was, ↑Obšč-ie zna[ni-ja kak by. General-PL knowledge-PL how PRT Like general knowledge. [Obšč-ie znani-ja da:. General-PL knowledge-PL yes Like general knowledge. (86) TI 35. <16:13> 01 E: 02 03 04 05 N: 06 E: No potom e↑ščë u nas by-l-i my uči-l-i-s’ Well then yet PRE 1PL.GEN be-PST-PL 1PL study-PST-PL-REFL do: odi:nnadcat-ogo dvenadcat-ogo čas-ov. PRE eleven-ORD.GEN.MASC twelve-ORD.GEN.MASC hour-PL.GEN Well then we had we studied until eleven or twelve. Potom kto hote-l-ø (.) é >kto kak-ie roditel-i< Then who want-PST-MASC (.) VOC who what-PL parent-PL ho↑te-l-i .h posle obed-a u nas want-PST-PL .h after lunch-GEN PRE 1PL.GEN prodlënka by-l-ø. .h continuation be-PST-MASC Then who wanted, or parents wanted, after lunch we had a continuation. Pro↑dlënka èt-o .h prodolženie urok-a. Continuation 3SG-NEU .h continuation lesson-GEN Continuation it is a continuation of a lesson. My [tam dela-l-i 1PL there do-PST-PL We did there [A:. VOC »domašn-ye zada:ni-ja my tam uči:-l-is’ home-PL work-PL 1PL there study-PST-PL my tam risova:-l-i [vs-ë vs-ë vs-ë. 1PL there draw-PST-PL all-NEU all-NEU all-NEU homework we drew there. 173 7 Candidate understandings – somewhere in the middle? 07 N: 08 E: 09 N: → 10 E: 11 N: [A vy s èt-imi s VOC 2PL PRE this-INST PRE (Musaraf) i (Musafa[r-om), (Musaraf) and Musafar-INST You were with these Musaraf and Musaraf, [Da: u nas taka-ja v- é: èt-o Yes PRE 1SG.GEN such-FEM VOC this-NEU by-l-o čto [my(-), be-PST-NEU that 1PL (-) Yes we had that kind of. It was that we, [Dopolnitel’n-ye. Extra-PL Extra. Dopolnitel’n-ye [urok-i by-l-i da. Extra-PL lesson-PL be-PST-PL yes Extra lessons were, yes. [Aga. Aha. 7.3 Summary Compared to collaborative productions and word searches, candidate understandings stand somewhere in the middle. When compared to collaborative productions, candidate understandings do not need to be completely backward-oriented, as the recipient can use them to rephrase the content of the previous turns and to introduce new elements to the conversation. Candidate understandings are also projected, but not as strictly as the collaborative productions. Interactionally, these understandings are used to check the recipient’s interpretation of the prior and to maintain mutual understanding. However, they are not prosodically or syntactically adapted to the previous utterance. They can be also preceded by hesitations or by other perturbations signs. When compared to word searches, candidate understandings exhibit some similarities. A candidate understanding functions as the first pair part of an adjacency pair, which means that it projects a confirmation as the next turn. Moreover, after a candidate understanding has been produced, the first speaker can evaluate its appropriateness. This can be undertaken either by repeating the offered understanding, or by producing a particle. Whereas there were few instances of not taking a stance on the produced understanding, when the participants moved on directly to a new topic, this was possible. It is interesting that there were proportionally more candidate understandings in the Russian than in the Finnish data. There could be various explanations with one being that the participants in the Russian 174 7 Candidate understandings – somewhere in the middle? conversations were more competent in the spoken lingua franca. That means that they have more opportunities to reformulate or paraphrase previous turns. Conversely, if a participant knows only one way of expressing something, he or she will adhere to that pattern only. However, if two or more participants are approximately at the same level, they can orient to linguistic structures and are therefore able to produce them collaboratively. The word-search sequences reveal that the participants were willing to help each other. 175 8 Conclusion and implications for further study In this chapter, I will first reiterate the central findings of this study (subchapter 8.1). The second subchapter (8.2) is devoted to presenting the possible methods of conducting lingua franca research concerning Finnish and Russian languages in the future. Future themes for research are based on features that are characteristic of non-native interaction on the basis of the data used in this study. Those features have not been systematically observed or analysed, but they have arisen otherwise from the data and they need to be analysed more thoroughly. 8.1 Collaborative productions and word searches This study concentrated on jointly produced structures by two or more participants in lingua franca interaction. The lingua francas were Finnish and Russian. However, few studies have been conducted on these languages as lingua francas. Thus, this study contributes to the research tradition of these languages. While the Finnish language is a member of the Finnic group of the Uralic family of languages, Russian belongs to the Slavic branch of the Indo-European languages. Although Finnish and Russian belong to different language groups, they share some characteristics (for instance, the absence of the article and comparatively free word order) and these make them more similar in some respects than some languages within the Indo-European family. In this respect, they also differ from the lingua franca that is used the most worldwide, English. More specifically, the focus of this analysis was primarily on word searches and collaborative productions. These categories were not defined in advance, but they arose instead from the data. This suggests that the actual choice of research theme was made only after some recordings were had been collected. This type of research is based on a methodology called conversation analysis (abbreviated CA; Heritage 1996 [1984], Psathas 1995, Sacks et al. 1974). In other words, the main objective for collecting the data was to obtain spoken Finnish and Russian lingua franca material, whereas the actual choice of theme for the research was made when some of the material had already been collected. Thus, the notion of ‘lingua franca interactions’ was taken as an analytical construct (Firth 1996: 241, Schegloff 1991). This type of CA-oriented research leaves the decision-making to the participants. The result is that the researcher does not need to make the 176 8 Conclusions and implications for further study hypothetical classifications, as he or she can rely on the participants’ actual linguistic behaviour. (Kurhila 2006: 32.) This also means that all the details of interaction can be important, and nothing in the analysis can be omitted as irrelevant or accidental (Heritage 1996 [1984]: 237–238, Psathas 1995: 2, Schegloff 1997: 502). Another methodological tool for conducting research was interactional linguistics. One of the goals of interactional linguistics is to obtain a better understanding of the types of influences that languages and interaction have on each other (Couper-Kuhlen & Selting 2001: 3). When studying typologically different languages such as Finnish and Russian, interactional linguistics research attempts to demonstrate that different languages might use different linguistic resources to address the common problems that arise in interaction (cf. ibid. 8). Word searches are usually attributed to part of the linguistic behaviour of language users who are not yet competent in a foreign language. However, participants in conversations can also become engaged in other activities that consist of two or more participants, such as collaborative productions. These two activities have been studied as constituting very similar phenomena, but certain characteristics differentiate collaborative productions from word searches. In the data, there were prototypical cases of both of the above-mentioned structures, i.e. word searches and collaborative productions, but different types of boundary cases also occurred. A prototypical collaborative production consisted of a preliminary part that was produced without any specific hesitations or perturbations and the final component that fitted syntactically, prosodically and semantically to the preliminary component. For example, a preliminary part could consist of a subject + verb combination as well as a final component that is a complement (in a broad sense), or the line between components could be situated between a subject and a verb. The latter cases were in contrast to previous research results. For example, according to Helasvuo (2003), it is more likely “to have an intonation unit boundary between the verb and the object than between the subject and the verb.” This seems to imply that concerning the most common grouping of constituents would be the subject and verb rather than the verb and. (see also Helasvuo 2004: 1332.) Although word order is often characterised as being relatively free in both Finnish and Russian, in Finnish spoken discourse data, a clear preference exists for subject – verb ordering (Helasvuo 2001). Thus, with respect to colloquial 177 8 Conclusions and implications for further study spoken Finnish, the ordering of the elements was the same, but the elements were grouped differently. A prototypical word search was initiated with pauses, hesitations or with specific questions. In these cases, the co-participant was considered to be a competent partner. The initiation affected the ongoing conversation; it caused perturbations and finding the correct word could have become what the participants were actually doing. In general, participants did not exhibit problems with linguistic structures in the sense, for example, so that they would have not known how to decline a noun, but the problem was that they did not know the noun they wanted to decline. Thus, the number of Finnish or Russian case endings or verbal morphology was not necessarily the factor that created problems. With respect to both collaborative productions and word searches, the initiator of the word search or the speaker who produced the preliminary component had an opportunity to evaluate the offered word or the final component that had been produced by the recipient. The initiator could accept them by repeating them, by producing a discourse particle or by moving on in his or her speech. It is important to note that overtly expressed rejections were rare. For example, if the first offered word was not the one that the first speaker had in mind, he or she would continue the search with the co-participant. The boundary cases consisted of those instances in which participants were co-constructing such grammatical information as the verb + object constructions. In those cases, the participants could negotiate the appropriate form. When comparing collaborative production and word searches, the candidate understandings were somewhere in the. It is interesting that in my data, proportionally more candidate understandings occurred in the Russian data than in the Finnish data. There could be a variety of explanations to account for this difference. One could be that the participants in the Russian conversations were more competent in their spoken lingua franca. This means that they had more possibilities to reformulate or to paraphrase previous turns. And the reverse was also possible: if a participant knew only one way to state something, he or she could continue with that pattern and only use that repeatedly. This also means that once the recognisability is established, the production of jointly produced structures becomes easier. Interactionally, candidate understandings were used for checking the recipient’s interpretation of the prior and for maintaining mutual understanding. When these understandings are compared to collaborative 178 8 Conclusions and implications for further study productions, understandings needed not be completely backward-oriented, but by using them, the recipient could have rephrased the content of the previous turns and introduce new elements into the conversation. They were also projected, but not as strictly as in the collaborative productions and they were not prosodically or syntactically adapted to the previous utterance. When compared to word searches, they shared some similarities. They could have been preceded with hesitations or other perturbations signs. From the perspective of sequentiality, a candidate understanding functioned as a first-pair part of an adjacency pair, which means that it projected a confirmation as the next turn. After a candidate understanding was produced, the first speaker could evaluate its appropriateness so that he or she had the same rights either as the initiator of the word-search sequence, or as the first speaker of the collaborative production. This could be achieved either by repeating the offered understanding, or by producing a discourse particle. Participants could also move straight to a new topic, but this rarely occurred in the data. All three ways emphasized that the participants wanted to cooperate with each other and that although the language used was foreign to all participants, the interaction need not to have been problematic. The situation was actually the opposite; it seemed that the non-native speakers could use language in a creative way. However, this did not mean that, for example, grammatical rules were neglected. The participants also searched for grammatical forms and were willing to help each other. As was evident in the examples, different types of collaborative productions could occur in one short extract. On the one hand, this meant that the extract must have been analysed thoroughly and meticulously, but on the other hand, this was evidence that the conversational participants were continuously following the conversation. In NNS interaction, no one can be said to be more knowledgeable than other participants. Although in most cases, the distinction could have been made as to whether the search was of a lexical item or whether it involved some grammatical issue, this distinction might not have been relevant. However, it was striking that there were grammatical searches involving both the current speaker and the recipient only in the Finnish data. When considering the grammatical structures of both languages, it is evident that both languages have the same type of overall structure. Traditionally thinking, both are Subject + Verb + Object (SVO) languages and they both possess a moderate number of morphological categories in declining nouns 179 8 Conclusions and implications for further study or in verbal morphology. In this sense, grammatical searches are possible in different positions during utterance formation. Intuitively, grammatical searches may appear to be an essential part of interaction in both languages, but only the participants in the Finnish data relied on other participants when they hesitated on grammatical issues. From a cross-linguistic perspective, the difference was also interesting, as it shows that lingua franca users were able to mobilize the resources of languages other than their mother tongues. This also contributes to the research in interactional linguistics research because it confirms that different languages indeed have different means to solve the problems that arise in interaction. The difference might be explained by the different environments in which languages have been learnt. Considering the Finnish language, the lingua franca speakers were usually adults who had learnt Finnish in different language courses, a factor that could have contributed to their orienting more to grammatical issues. Most of the speakers had moved to Finland and thus Finnish language had not been a part of their life before. The lingua franca speakers of Russian had had different histories. In a certain sense, the Russian material was versatile because it consisted of the data of both those who had learnt Russian in their adult life and those who had had Russian language in their lives since childhood. Furthermore, historical circumstances affected the differences between these languages. For example, the lingua franca speakers of Russian might have established a more native-speaker relationship to the Russian language, which means that they did not consider the grammatical appropriateness to be important as the lingua franca users of Finnish language. (See, e.g. Kurhila 2006: 132–133.) As mentioned previously in this analysis, when concerning the Finnish data, there were some cases arose in the Finnish data in which the grammatical appropriateness surfaced in the conversational so that it that caused disturbances in the ongoing conversation. In grammatical searches, the participants could have searched together for the correct form and decide on it by cooperating even though the form might have been ungrammatical in the native language. In these cases, the grammar could have been interpreted as real-time phenomenon (Hopper 1987: 141), which means that syntactic structures unfolded in real time (Auer & Pfänder 2011: 3, Hopper 2011: 27). Emergence of clauses as syntactic units were achieved through mutual activity by the participants (Helasvuo 2001b: 33–35). I have emphasised the cooperative nature of lingua franca interaction and have not made observations on perhaps one of the obvious research areas: 180 8 Conclusions and implications for further study communicative failures or miscommunication (for example, see Firth 1996, House 1999, and Mauranen 2006b). In some of the cases concerning prolonged search sequences or unsolved cases, a communicative failure was possible. However, the number of those instances was not high. This might be explained by the overall behaviour of the participants. They might be eager to monitor the co-participant’s speech and to attempt to avoid problems beforehand. The cooperative nature of conversations also suggests that the participants were willing to act together in order to ensure a smooth conversational flow. Even though the language that the participants used was not their mother tongue, they were able to cooperate and avoid misunderstandings. In this sense, the results of this study are in line with Russian scholars who have reported that miscommunication or communicative failures are as common in interaction between friends and family members as they are in intercultural interaction (Ermakova & Zemskaja 1993). 8.2 Further research One of the obvious future research areas is the actual language form that is used in lingua franca conversations. Although the focus of this study was not primarily on the actual language form, some hypotheses can be drawn on the basis of the data presented in this study. While different types of word-search sequences occur in the data, the nominative case or the infinitive form of the verb dominated. These are referred to as the basic forms, or the dictionary forms. There are three explanations for the occurrence of these basic forms. First, the sequence may have been initiated so that it required this type of basic form only, and other possibilities being that either the offered word was in the nominative case, or the verb was in the infinitive. After the offer, the initiator of the sequence had an opportunity to decline the noun in the correct case. For example, in the Finnish examples, the question mikä se on, ‘what is it’, usually indicates that the searched word is a noun and that it can be produced in the nominative case as a solution to the search. With special reference to the Russian language, certain categories might undergo changes in Russian lingua franca interaction. Firstly, the grammatical form of the noun fluctuates after a cardinal number. Another category that is liable to change is gender. For example, the third-person pronoun, which usually reflects the referent’s gender, might be used differently. When discussing the semantics of the language, one tendency in 181 8 Conclusions and implications for further study lingua franca interaction is to use of semantically close words and the participants can even use two semantically close verbs. While from a native speaker’s perspective this use of the verb might be inappropriate, the nonnative speakers accept the usage without hesitation. One of the interesting tendencies in lingua franca interaction is the connection between lexico-grammatical correctness and communicative effectiveness (for example, see Hülmbauer 2009). The utterances that the interlocutors produced were typically lexico-grammatically appropriate (either formulated through individual or joint effort), but they might have sounded rather odd. In other words, a native speaker would probably not use those particular constructions even though they are grammatically suitable and no problems were encountered, such as in the choice of nouns. As pointed out in the introduction, both Finnish and Russian differ structurally from the lingua franca that is most used, English. In terms of morphological typology, English as a lingua franca is analytic, which is in line with the linguistic structure of Standard English. Furthermore, the different linguistic backgrounds of the speakers do not necessarily imply that the structure of the used lingua franca would change. Rather, it seems that the speakers are able to mobilize the different resources of the spoken lingua franca, but they differ in their usage from native speakers. When participants utilise resources, they are also negotiate the language use (Cogo 2009, Jenkins 2003). Those resources include the controversial use of the third-person singular –s, and the extended use of the progressive –ing verb forms. Speakers also tend to insert redundant prepositions and overuse certain words that have high semantic generality (do, have, make). (Björkman 2009; Cogo & Dewey 2006; Dewey 2009; Ranta 2009; Seidlhofer 2004, 2011.) In terms of morphological typology, Finnish and Russian resemble synthetic languages. However, the same observation is also valid for English as a lingua franca. The overall structure of the languages that are used remains the same. In other words, the participants were able to use the spoken lingua franca and its resources in speech production. In some instances, this was accomplished through joint activity when the participants formulated utterances together in order to attain mutual understanding. However, additional data are needed to ensure that the observation on the synthetic language type is valid. As I have mentioned in the introduction, the historical backgrounds are different for the Finnish and Russian languages as lingua francas. Finnish as 182 8 Conclusions and implications for further study a lingua franca has not had a long history. In contrast, the Russian language reached its pinnacle as a lingua franca during the Soviet era, but the language lost its importance after the dissolution of the Soviet Union. However, the future may hold same progression for both languages, but for different reasons. The number of those who use Finnish language as a foreign language is rising due to the rising number of immigrants. The Russian language may regain its status as a language of interethnic communication both in Russia and in what is referred to as the Near Foreign countries. Concerning both Finnish and Russian, bilingual and multilingual speakers also pose challenges to language education. 183 References AG-80 = N. Ju. Švedova, N. D. Arutjunova, A. V. Bondarko, B. B. Ivanov, V. V. Lopatin, I. S. Uluhanov, F. P. Filin: Russkaja grammatika I & II. Akademia nauk SSSR: Institut russkogo jazyka. [Available at: http://rusgram.narod.ru/.] ALPATOV, V. M. 2003. Russkij jazyk v sovremennom mire. Rešenie nacional’nojazykovyh voprosov v sovremennom mire. Saint-Petersburg: Zlatoust. pp. 426–430. ANDREEVA, S. V. 2007. Razgovornaja russkaja reč’ i eë edinicy. Analiz razgovornoj russkoj reči, pp. 37–42. AREF'EV A. L. 2012. Russkij jazyk na rubeže XX-XXI vekov. Electronic resources. Moscow: Centr social’nogo prognozirovanija i marketinga. ARNOLD, JENNIFER E., MARIA FAGNANO & MICHAEL K. TANENHAUS 2003. Disfluencies signal theee, um, new information. Journal of Psycholinguistic Research 32 (1), pp. 25–36. ASINOVSKIJ, A. C., N. V. BOGDANOVA, S. B. STEPANOVA, T. JU. ŠERSTINOVA, E. V. MARKASOVA & A. V SUPRUNOVA 2010. Zvukovoj korpus russkogo jazyka “Odin rečevoj den’”: puti popolnenija i intellektual’nye tehnologii. In: A. E. Kibrik, ed., Komp’juternaja lingvistika i intellektual’nye tehnologii: Po materialam ežegodnoj Meždunarodnoj konferencii “Dialog”, 9(16), pp. 41–47. Moscow: Izdatel’stvo RGGU. [available at: http://www.dialog-21.ru/digest/2010/pdf/] AUER, PETER 1996. On the prosody and syntax of turn-taking. In: Elizabeth CouperKuhlen & Margret Selting, eds., Prosody in conversation. Interactional studies, pp. 57–100. Cambridge: Cambridge University Press. AUER, PETER 1999. From codeswitching via language mixing to fused lects: toward a dynamic typology of bilingual speech. International Journal of Bilingualism 3/4, pp. 309–332. AUER, PETER 2005. Projection in interaction and projection in grammar. Text 25 (1), pp. 7–36. AUER, PETER 2009. On-line syntax: Thoughts on the temporality of spoken language. Language Sciences 31, pp. 1–13. AUER, PETER & STEFAN PFÄNDER 2011. Constructions: Emergent or emerging? In: Peter Auer & Stefan Pfänder, eds., Constructions: Emerging and emergent, pp. 1–21. Berlin: De Gruyter. AVINA, N. JU. 2000. Ob osobennnostjah russkoj reči v Litve. Izvestija an. serija literatury i jazyka, 2000, 59(3), pp. 52–57. AVINA, N. JU. 2002. Nekotorye semantičeskie izmenenija v jazyke russkih Litvy. Žmogus ir žodis. Vilnius 2002, 4:3, pp. 1–6. BARTH-WEINGARTEN, DAGMAR & ELIZABETH COUPER-KUHLEN 2011. Action, prosody and emergent constructions: The case of and. In: Peter Auer & Stefan Pfänder, eds., Constructions: emerging and emergent, pp. 262–292. Berlin: De Gruyter. BAUMGARTEN, NICOLA & JULIANE HOUSE 2010. I think and I don’t know in English as lingua franca and native English discourse. Journal of Pragmatics 42, pp. 1184–1200. BAZZANELLA, CARLA & ROSSANA DAMIANO 1999. The interactional handling of misunderstanding in everyday conversations. Journal of Pragmatics 31, pp. 817–836. BJÖRKMAN, BEYZA 2009. From code to discourse in spoken ELF. In: Anna Mauranen & Elina Ranta, eds., English as a Lingua Franca: Studies and Findings, pp. 225–251. Newcastle upon Tyne: Cambridge Scholars. 184 References BOGDANOVA, N. V., S. B. STEPANOVA & T. JU. ŠERSTINOVA 2010. Zvukovoj korpus russkogo jazyka “Odin rečevoj den’” kak sposob lingvističeskogo monitoringa reči regiona. In: Integracionnye processy v kommunikativnom prostranstve regionov, pp. 50–54. Volgograd: Izdatel’stvo VGU. [available at: http://lingva.volsu.ru/doc/conf/sbornik-12-14-april-2010.pdf] BOLDEN, GALINA 2004. The quote and beyond: Defining boundaries of reported speech in conversational Russian. Journal of Pragmatics 36 (6), pp. 1071–1118. BONDARKO. LIYA 2009. Short description of Russian sound system. In: Viola de Silva & Riikka Ullakonoja, eds., Phonetics of Russian and Finnish.General description of phonetic systems. Experimental studies on spontaneous and read-aloud speech, pp. 23–35. Frankfurt am Main: Peter Lang. BREITENEDER, ANGELIKA, MARIE-LUISE PITZL, STEFAN MAJEWSKI & THERESA KLIMPFINGER 2006. VOICE Recording – methodological challenges in the compilation of a corpus of spoken ELF. Nordic Journal of English Studies 5 (2), pp. 161–187. BROUWER, CATHERINE, E. 2004. Doing pronunciation: A specific type of repair sequence. In: Rod Gardner & Johannes Wagner, eds., Second language conversations, pp. 93– 114. London, New York: Continuum. BROUWER, CATHERINE E., GITTE RASMUSSEN & JOHANNES WAGNER 2004. Embedded corrections in second language talk. In: Rod Gardner & Johannes Wagner, eds., Second language conversations, pp. 75–92. London, New York: Continuum. CANAGARAJAH, A. SURESH & ADRIAN J. WURR 2011. Multilingual communication and language acquisition: New research directions. The Reading Matrix 11(1), pp. 1–15. CHESTERMAN, ANDREW 2005. Englanti globaalina ja eurooppalaisena lingua francana. In: Marjut Johansson & Riitta Pyykkö, eds., Monikielinen Eurooppa. Kielipolitiikkaa ja käytäntöä, pp. 115–130. Helsinki: Gaudeamus. CHEVALIER, FABIENNE H. G. & REBECCA CLIFT 2008. Unfinished turns in French conversation: Projectability, syntax and action. Journal of Pragmatics 40, pp. 1731– 1732. CLAYMAN S.E. 1992. Footing in the achievement of neutrality: The case of newsinterview discourse. In: Paul Drew & John Heritage, eds., Talk at work. Interaction in institutional settings, pp. 163–198. Cambridge: Cambridge University Press. COGO, ALESSIA 2009. Accomodating difference in ELF conversations: A study of pragmatic strategies. In: Anna Mauranen & Elina Ranta, eds., English as a lingua franca: Studies and findings, pp. 254–273. Newcastle upon Tyne: Cambridge Scholars. COGO ALESSIA & MARTIN DEWEY 2006. Efficiency in ELF communication: From pragmatic motives to lexico-grammatical innovation. Nordic Journal of English Studies 5 (2), pp. 59–93. COUPER-KUHLEN, ELIZABETH & TSUYUSHI ONO 2007. 'Incrementing’ in conversation. A comparison of practices in English, German and Japanese. Pragmatics 17 (4), pp. 513–552. COUPER-KUHLEN, ELIZABETH & MARGRET SELTING 1996. Towards an interactional perspective on prosody and a prosodic perspective on interaction. In: Elizabeth Couper-Kuhlen & Margret Selting, eds., Prosody in conversation. Interactional studies, pp. 11–56. Cambridge: Cambridge University Press. COUPER-KUHLEN, ELIZABETH & MARGRET SELTING 2001. Introducing interactional linguistics. In: Margret Selting & Elizabeth Couper-Kuhlen, eds., Studies in interactional linguistics, pp. 1–22. Benjamins. 185 References COUPER-KUHLEN, ELIZABETH & BARBARA E. FORD (EDS.) 2004. Sound patterns in interaction. Amsterdam: Benjamins. COUPER-KUHLEN, ELIZABETH & MARGRET SELTING (EDS.) 1996. Prosody in conversation. Interactional studies. Cambridge: Cambridge University Press. CRYSTAL, DAVID 2003. English as a global language. Second edition. Cambridge University Press. CURL, TRACI S. 2004. Practices in other-initiated repair resolution: The phonetic differentiation of ‘repetitions’. Discourse Processes 39 (1), pp. 1–43 DANIEL, M., N. DOBRUSHINA & S. KNYAZEV 2010. Highlanders’ Russian: Case study in bilingualism and language interference in Central Daghestan. In: A. Mustajoki, E. Protassova & N. Vahtin, eds., Instrumentarium of Linguistics. Sociolinguistic Approaches to Non-Standard Russian, pp. 65–93. Helsinki. DARAGAN, JU. V. 2003. Paratizm ili simbioz: mehanizm preodolenija kommunikativnyh sboev i obsluživajuščie ego verbal’nye sredstva. In: Doklady meždunarodnoj konferencii “Dialog 2003” po kom’pjuternoj lingvistike i ee priloženijam, pp. 166– 178. Moscow: Nauka. [Available at: http://www.dialog21.ru/Archive/2003/Daragan.htm.] DARQUENNES, JEROEN & PETER NELDE 2006. German as a lingua franca. Annual Review of Applied Linguistics 26, pp. 61–77. DEWEY, MARTIN 2009. English as a lingua franca: Heightened variability and theoretical implications. In: Anna Mauranen & Elina Ranta, eds., English as a lingua franca: Studies and findings, pp. 60–83. Newcastle upon Tyne: Cambridge Scholars. D’JAČOK, M. T. 2003. Russkoe prostorečie kak sociolingvističeskoe javlenie. Gumanitarnye nauki 21, pp. 102–113. Moscow. DOETJES, D. 2007. Understanding differences in inter-Scandinavian language understanding. In: J. D. ten Thije & L. Zeevaert, eds., Receptive multilingualism, pp. 217–230, Amsterdam: John Benjamins. DREW, PAUL 1997. ‘Open’class repair initiators in response to sequential sources of troubles in conversation. Journal of Pragmatics 28, pp. 69–101. DREW, PAUL & JOHN HERITAGE 1992. Analyzing talk at work: an introduction. In: Paul Drew & John Heritage, eds., Talk at work. Interaction in institutional settings, pp. 3– 65. Cambridge: Cambridge University Press. DUBININA, I. E. & M.C. POLINSKAJA 2013. Russkogovorjaščie amerikancy: lingvističeskie portrety. Language in a global context. Moscow: INION, pp. 95-123. DUNCAN JR., STARKEY 1972. Some signals and rules for taking speaking turns in conversations. Journal of Personality and Social Psychology 23 (2), pp. 283–292. DURANTI, ALESSANDRO 1997. Linguistic anthropology. Cambridge: Cambridge University Press. DUVALLON, OUTI & SARA ROUTARINNE 2001. Parenteesi keskustelun kieliopin voimavarana. In: Mia Halonen & Sara Routarinne, eds., Keskustelunanalyysin näkymiä. Kieli 13, pp. 122–154. University of Helsinki. EGBERT, MARIA, LILO NIEBECKER & SABRINA REZZARA 2004. Inside first and second language speakers’ trouble in understanding. In: Rod Gardner & Johannes Wagner, eds., Second language conversations, pp. 178–200. London, New York: Continuum. EJZENBERG, ROSELI 2000. The juggling act of oral fluency: A psycho-sociolinguistic metaphor. In: Heidi Riggenbach, ed., Perspectives on fluency, pp. 287–314. Ann Arbor: University of Michigan Press. ERMAKOVA, O. N. 1984. Nominacii v prostorečii. In: E. A. Zemskaja & D. N. Šmelëv, eds., Gorodskoe prostorečie. Problemy izučenija, pp. 130–140.Moscow: Nauka. 186 References ERMAKOVA, O. N. & E. A. ZEMSKAJA 1993. K postroeniju tipologii kommunikativnyh neudač (na materiale estestvennogo russkogo dialoga. In: E. A. Zemskaja & D. N Šmelëv, eds., Russkij jazyk v funkcionirovanii. Kommunikativno-pragmatičeskij aspekt, pp. 30–64. Moscow: Nauka. EUROBAROMETER 243 = Europeans and their languages. Special Eurobarometer. European Commission. Available at: http://ec.europa.eu/public_opinion/archives/ebs/ebs_243_en.pdf [Accessed January 20, 2014]. FÆRCH, CLAUS & GABRIELLE GABRIELE 1983A. Plans and strategies in foreign language communication. In: Claus Færch & Gabriele Kasper, eds., Strategies in interlanguage communication, pp. 20–60. London: Longman. FÆRCH, CLAUS & GABRIELE KASPER 1983B. On identifying strategies in interlanguage production. In: Claus Færch & Gabriele Kasper, eds., Strategies in interlanguage communication, pp. 210–238. London: Longman. FEDOROVA, K. & T. GAVRILOVA 2010. Native speakers of Russian in interethnic communication: Sociolinguistic situations and linguistic strategies. In: A. Mustajoki, E. Protassova & N. Vahtin, eds., Instrumentarium of Linguistics. Sociolinguistic Approaches to Non-Standard Russian, pp. 50–64. Helsinki. FERRARA, KATHLEEN 1994. Therapeutic ways with words. Oxford University Press. FERREIRA, VICTOR S., LLOYD ROBERT SLEVC & ERIN S. ROGERS 2004. How do speakers avoid ambiguous linguistic expressions? Cognition 96, pp. 263–284. FILLMORE, CHARLES J. 1979 [2000]. On fluency. In: Heidi Riggenbach, ed. Perspectives on fluency, pp. 43–60. Ann Arbor: University of Michigan Press. FIRTH ALAN 1996. The discursive accomplishment of normality: On 'lingua franca' English and conversation analysis. Journal of Pragmatics 25 (2), pp. 237–259. FIRTH, ALAN & JOHANNES WAGNER 1997. On discourse, communication, and (some) fundamental concepts of SLA research. The Modern Language Journal 81 (3), pp. 285–300. FOLEY, JOSEPH 2006. English as a lingua franca: Singapore. International Journal of the Sociology of Language 177, pp. 51–65. FORD, CECILIA E. 2001. At the intersection of turn and sequence: negation and what comes next. In: Margret Selting & Elizabeth Couper-Kuhlen, eds., Studies in interactional linguistics, pp. 51–79. Amsterdam: John Benjamins. FORD, CECILIA E., BARBARA A. FOX & SANDRA A. THOMPSON 1996. Practices in the construction of turns: the “TCU” revisited. Pragmatics 6, 427–454. FORD, CECILIA E., BARBARA A. FOX & SANDRA A. THOMPSON 2002. Constituency and the grammar of turn increments. In: Cecilia E. Ford, Barbara A. Fox & Sandra A. Thompson, eds., The language of turn and sequence, pp. 14–38. Oxford: Oxford University Press. FORD, CECILIA E., BARBARA A. FOX & SANDRA A. THOMPSON (EDS.) 2002. The language of turn and sequence. Oxford: Oxford University Press FORD, CECILIA A. & SANDRA A. THOMPSON 1996. Interactional units in conversation: syntactic, intonational, and pragmatic resources for the management of turns. In: Elinor Ochs, Emanuel A. Schegloff & Sandra A. Thompson, eds., Interaction and grammar, pp. 134–183. Cambridge: Cambridge University Press. FOX, BARBARA 2007. Principles shaping grammatical practices: and exploration. – Discourse Studies 9 s. 299–318. FOX, BARBARA, A., MAKOTO HAYASHI & ROBERT JASPERSON 1996. Resources and repair: a cross-linguistic study of syntax and repair. In: Elinor Ochs, Emanuel A. 187 References Schegloff & Sandra A. Thompson, eds., Interaction and grammar, pp. 185–237. Cambridge: Cambridge University Press. FOX, BARBARA & ROBERT JASPERSON 1995. A syntactic exploration of repair in English conversation. In: Philip W. Davis, ed., Alternative Linguistics: Descriptive and Theoretical Modes. Current Issues in Linguistic, Theory, vol. 102, pp. 77–134. Benjamins, Amsterdam. GLOVINSKAJA, M. JA. 2001. Obščie i specifičeskie processy v jazyke metropolii i èmigracii. In: E. A. Zemskaja, ed., Jazyk russkogo zarubež’ja: Obščie processy i rečevye portrety, pp. 339–492. Moscow, Wien: Jazyki slavjanskoj kul’tury: Venskij slavističeskij al’manah. GOODWIN CHARLES 1981. Conversational organization: Interaction between speakers and hearers. New York, NY: Academic Press. GOODWIN, CHARLES 1987. Forgetfulness as an interactive resource. Social Psychology Quarterly 50 (2), pp. 115–131. GOODWIN, CHARLES & JOHN HERITAGE 1990. Conversation analysis. Annual Review of Anthropology 19, pp. 283–307. GOODWIN, MARJORIE H. & CHARLES GOODWIN 1986. Gesture and coparticipation in the activity of searching for a word. Semiotica 62 (1–2), pp. 51–75. GRENOBL’, LENORE A. 2008. Sintaksis i sovmestnoe postronie repliki v russkom dialoge. Voprosy jazykoznanija 2008, 1, pp. 25–36. GRIŠINA, E. A. & S. O. SAVČUK 2009. Korpus ustnyh tekstov v NKRJA: sostav i struktura. In: V. A. Plungjan, ed., Nacional’nyj korpus russkogo jazyka: 2006–2008. Novye rezul’taty i perspektivy, pp. 129–149. Saint-Petersburg: Nestor-Istorija. [Available at: http://ruscorpora.ru/sbornik2008/07.pdf] GUMPERZ, JOHN J. 1982. Discourse strategies. Cambridge: Cambridge University Press. HABERLAND, HARTMUT 2011. Ownership and maintenance of a language in transnational use: Should we leave our lingua franca alone? Journal of Pragmatics 43, pp. 937– 949. HAKIMOV, È. R. & E.G. TRUSOVA 2010. Russkij jazyk kak lingva franka v respublike Udmurtija. In: A. Mustajoki, E. Protassova & N. Vahtin, eds., Instrumentarium of Linguistics. Sociolinguistic Approaches to Non-Standard Russian, pp. 127–138. Helsinki. HAKULINEN, AULI 1997. Vuorottelujäsennys. In: Liisa Tainio, ed., Keskustelunanalyysin perusteet, pp. 32–55. Tampere: Vastapaino. HAKULINEN, AULI & MARGRET SELTING (EDS.) 2005. Syntax and lexis in conversation. Amsterdam: John Benjamins. HAYASHI, MAKOTO 2003. Language and the body as resources for collaborative action: a study of word searches in Japanese conversation. Research on Language and Social Interaction 36 (2), pp. 109–141. HELASVUO, MARJA-LIISA 2001A. Syntax in the Making. The emergence of syntactic units in Finnish conversation. Amsterdam: Benjamis. HELASVUO, MARJA-LIISA. 2001B. Emerging syntax for interaction: Noun phrases and clauses as a syntactic resource for interaction. In: Margret Selting & Elizabeth Couper-Kuhlen, eds., Studies in interactional linguistics, pp. 25–50. Amsterdam: John Benjamins. HELASVUO, MARJA-LIISA 2004. Shared syntax: the grammar of co-construction. Journal of Pragmatics 36 (8), pp. 1315–1336. HELASVUO, MARJA-LIISA, MINNA LAAKSO & MARJA-LEENA SORJONEN 2004. Searching for Words: Syntactic and Sequential Construction of Word Search in Conversations 188 References of Finnish Speakers with Aphasia. Research on Language and Social Interaction 37 (1), pp. 1–37. HENNOSTE, TILT, LEELO KEEVALLIK & KARL PAJUSALU 1999. Introduction. International journal of the Sociology of Language 139, pp. 1-16. HERITAGE, JOHN 1985. Analyzing news interviews: aspects of the production of talk for an overhearing audience. In: T. van Dijk, ed., Handbook of discourse analysis Vol. 3, pp. 95–117. London: Academic Press. HERITAGE, JOHN 1996. Harold Garfinkel ja etnometodologia. Helsinki: Gaudeamus. (J. Heritage. Garfinkel and ethnomethodology. Cambridge: Polity Press, 1984.) HERITAGE, JOHN 2002. The limits of questioning. Negative interrogatives and hostile question content. Journal of Pragmatics 34 (10–11), pp. 1427–1446. HOPPER, PAUL 1987: Emergent grammar. Berkeley Linguistics Society 13 s. 139–157. HOPPER, PAUL 1998: Emergent grammar. In: Michael Tomasello, ed., The new psychology of language s. 155–175. Mahwah, NJ: Lawrence Erlbaum. HOPPER, PAUL 2011: Emergent grammar and temporality in interactional linguistics. In: Peter Auer & Stefan Pfänder, eds., Constructions: emerging and emergent, pp. 22– 44. Berlin: De Gruyter. HOPPER, PAUL & SANDRA A. THOMPSON 2008. Projectability and clause combining in interaction. In: Ritva Laury, ed., Crosslinguistic studies of clause combining. The multifunctionality of conjunctions, pp. 99–123. Amsterdam: John Benjamins. HOUSE, JULIANE 1999. Misunderstanding in Intercultural Communication. Interactions in English as Lingua Franca and the Myth of Mutual Intelligibility. In: C. Gnutzmann, ed. Teaching and Learning English as a Global Language, pp.72–88. Tübingen: Stauffenburg. HOUSE, JULIANE 2002. Communicating in English as a lingua franca. EUROSLA Yearbook 2, pp. 243–261. John Benjamins Publishing Company. HOUSE, JULIANE 2003. English as a lingua franca: A threat to multilingualism? Journal of Sociolinguistics, 7(4), pp. 556–578. HÜLMBAUER, CORNELIA 2009. ”We don’t take the right way. We just take the way that we think you will understand” – The Shifting Relationship between Correctness and Effectiveness in ELF. In: Anna Mauranen & Elina Ranta, eds., English as a lingua franca: Studies and findings, pp. 323–347. Newcastle upon Tyne: Cambridge Scholars. HÜTTNER, JULIA 2009. Fluent speakers – fluent interactions: On the creation of (co)Fluency in English as a lingua franca. In: Anna Mauranen & Elina Ranta, eds., English as a lingua franca: Studies and findings, pp. 274–297. Newcastle upon Tyne: Cambridge Scholars. IIVONEN, ANTTI 2009. Major features of standard Finnish phonetics. In: Viola de Silva & Riikka Ullakonoja, eds., Phonetics of Russian and Finnish.General description of phonetic systems. Experimental studies on spontaneous and read-aloud speech, pp. 47–65. Frankfurt am Main: Peter Lang. ISUPOVA, O. G. 2002. Konversacionnyj analiz: predstavlenie metoda. Sociolingvistika, 15, pp. 33–52. [Available: http://www.isras.ru/files/File/4M/15/Isupova.pdf] JEFFERSON, GAIL 1987. On exposed and embedded correction in conversation. In: Graham Button & John R. E. Lee, eds., Talk and social organization, pp. 86–100. Clevedon: Multilingual Matters. JEFFERSON, GAIL 1990. List-construction as a task and resource. In: G. Psathas, ed., Interactional competence, pp. 63–92. New York, NY: Irvington Publishers. 189 References JENKINS, JENNIFER 2000. The Phonology of English as an International Language: New Models, New Norms, New Goals. Oxford: Oxford University Press. JENKINS, JENNIFER 2002. A sociolinguistically based, empirically researched pronunciation syllabys for English as an international language. Applied Linguistics 23 (1), pp. 83–103. JENKINS, JENNIFER 2003. World Englishes. A resource book for students. London: Routledge. JENKINS, JENNIFER 2006. Points of view and blind spots: ELF and SLA. International Journal of Applied Linguistics 16 (2), pp. 137–162. JENKINS, JENNIFER 2007. English as a lingua franca: Attitude and identity. Oxford: Oxford University Press. JENKINS, JENNIFER 2009. (Un)pleasant? (In)correct? (Un)intelligible? ELF speakers’ perceptions of their accents. In: Anna Mauranen & Elina Ranta, eds., English as a lingua franca: Studies and findings, pp. 10–36. Newcastle upon Tyne: Cambridge Scholars. JENKINS, JENNIFER 2011. Accommodating (to) ELF in the international university. Journal of Pragmatics 43, pp. 926–936. JÖNSSON-KORHOLA, HANNELE & ANNA-RIITTA LINDGREN (EDS.) 2003. Monena suomi maailmalla. Suomalaisperäisiä kielivähemmistöjä. (=Tietolipas 190.) Helsinki: Suomalaisen kirjallisuuden seura. KACHRU, BRAJ B. 1992 [1982]. Teaching World Englishes. In: Braj. B. Kachru, ed., The other tongue. English across cultures, pp. 355–366. University of Illinois Press. KAIVAPALU, ANNEKATRIN 2005. Lähdekieli kielenoppimisen apuna. Jyväskylä: Jyväskylän yliopisto. KAPANADZE, L. A. 1984. Sovremennoe gorodskoe prostorečie i literaturnyj jazyk. In: E. A. Zemskaja & D. N. Šmelëv, eds., Gorodskoe prostorečie. Problemy izučenija, pp. 5–12. Moscow: Nauka. KIBRIK, A.A. & V. I. PODLESSKAJA 2003. K sozdaniju korpusov ustnoj russkoj reči: principy transkribirovanija. Naučno-tehničeskaja informacija. Serija 2:6, pp. 5–12. KLIMPFINGER, THERESA 2009. ”She’s mixing the two languages together” – Forms and functions of code-switching in English as a lingua franca. In: Anna Mauranen & Elina Ranta, eds., English as a lingua franca: Studies and findings, pp. 348–371. Newcastle upon Tyne: Cambridge Scholars. KOIVISTO, AINO 2011. Sanomattakin selvää? Ja, mutta ja että puheenvuoron lopussa. Doctoral dissertation. University of Helsinki. KOIVISTO, HELINÄ 1994. Ulkomaalaissuomen syntaksia. (=Tampereen yliopiston suomen kielen ja yleisen kielitieteen laitoksen julkaisuja 17.) Tampere: Tampereen yliopisto. KOPELIOVICH, S. 2010. The variety of Russian as a heritage language in Israel: Hebrew-induced changes at the abstract level. In: A. Mustajoki, E. Protassova & N. Vahtin, eds., Instrumentarium of Linguistics. Sociolinguistic Approaches to NonStandard Russian, pp. 403–417. Helsinki. KOSHIK, IRENE 2002. Designedly incomplete utterances: a pedagogical practice for eliciting knowledge displays in error correction sequences. Research on Language and Social Interaction 35 (3), pp. 277–309. KRASHEN, STEPHEN D. 1978. Individual variation in the use of the monitor. In: William C. Ritchie, ed., Second language acquisition research, pp. 175–183. New York: Academic Press. KURHILA, SALLA 2003. Co-constructing understanding in second language conversation. Helsinki: University of Helsinki. 190 References KURHILA, SALLA 2004. Clients or language lerners – Being a second language speaker in institutional interaction. In: Rod Gardner & Johannes Wagner, eds., Second language conversations, pp. 58–74. London, New York: Continuum. KURHILA, SALLA 2006. Second language interaction. Amsterdam: John Benjamins. LAAKSO, MINNA 1997. Self-initiated repair by fluent aphasic speakers in conversation. Studia Fennica Linguistica 8. Helsinki: Finnish Literature Society. LAAKSO, MINNA 2006. Kaksivuotiaiden lasten oman puheen korjaukset keskustelussa. Puhe ja kieli 2006 (2), pp. 123–136. LAPPALAINEN, HANNA 2004. Variaatio ja sen funktiot: erään sosiaalisen verkoston jäsenten kielellisen variaation ja vuorovaikutuksen tarkastelua. Helsinki: SKS. LEHTONEN, HEINI 2006. Morfologinen variaatio maahanmuuttajataustaisten helsinkiläisnuorten puheessa. In: Kaisu Juusela & Katariina Nisula, eds., Helsinki kieliyhteisönä, pp. 255–274. University of Helsinki. LEISIÖ, LARISA 2001. Morphosyntactic convergence and integration in Finland Russian. Doctoral dissertation. Tampere: University of Tampere. LEMIVAARA-KHUDOIKULOVA, NOORA 2011. Sociolingvističeskij analiz roli russkogo jazyka v Tadžikistane: jazykovaja povsednevnost’ odnoj tadžikskoj sem’i (na osnove bytovyh pis’mennyh tekstov. University of Helsinki. LERNER, GENE H. 1989. Notes on overlap management in conversation: The case of delayed completion. Western Journal of Speech Communication 53, pp. 167–177. LERNER, GENE H. 1991. On the syntax of sentences-in-progress. Language in Society 20, pp. 441–458. LERNER, GENE H. 1992. Assisted Storytelling: Deploying Shared Knowledge as a Practical Matter. Qualitative Sociology 15(3), pp. 247–271. LERNER, GENE H. 1994. Responsive list construction. A conversational resource for accomplishing multifaceted social action. Journal of Language and Social Psychology 13, pp. 20–33. LERNER, GENE H. 1996. On the “semi-permeable” character of grammatical units in conversation: conditional entry into the turn space of another speaker. In: Elinor Ochs, Emanuel A. Schegloff & Sandra A. Thompson, eds., Interaction and grammar, pp. 228–276. Cambridge: Cambridge University Press. LERNER, GENE H. 2004A. Collaborative turn sequences. In: Gene H. Lerner, ed., Conversation Analysis. Studies from the first generation, pp. 225–256. John Benjamins Publishing Company. LERNER, GENE H. 2004B. On the place of linguistic resources in the organization of talkin-interaction: Grammar as action in prompting a speaker to elaborate. Research on Language and Social Interaction 37 (2), pp. 151–184. LERNER, GENE H. & TAKAGI TOMOYO 1999. On the place of linguistic resources in the organization of talk-in-interaction: a co-investigation of English and Japanese grammatical practices. Journal of Pragmatics 31 (1), pp. 49–75. LILJA, NIINA 2010. Ongelmista oppimiseen. Toisen aloittamat korjausjaksot kakkoskielisessä keskustelussa. Jyväskylä: Jyväskylän yliopisto. LOCAL, JOHN & GARETH WALKER 2004. Abrupt-joins as a resource for the production of multi-unit, multi-action turns. Journal of Pragmatics 36 (8), pp. 1375–1403. LUKINA M. 2005 [2003]. Tehnologija interv’ju. Moscow: Aspekt Press. LÜDI, GEORGES 2006. Multilingual repertoires and the consequences for linguistic theory. In: Kristin Bührig & Jan D. ten Thije, eds., Beyond misunderstanding. Linguistic analyses of intercultural communication, pp. 11–42. Amsterdam, John Benjamins. MAKAROV, M. M. 2003. Osnovy teorii diskursa. Moscow: ITDGK Gnozis. 191 References MARGUTTI, PIERA 2010. On designedly incomplete utterances: What counts as learning for teachers and students in primary classroom interaction. Research on Language and Social Interaction 43 (4), pp. 315–345. MARKEE, NUMA 2000. Conversation analysis. Mahwah: Lawrence Erlbaum Associates. MAURANEN, ANNA 2006A. Signaling and preventing misunderstanding in English as lingua franca communication. International Journal of the Sociology of Language 177, pp. 123–150. MAURANEN, ANNA 2006B. A Rich Domain of ELF – the ELFA Corpus of Academic Discourse. Nordic Journal of English Studies. Special issue: English as a Lingua Franca 5 (2), pp. 145–159. MAURANEN, ANNA 2009A. Englanti lingua francana – onko sillä kulttuuria? In: Anna Idström & Sachiko Sosa, eds., Kielissä kulttuurin ääni, pp. 290–305. Helsinki: SKS. MAURANEN, ANNA 2009B. Introduction. English as a Lingua Franca: Studies and Findings. In: Anna Mauranen & Elina Ranta, eds., pp. 1–7. Newcastle upon Tyne: Cambridge Scholars. MAURANEN, ANNA & MARIA METSÄ-KETELÄ 2006. Introduction: English as a Lingua Franca. Nordic Journal of English Studies. Special issue: English as a Lingua Franca 5 (2), pp. 1–8. MAURANEN, ANNA & ELINA RANTA (EDS.) 2009. English as a lingua franca: Studies and findings. Newcastle upon Tyne: Cambridge Scholars. MAZELAND, HARRIE 2007. Parenthetical sequences. Journal of Pragmatics 39 (10), pp. 1816–1869. MAZELAND, HARRIE & MINNA ZAMAN-ZADEH 2004. The logic of clarification: Some observations about word-clarification repairs in Finnish-as-a-lingua franca interactions. In: Rod Gardner & Johannes Wagner, eds., Second language conversations, pp. 132–156. London, New York: Continuum. MEIERKORD, CHRISTIANE 1998. Lingua Franca English. Characteritics of successul nonnative – nonnative speaker discourse. Erfurt Electronic Studies in English (EESE), 7 Available at: http://webdoc.sub.gwdg.de/edoc/ia/eese/rahmen22.html [Accessed January 15, 2013]. MEIERKORD, CHRISTIANE 2004. Syntactic variation in interactions across international Englishes. English World-Wide 25 (1), pp. 109–132. MEIERKORD, CHRISTIANE 2006. Lingua franca communication past and present. International Journal of the Sociology of Language 177, pp. 9–30. MOLLIN, SANDRA 2006. English as a lingua franca: A new variety in the new expanding circle? Nordic Journal of English Studies 5 (2), pp. 41–57. MONDADA, LORENZA 2004. Ways of ’doing being plurilingual’ in international work meetings. In: Rod Gardner & Johannes Wagner, eds., Second language conversations, pp. 18–39. London, New York: Continuum. MONONEN, KAARINA 2013. Inkerinsuomalaisten suomen kielen käyttö Pietarissa ja sen lähialueella. Doctoral dissertation. University of Helsinki. MUSTAJOKI, ARTO 2010. Types of non-standard communication encounters with special reference to Russian. In: Mika Lähteenmäki & Marjatta Vanhala-Aniszewski, eds., Language ideologies in transition multilingualism in Russia and Finland, pp. 35–55. Frankfurt am Main: Peter Lang. MUSTAJOKI, ARTO 2011. Počemu obščenie na lingva franka udaëtsja tak horošo? In: N. Vahtin, ed., Jazyki sosedej: mosty ili bar’ery? Problemy dvujazyčnoj kommunikacii, pp. 10–31. Saint Petersburg. 192 References MUSTAJOKI, ARTO 2012. A speaker-oriented multidimensional approach to risks and causes of miscommunication. Language and Dialogue 2:2, pp. 216–243. John Benjamins Publishing Company. MUSTAJOKI, ARTO 2013A. Raznovidnosti russkogo jazyka: analiz i klassifikacija. Voprosy jazykoznanija, 5, pp. 3–27. MUSTAJOKI, ARTO 2013B. Risks of miscommunication in various speech genres. In: E. Borisova & O. Souleimanova, eds., Understanding by Communication, pp. 33–53. Cambridge: Cambridge Scholars Publishing. MYERS-SCOTTON, CAROL 1992. Comparing codeswitching and borrowing. Journal of Multilingual OCHS, ELINOR, EMANUEL A. SCHEGLOFF & SANDRA A. THOMPSON (EDS.) 1996. Interaction and grammar. Cambridge: Cambridge University Press. ONO, TSUYOSHI & SANDRA A. THOMPSON 1995. What Can Conversation Tell Us About Syntax? In: Philip W. Davies, ed., Descriptive and Theoretical Modes in the Alternative Linguistics, pp. 213–271. John Benjamins Publishing Company. Amsterdam. ONO, TSUYOSHI & SANDRA A. THOMPSON 1996. Interaction and Syntax in the Structure of Conversational Discourse. In: Eduard H. Hovy & Donia R. Scott, eds., Collaboration, Overlap, and Syntactic Dissociation. Computational and Conversational Discourse: Burning Issues – An Interdisciplinary Account. Springer, Berlin. ORESTRÖM, BENGT 1983. Turn-taking in English conversation. Lund: Gleerup. ORLETTI, FRANCA 2000. The conversational construction of social identity in native/nonnative interaction. In: Aldo Di Luzio, Susanne Günthner & Franca Orletti, eds., Culture in communication: Analyses of intercultural situations, pp. 271–294. John Benjamins Publishing Company. OSF = OFFICIAL STATISTICS OF FINLAND: Population structure [e-publication]. ISSN=1797-5395. annual review 2011. Helsinki: Statistics Finland [referred: 3.12.2012]. [Available at: http://www.stat.fi/til/vaerak/2011/01/vaerak_2011_01_2012-1130_tie_001_en.html.] PAUKKERI, PIRKKO 2006. Recipient v russkom razgovore: o raspredelenii funkcij meždu otvetami da, nu i tak. Doctoral dissertation. University of Helsinki. [Available at: http://ethesis.helsinki.fi/julkaisut/hum/slavi/vk/paukkeri/recipien.pdf.] PAVLENKO, ANETA 2006. Russian as a lingua franca. Annual Review of Applied Linguistics 26, pp. 78–99. PAVLENKO, ANETA 2008. Russian in post-Soviet countries. Russian Linguistics 32, pp. 59–80. PEKAREK DOEHLER, SIMONA 2011. Emergent grammar for all practical purposes: The online formating of dislocated constructions in French conversation. In: Peter Auer & Stefan Pfänder, eds., Constructions: emerging and emergent, pp. 46–88. Mouton de Gruyter. PERÄKYLÄ, ANSSI 1997. Institutionaalinen keskustelu. In: Liisa Tainio, ed., Keskustelunanalyysin perusteet, pp. 177–203. Tampere: Vastapaino. PHILLIPSON, ROBERT 2008. Lingua franca or lingua frankensteinia? English in European integration and globalisation. World Englishes 27 (2), pp. 250–267. PODLESSKAJA, V. I. & A. A. KIBRIK 2009. Rečevye sboi i zatrudnenija. In: A. A. Kibrik & V. I. Podlesskaja, eds., Rasskazy o snovidenijah. Korpusnoe issledovanie ustnogo russkogo diskursa, pp. 177–218. Moscow: Jazyki slavjanskih kul’tur. 193 References PODPORENKO, JU. 2001. Bespraven, no vostrebovan. Russkij jazyk v Uzbekistane. Družba narodov. POLINSKY MARIA & OLGA KAGAN 2007. Heritage Languages: In the ‘Wild’ and in the Classroom. Language and Linguistics Compass 1/5: 368–395. POMERANTZ, ANITA 1984. Agreeing and disagreeing with assessments: some features of preferred/dispreferred turn shapes. In: J. Maxwell Atkinson & John Heritage, eds. Structures of social action: Studies in Conversation Analysis, pp. 57–101. Cambridge: Cambridge University Press. PRODROMOU, LUKE 2007. Bumping into creative idiomaticity. English Today 89, 23 (1), pp. 14–25. PROTASOVA, E. JU. 2004. Fennorossy: žizn’ i upotreblenie jazyka. Saint-Petersburg: Zlatoust. PSATHAS, GEORGE 1995. Conversation Analysis. The study of talk-in-interaction. Thousand Oaks, California: Sage. RAEVAARA, LIISA 1997. Vierusparit – esimerkkinä kysymys ja vastaus. In: Liisa Tainio, ed., Keskustelunanalyysin perusteet, pp. 75–92. Tampere: Vastapaino. RAEVAARA, LIISA, JOHANNA RUUSUVUORI & MARKKU HAAKANA 2001. Institutionaalinen vuorovaikutus ja sen tutkiminen. In: Johanna Ruusuvuori, Markku Haakana & Liisa Raevaara, eds., Institutionaalinen vuorovaikutus: Keskustelunanalyyttisiä tutkimuksia, pp. 11–38. Helsinki: SKS. RAMPTON, BEN 1998. Language crossing and the redefinition of reality. In: Peter Auer, ed., Code-switching in conversation. Language, interaction and identity, pp. 290– 317. London: Routledge. RANTA, ELINA 2006. The ‘attractive’ progressive – Why use the -ing form in English as a lingua franca? Nordic Journal of English Studies 5 (2), pp. 95–116. RANTA, ELINA 2009. Syntactic features in spoken ELF – Learner language or spoken grammar? In: Anna Mauranen & Elina Ranta, eds., English as a lingua franca: Studies and findings, pp. 84–106. Newcastle upon Tyne: Cambridge Scholars. RIGGENBACH, HEIDI 1991. Toward an Understanding of Fluency: A Microanalysis of Nonnative Speaker Conversations. Discourse Processes 14 (4), pp. 423–441. ROUTARINNE, SARA 2003: Tytöt äänessä. Parenteesit ja nouseva sävelkulku kertojien vuorovaikutuskeinoina. Helsinki: Suomalaisen Kirjallisuuden Seura. SACKS, HARVEY 1992A. Lectures on conversation I. Ed. by Gail Jefferson. Oxford: Blackwell. SACKS, HARVEY 1992B. Lectures on conversation II. Ed. by Gail Jefferson. Oxford: Blackwell. SACKS, H. & E.A. SCHEGLOFF 1979. Two preferences in the organization of reference to persons in conversation and their interaction. In: G. Psathas, ed., Everyday Language, pp. 15–21. New York: Irvington. SACKS, HARVEY, SCHEGLOFF, EMANUEL A. & JEFFERSON, GAIL 1974. A simplest systematics for the organization of turn-taking in conversation. Language 50 (4), pp. 696–735. SAMMALKORPI, ILONA 2006. Analiz institucional’nogo obščenija v slučae kommunikativnoj neudaci. Pro gradu -tutkielma. Historiallis-kielitieteellinen kirjasto. Helsinki. MARJO 2011. Yhteisestä toiminnasta yhteiseen kieleen. SAVIJÄRVI, Keskustelunanalyyttinen tutkimus toisen kielen oppimisesta kielikylpypäiväkodin arkitilanteissa. Doctoral dissertation. University of Helsinki. 194 References SCHEGLOFF, EMANUEL A. 1979. The relevance of repair to syntax-for-conversation. In: Talmy Givón, ed., Syntax and semantics, volume 12: discourse and syntax, pp. 261– 286. New York: Academic Press. SCHEGLOFF, EMANUEL A. 1982. Discourse as an interactional achievement: Some uses of 'uh huh' and other things that come between sentences. In: Deborah Tannen, ed., Analyzing discourse. Text and talk, pp. 71–93. Washington D.C.: Georgetown University Press. SCHEGLOFF, EMANUEL A. 1984. On Some Questions and Ambiguities in Conversation. In: J. Maxwell Atkinson & John Heritage, eds., Structures of Social Action. Studies in Conversation Analysis, pp. 28–52. Cambridge University Press. Cambridge. SCHEGLOFF, EMANUEL A. 1987. Analyzing single episodes of interaction: An exercise in Conversation Analysis. Social Psychology Quarterly 50 (2), pp. 101–114. SCHEGLOFF, EMANUEL A. 1991. Reflections on talk and social structure. In: Deirdre Boden & Don H. Zimmerman, eds., Talk and social structure. Studies in ethnomethodology and Conversation Analysis, pp. 44–70. Cambridge: Polity Press. SCHEGLOFF, EMANUEL A. 1992. Repair after next turn: The last structurally provided defense of intersubjectivity in conversation. American Journal Of Sociology 97(5), pp. 1295–1345. SCHEGLOFF, EMANUEL A. 1996. Turn organization: one intersection of grammar and interaction. In: Elinor Ochs, Emanuel A. Schegloff & Sandra A. Thompson, eds., Interaction and grammar, pp. 52–133. Cambridge: Cambridge University Press. SCHEGLOFF, EMANUEL A. 1997. Practices and actions: Boundary cases of other-initiated repair. Discourse Processes 23 (3), pp. 499–545. SCHEGLOFF, EMANUEL A. 2000. When ’others’ initiate repair. Applied Linguistics 21 (2), pp. 205–243. SCHEGLOFF, EMANUEL A. 2007. Sequence organization in interaction. A Primer in Conversation Analysis I. Cambridge: Cambridge University Press. SCHEGLOFF, EMANUEL A., GAIL JEFFERSON & HARVEY SACKS 1977. The preference for self correction in the organization of repair in conversation. Language 53 (2), pp. 361–382. SCHEGLOFF, EMANUEL A. & HARVEY SACKS 1973. Opening up closings. Semiotica 8 (4), pp. 289–327. SEEDHOUSE, PAUL 1998. CA and the analysis of foreign language interaction: A reply to Wagner. Journal of Pragmatics 30 (1), pp. 85–102. SEIDLHOFER, BARBARA 2000. Mind the gap: English as a mother tongue vs English as a lingua franca. Vienna English Working Papers 9 (1), pp. 51–68. SEIDLHOFER, BARBARA 2001. Closing a conceptual gap: The case for a description of english as a lingua franca. International Journal of Applied Linguistics 11 (2), pp. 133–158. SEIDLHOFER, BARBARA 2004. Research perspectives on teaching English as a lingua franca. Annual Review of Applied Linguistics 24, pp. 209–239. SEIDLHOFER BARBARA 2005. English as a lingua franca. ELT Journal Volume 59 (4), pp. 339–341. SEIDLHOFER, BARBARA 2009. Orientations in ELF research: Form and function. In: Anna Mauranen & Elina Ranta, eds., English as a lingua franca: Studies and findings, pp. 37–59. Newcastle upon Tyne: Cambridge Scholars. SEIDLHOFER, BARBARA 2011. Understanding Englihs as a Lingua Franca. Oxford University Press. 195 References SELTING, MARGRET 1996. On the interplay of syntax and prosody in the constitution of turn-constructional units and turns in conversation. Pragmatics 6, pp. 357–388. SELTING, MARGRET 2000. The construction of units in conversational talk. Language in Society 29, pp. 477–517. SELTING, MARGRET 2005. Syntax and prosody as methods for the construction and identification of turn-constructional units in conversation. In: Auli Hakulinen & Margret Selting, eds., Syntax and lexis in conversation. Studies on the use of linguistic resources in talk-in-interaction, pp. 17–44. Philadelphia: John Benjamins. SELTING, MARGRET & ELIZABETH COUPER-KUHLEN (EDS.) 2001. Studies in interactional linguistics. Amsterdam: John Benjamins. SEPPÄNEN, EEVA-LEENA 1997. Vuorovaikutus paperilla. In: Liisa Tainio, ed., Keskustelunanalyysin perusteet, pp. 18–31. Tampere: Vastapaino. SIITONEN, KIRSTI & MAISA MARTIN 2001. Mämmilän Mukun suomi – ulkomaalaispuhetta vai pelkkää fiktiota? XXVI Kielitieteen päivät Oulussa 19.– 20.5.2000, pp. 256–264. Suomen ja saamen kielen ja logopedian laitos. Oulu: Oulun yliopisto. SORJONEN, MARJA-LEENA 1997. Korjausjäsennys. In: Liisa Tainio, ed., Keskustelunanalyysin perusteet, pp. 111–137. Tampere: Vastapaino. SORJONEN, MARJA-LEENA 1999. Dialogipartikkelien tehtävistä. Virittäjä 2/1999, pp. 170–194. SORJONEN, MARJA-LEENA 2001. Responding in conversation: A study of response particles in Finnish. Amsterdam: John Benjamins. SORJONEN, MARJA-LEENA & MINNA LAAKSO 2005. Katko vai eiku? Itsekorjauksen aloitustavat. Virittäjä 2/2005, pp. 244–271. SUVINIITTY, JAANA 2012. Lectures in English as a Lingua Franca: Interactional Features. PhD. University of Helsinki. SVENNEVIG, JAN 2008. Trying the easiest solution first in other-initiation of repair. Journal of Pragmatics 40 (2), pp. 333–348. SZCZEPEK, BEATRICE 2000A. Formal aspects of collaborative productions in English conversation. InLiSt – Interaction and Linguistic Structures, No. 17. Saatavilla: http://www.inlist.uni-bayreuth.de/ SZCZEPEK, BEATRICE 2000B. Functional aspects of collaborative productions in English conversation. InLiSt – Interaction and Linguistic Structures, No. 21. Saatavilla: http://www.inlist.uni-bayreuth.de/ SZCZEPEK, BEATRICE 2010. Speech rhythm across turn transitions in cross-cultural talkin-interaction. Journal of Pragmatics 42 (4), pp. 1037–1059. TARONE, ELAINE 1983. Some thoughts on the notion of ’communication strategy’. In: Claus Færch & Gabriele Kasper, eds., Strategies in interlanguage communication, pp. 61–74. London: Longman. THIJE, JAN D. TEN 2003. The Transition from Misunderstanding to Understanding in Intercultural Communication. In: László I. Komlósi, Peter Houtlosser & Michiel Leezenberg, eds., Communication and culture. Argumentative, cognitive and linguistic perspectives, pp. 197–213. Amsterdam: Sic Sac. THOMASON, SARAH 2001. Language Contact. An Introduction. Edinburgh University Press. THOMPSON, SANDRA A. & ELIZABETH COUPER-KUHLEN 2005. The clause as a locus of grammar and interaction. – Discourse Studies 7, pp. 481–505. TIITTULA, LIISA 1985. Puheenvuorojen vaihtuminen keskustelussa. Virittäjä 1985, pp. 319–336. 196 References VAHTIN, N., A. MUSTAJOKI & E. PROTASOVA 2010. Russkie jazyki. In: A. Mustajoki, E. Protassova & N. Vahtin, eds., Instrumentarium of Linguistics. Sociolinguistic Approaches to Non-Standard Russian, pp. 5–16. Helsinki. VERSCHIK, ANNA 2004. Aspects of Russian-Estonian codeswitching: Research perspectives. International Journal of Bilingualism 8(4), pp. 427–448. VERSCHIK, ANNA 2006. Convergence in Estonia's Russian: Directional vs. static vs. separative verb. International Journal of Bilingualism 10(4), pp. 383–404. VERSCHIK, ANNA 2007. The language situation in Estonia. Journal of Baltic Studies (Baltic Sociolinguistic Review, Special Issue), 36(3), pp. 283–317. VERSCHIK, ANNA 2011. Practicing receptive multilingualism: Estonian-Finnish communication in Tallinn. International Journal of Bilingualism 16(3), pp. 265–286. VICKERS, CAROLINE H. 2010. Language competence and construction of expert-novice in NS-NNS interaction. Journal of Pragmatics, 42 (1), pp. 116–138. VISK = Auli Hakulinen, Maria Vilkuna, Riitta Korhonen, Vesa Koivisto, Tarja Riitta Heinonen ja Irja Alho 2004: Iso suomen kielioppi. Helsinki: Suomalaisen Kirjallisuuden Seura. [Available at: http://scripta.kotus.fi/visk. URN:ISBN:978-9525446-35-7.] WALKER, GARETH 2012. Coordination and interpretation of vocal and visible resources: ‘Trail-off’ conjunctions. Language and Speech 55 (1), pp. 141–163. WERLEN, I. 2007. Receptive multilingualism in Switzerland and the case of Biel/Bienne. In: J. D. ten Thije & L. Zeevaert, eds., Receptive multilingualism, pp. 137–157, Amsterdam: John Benjamins. WIDDOWSON, H., 1994. The ownership of English. TESOL Quarterly, 28(2), pp. 377– 389. WONG, JEAN 2004. Some preliminary thoughts on delay as an interactional resource. In: Rod Gardner & Johannes Wagner, eds., Second language conversations, pp. 114– 131. London, New York: Continuum. WONG, JEAN & DAVID OLSHER 2000. Reflections on conversation analysis and nonnative speaker talk: An interview with Emanuel A. Schegloff. Issues in Applied Linguistics 11 (1), pp. 11–128. WOOD, DAVID 2006. Uses and functions of formulaic sequences in second language speech: An exploration of the foundations of fluency. The Canadian Modern Language Review/La Revue canadienne des langues vivante 63 (1) (September/septembre), pp. 13–33. WRIGHT, SUE 2006. French as a lingua franca. Annual Review of Applied Linguistics, 26, pp. 35–60. YULE, GEORGE & ELAINE TARONE 1997. Investigating L2 reference: Pros and cons. In: G. Kasper & E. Kellerman, eds., Advances in Communication Strategy Research, pp. 17–30. New York: Longman. ZABRODSKAJA, ANASTASSIA 2009. Russian-Estonian language contacts: Grammatical aspects of language use and change. Tallinna Ülikool. ZAPRUDSKI, SIARHIEJ 2007. In the grip of replacive bilingualism: the Belarusian language in contact with Russian. International Journal of the Sociology of Language 183, pp. 97–118. ZEEVAERT, L & J.D. TEN THIJE 2007. Introduction. In: J.D. ten Thije & L Zeevaert, eds., Receptive Multilingualism. Linguistic analyses, language policies and didactic concepts, pp. 1–25, Amsterdam: Benjamins. 197 References ZEMSKAJA, E. A. 2001. Obščie jazykovye processy i individual’nye jazykovye processy. In: E. A. Zemskaja, ed., Jazyk russkogo zarubež’ja: Obščie processy i rečevye portrety, pp. 23–338. Moscow, Wien: Jazyki slavjanskoj kul’tury: Venskij slavističeskij al’manah. ZEMSKAJA, E. A, M. B. KITAJGORODSKAJA & E. N. ŠIRJAEV 1981. Russkaja razgovornaja reč’. Obščie voprosy, slovoobrazovanie, sintaksis. Moscow: Nauka. ZEVELEV, IGOR 2001. Russia and its new diasporas. Washingon: United States Institute of Peace Press. ZHDANOVA, V. V. 2012. Jazyk russkoj diaspory: k probleme tipologii morfologičeskih i sintaksičeskih harakteristik. In: Ju. Apresjan, I. Boguslavsky, M.-C. L'Homme, L. Iomdin, J. Milic'evic', A. Polguere, L. Wanner, eds., Meanings, Texts and Other Exciting Things. A Festschrift to Commemorate the 80th Anniversary of Professor Igor Alexandrovich Mel'chuk., pp. 682–695. Moscow: Languages of Slavic Culture. 198 Appendix Symbols used in transcription . falling intonation , slightly falling intonation ? rising intonation ↑ next word or syllable pronounced higher than surrounding speech ↓ next word or syllable pronounced lower than surrounding speech [ the beginning of overlap ] the end of overlap : lengthening of sound ° soft voice <> slow speech >< fast speech haluavat emphasis .hh inhalation (h) exhalation within a word @@ laughter, smile (.) micropause (less than 0.2 seconds) (1) pause (length of the pause indicated in brackets) » latching -- cut-off word LEIPÄÄ loud voice (-) word in doubt ʔ glottal sound Grammatical glosses 1SG, 2SG, 3SG first person singular (ending or personal pronoun), etc. 1PL, 2PL, 3PL first person plural (ending or personal pronoun), etc. 4 passive person ending ABL ablative (‘from’) ACC accusative 199 Appendix ADE adessive (‘at, on’) ADJ adjectivizing derivational suffix ADV adverb ALL allative (‘to’) CLI clitic COMP comparative CON conditional mood DAT dative ELA elative (‘out of, from’) ESS essive FEM feminine GEN genitive ILL illative (‘into, to’) IMP imperative INE inessive (‘in’) INF infinitive INSTR instrumental MASC masculine NEG negation NEU neuter PASS passive PCP participle PL plural PRE preposition PREP prepositional PRT particle PST past tense PTV partitive (‘part of’) PX possessive suffix Q question clitic REFL reflexive TRANS translative (‘into’, ‘for’) VOC vocalism 200