Interactional Perspectives on Discourse Proceedings from the Organization in Discourse 3 Conference Mia Raitaniemi Sanna-Kaisa Tanskanen Marja-Liisa Helasvuo Marjut Johansson 2Interactional Perspectives on Discourse: Proceedings from the Organization in Discourse 3 Conference. Edited by Mia Raitaniemi, Sanna-Kaisa Tanskanen, Marja-Liisa Helasvuo, Marjut Johansson Anglicana Turkuensia No. 30 ISSN 1236-4754 University of Turku Turku, Finland 2010 ISBN 978-951-29-4501-6 Copyright @ individual authors Layout by Mia Raitaniemi 3Interactional Perspectives on Discourse: Proceedings from the Organization in Discourse 3 Conference. Mia Raitaniemi Introduction _____________________________________________________ 4 Hatice Çubukçu and Hatice Sofu Asking for the hand of the girl - marriage arrangement discourse in Turkish culture _________________________________________________________ 7 Miros?awa Kaczmarek Expressing social distance in Polish discourse - the use of the grammatical category of person _______________________________________________ 28 Nely Keinänen and Mari Pakkala-Weckström Linguistic Survival Strategies in Geoffrey Chaucer’s Clerk’s Tale, William Shakespeare’s Othello and Elizabeth Cary’s The Tragedy of Mariam _______ 60 Jean-Yves Malherbe La dérision comme acte de langage de l’art belge francophone ____________ 80 Helena Nilsson: Okay als Rückmeldesignal in deutscher und schwedischer Chat-Kommunikation ____________________________________________102 Mayumi Nishikawa: Oh as a discourse marker __________________________________________125 4Introduction Mia Raitaniemi This online volume contains six papers originally presented at a conference entitled “Organization in Discourse 3: The Interactional Perspective”. The conference was a cooperative venture by four language departments of the University of Turku: English, Finnish, French and German. OID3 was the third in a series of “Organization in Discourse” conferences, the first of which took place in 1994 on the initiative of the English Department. The focus of the most recent conference was on an interactional perspective, and papers and posters were invited from several research fields: discourse studies, dialogical linguistics, interactional linguistics, pragmatics, sociopragmatics, diachronic linguistics and literary studies. The organizers orient themselves in different ways with regard to interaction research, and it was their deliberate intention to establish new connections between different scholarly fields. The conference consequently welcomed papers dealing with discourse from any interactional perspective in any language from any period, representing all types of interaction: spoken, written or electronic. Accordingly, in these proceedings various linguistic topics dealing with discourse practices are represented, ranging from syntactic analyses of speech and linguistic strategies in diachronic texts to the literary and cultural expression of interaction. The purpose is to show the pervasiveness of the notion of interaction structure within linguistic studies, and the great range of phenomena it can comprehend. The notion of interaction can in fact be regarded as central in most linguistic work today, so much so that Helasvuo, Johansson and Tanskanen (2010) suggest the notion of interactional turn to describe the most recent developments in linguistics. They refer to the increased interest in studies of language as communication, ranging from the role of different communicative contexts to ideological underpinnings. In the first article, Hatice Çubukçu and Hatice Sofu investigate the discourse of Turkish marriage arrangement practices as a communicative event. 5The data are naturally occurring events, analyzed in terms of interactional moves resulting in an interactional pattern of the ceremony of k?z isteme (“asking for the girl’s hand”). The authors propose a model for the move sequence of the event. Miros?awa Kaczmarek has studied social distance in Polish discourse, focusing on the grammatical category of person. The spoken language data consist on the one hand of academic encounters between staff members and students, on the other of conversations between female acquaintances. The author analyzes shifts in the choice of personal forms of verbs, including the absence of personal forms, and concludes that their use reflects the socio-psychological distance of the interlocutors. Next, Nely Keinänen and Mari Pakkala-Weckström discuss the linguistic survival strategies of fictional married couples under the threat of marital violence in three dramas from the medieval to early modern period. They challenge the view that because of their institutional role subordinated women mainly use negative politeness strategies in defending themselves. At the intersection of gender, power and politeness, the authors combine literary and linguistic methods. They conclude that some of the linguistic strategies used by the characters show variation beyond the previous reading of these dramas. The fourth article, by Jean-Yves Malherbe, presents Belgian nationally- shared expressions of derision, in other words ways of deconstructing standard conceptions of linguistic structures. The article explores the historical roots of this deconstructing and ridiculing attitude and its manifestation in surrealistic and dadaistic poetry, comics and other forms of written and visual arts. Derision is defined as a linguistic act present in these forms of discourse and in many ways pervasive in the Belgian identity as one of its playful components. Helena Nilsson explores the use of okay as a receipt token from a comparative angle, reflecting on its use in Swedish and German chat communication. She analyzes one hundred occurrences of okej and ok in framework of conversation analysis. In Swedish chat the discourse particle okej is commonly used to signal receipt of the previous utterance, whereas the German ok is not distinctly reserved for this purpose. In German, ok is typically used for 6agreeing with the previous utterance, a usage which in the Swedish material was found only twice. In the final paper of the volume, Mayumi Nishikawa discusses the English discourse marker oh. The function of this discourse marker as indicative of such feelings as surprise and joy is contested with the interpretation of oh as a context sensitive discourse marker. Using film dialogue as material, Nishikawa investigates oh within a framework of cognitive pragmatics. The results suggest that the function of oh is to lead the hearer to interpret the subsequent utterance in the way intended by the speaker by expressing the speaker’s mental state. Nishikawa regards the context-sensitive meaning, rather than the semantic one, as specific features of this discourse marker, along with the interpersonal dimension. By examining interactional practices across several types of discourse, the six papers in this collection generate interaction between various points of view within studies of interaction. Reference: Helasvuo, Marja-Liisa, Marjut Johansson & Sanna-Kaisa Tanskanen 2010. Discourse and the interactional turn. In Tanskanen, Sanna-Kaisa, Marja- Liisa Helasvuo, Marjut Johansson & Mia Raitaniemi (eds): Discourses in Interaction (Pragmatics & Beyond New Series 203). Amsterdam: John Benjamins. 1–9. 7Asking for the hand of the girl - marriage arrangement discourse in Turkish culture Hatice Çubukçu and Hatice Sofu, Çukurova University 1. Introduction The rich array of traditional marriage practices in Turkey have been a focus of interest for researchers and provided fruitful data especially for the fields of sociology and social psychology (e.g. Timur, 1972; Ka??tç?ba??, 1982; Duben&Behar, 1991; Hortacsu, 1997). Nevertheless, the discourse dimension which is an inextricable component of such practices has remained as a vast area that needs to be explored by linguists. With this rationale in mind, the present study aims to investigate the communicative patterns of a specific type of speech event in Turkish culture, so known as ??z isteme or ‘asking for the hand of the girl’. The purpose of the ??z isteme ritual is to get the consent of the prospective bride’s family for the girl’s hand. These ceremonies as a part of the traditional marriage-arrangement procedures, are practiced both in strictly traditional settings, and in couple- initiated marriages symbolically, still preserving their specific discourse and some of the other traditional components. The paper is also intended to be complementary to the recent work conducted on the betrothal speech event which involves the formal announcement of the agreement of the families for the marriage and thus, constitutes the final phase of the marriage arrangement process (Çubukçu, 2007). The study holds an ethnographic perspective to communication that situates language in the social life and cultural systems of particular speech communities. Accordingly, understanding human communication involves investigating diversity of patterns of and practices of communities, i.e; communicative acts and events that constitute the life of a 8society (Hymes, 1962:.30).1 This view has been well received by a large body of scholars, yielding into a strong claim that human communication is organized by speech events (or, genres, as often called) in almost the same way as grammatical forms do (Bakhtin 2005; Philipsen and Coutu 2005). However, exploration of these ‘typical forms of constructions’ of everyday conversation still seems to be a challenging task for the linguist, considering Bakhtin’s (2005:128) observation that ”...no list of oral speech genres yet exists, or even a principle on which such a list may be based”. As a means of facilitating the exploration of the communicative structures of everyday conversation, scholars have been encouraged to examine communicative events which bear more stable and identifiable structures (Ventola,1983; Saville-Troike, 1989,1997). It is basically this motivation that our work has been triggered by, along with the belief that uncovering the patterns of traditional rituals enables us to see more strikingly, how culture and linguistic activity are interwoven through the human interaction. 1.1 Marriage arrangement practices in Turkey Research shows that the traditional type of marriage in Turkey is the ‘arranged marriage’ although the concept of love marriage entered urban centers starting from the 19th century and has largely increased in popularity (Duben & Behar,1991). Family-initiated marriages are more prevalent in rural areas and among less educated people. According to Hortacsu (1997), ‘traditional marriage- arrangements’ have undergone considerable changes over the time, and today, the role of the families in arranged marriages may vary from arranging meetings between marriageable individuals and leaving the decision to the couple, to being the primary decision maker about everything concerning the marriage. Hortacsu (1997) additionally points out that families also play a role in love marriages by providing large amount of the expenses required during home making. They may 1 The term speech event, is used interchangeably in the related literature, with other terms such as ‘communicative event’, ‘spoken genre’ and ‘speech genre’ (Coulthard, 1988). 9also take part in negotiating marriage arrangements with the other family. Briefly, regardless of the type of marriage -arrangement procedures, families still play an important role during the marriage arrangement process. Traditional rituals are practiced more extensively in family-initiated marriages; because this kind of marriage is more likely to take place within the same community, lineage or neighborhood ( Stirling, 1965; Ka??tç?ba??, 1982). To specify, the marriage process, in such settings, is usually initiated by ‘viewing visits’, that is, the mother of the boy and women close to the family visit the home of the girl to see her, if not seen otherwise. This may be followed by other informal visits known as the ‘inspecting visit’ to have a better idea of the girl. If they think that she is a suitable match, she is then shown to the groom, and the family arranges a ??z isteme visit to the girl’s house provided that the boy also likes the girl. A few visits are made giving the family of the girl enough time to think about it. The final visit where the family gives their overt consent, takes the form of betrothal ceremony, i.e. the formal verbal agreement for the marriage which constitutes a different speech event with its own characteristics. However, ??z isteme visits in couple-initiated marriages are not normally repeated as in the traditional settings; in contrast, they may even end up with the formal recognition of the prospective marriage i.e., the betrothal ceremony, during the same visit; and, it is more often than not, that such cases are finalized with the announcement of the engagement of the couple, thus, constituting a typical ‘hybrid’ speech event to use Hymes’s (1974) term. 1.2 Background of the study Whereas the starting point of ethnographic studies was “exploring the patterned regularity….in the activity of speaking” as restated by Hymes (2000:34), the studies in the field have taken different directions, for example, towards contrastive rhetoric and cross cultural studies which prioritized more the examination of written genres. To name only a few, analyses on academic texts (e.g., Swales,1991, Feak and Swales,2003), on educational materials (Mc Carty and Carter,1992), advertisement and electronic communication (Cook, 1992) and 10 others. The field also flourished with numerous studies on political debates, TV talk shows, quiz programs etc, as some examples of the spoken genres (Chilton &Schaffer, 1997; Holtgraves& Lasky,1999). On the other hand, interest in exploring local practices in a diversity of cultures have remained relatively limited in number and scope. Bauman & Sherzer (1974), Tannen & Saville-Troike (1985), Gumperz & Hymes (1986) and, Philipsen & Coutu (2005) have provided some ethnographic work concerning traditional marriage customs in various cultures such as, viewing visits (e.g., in Newari culture), marriage proposals (e.g., in a Japanese village). The field also includes research on other ritualistic communicative events such as politeness rituals and fortune telling in Chinese culture (Greenblatt,1979), condolence visits in Nigeria (Nwoye (1985), an interrogation of an accused by the villagers and fortune telling in Calu community (Hill and Hill, 1986), and greeting rituals among Igbo people (Nwoye, 1993). Related literature on Turkish cultural practices also seems to include few studies. Demircan (1999) as a guide to ethnographic studies, provides an inventory of 49 traditional speech situations among which communicative events such as ‘betrothal visits’ and ‘engagement ceremonies’ take place, as well as ‘funerals’ , ‘fortune telling’ ‘circumcision’ rituals etc. Alagözlü (2007), in her analysis of ‘coffee cup reading’ among women, presents a description of the traditional fortune telling practice. Finally, as mentioned earlier in this study, Çubukçu (2007), includes a study on the betrothal speech event, so known as söz kesme ceremony which corresponds to the final stage of the marriage arrangement rituals, in Turkish culture. 2. Method of Analysis 2.1 Data We have gathered the naturally occurring data from 4 different cases of ??z isteme visits when we participated in these occasions as a guest, a friend or a relative. For the sake of naturalness of the data, covert audio recordings and field notes were used as a method of data collection. Permission for using the data was later granted from the hosts. And our intention for sharing the findings of the study 11 with the couples, which eventually we did, was especially appreciated by them. We also benefited from informal conversations with the people who participated in the event as spokespersons. Although in earlier ethnographic studies, scholars obtained data as ‘participant-observers’ while conducting research in other cultures with the help of a native ‘informant’ who informed them about the cultural norms of the community; the advantages of doing research in the researchers’ own culture were specifically emphasized (E.g.,Milroy,1987; Schiffrin,1994; Saville-Troike,1997). It was stated that, ”combining observation and self knowledge, the researcher can explore subtle interconnections of meaning that the outsider could attain only with great difficulty, if at all” (Saville-Troike,1997:136). It is to say that the ethnographer can function both as native observer and informant, as the case is in our study. The same argument had earlier found grounds in etnomethodology as pronounced by Garfinkel (1967:76-7), “The discovery of common culture consists of the discovery from within the society by social scientists of the existence of commonsense knowledge of social structures”. Also, in later studies, various methods of obtaining ethnographic data have been suggested such as using ‘focus groups (see, Suter, 2000), along with additional ways for in-depth information gathering to supplement the data in hand (Saville-Troike, 2003). 2.2 Ethnographic analysis Ethnography of communication, the approach employed in this research, concerns itself with investigating how social organization of a community is reflected in language use. To this end, it has set out to identify the micro-scale organizations of discourse, and to describe the patterns in these organizations, as the key to exploring human communication in the broader sense. Hymes’ (1962,1974) categorization of communication into definable units in a hierarchical order; have established the first basis for a methodology enabling the researcher to explore these patterns. The definable units of communication includes; a) speech situation: the social occasion in which speech occurs, (e.g. a party) b) speech 12 event: activities or aspects of activities that are directly governed by rules and norms of the use of speech (e.g. a conversation in the party), c) speech act: acts that can be identified by their illocutionary force (e.g. a joke during the conversation). Speech events, according to Hymes (1974:52), are the largest units for which one can discover linguistic structures, and the term speech event is “...restricted to activities or aspect of activities that are directly governed by rules or norms for use of speech.”. Hymes (1962:132) also identified the speech event as the “primary object of analysis” and proposed a classificatory scale known as the SPEAKING Grid. Each letter in the acronym S-P-E-A-K-I-N-G corresponds to one possible component of the speech event to be analyzed; i.e.; Setting, Participants, Ends, Act- sequence, Key, Instrumentalities, Norms and Genre. (see Appendix for the details). In this study, we will use the SPEAKING Grid for analyzing the non-verbal (cultural) components of the context shaping the verbal behavior. Regarding the discovery of the interactional structures in the ??z isteme event, however, we benefit from the method of Move Analysis following the insights in some recent work (e.g Motta-Roth,1998; De Carvalho 2001; Suarez, 2005). Though this method has commonly been used in studying rhetorical structures in written genres, we find it relevant for this study since the act- sequence models fall short in describing the interactional patterns in the event. 3. Data Analysis Towards a detailed analysis of the contexts of communicative event, we have used the SPEAKING Grid, for the initial stage to identify the verbal and non-verbal components of each conversational text. In section 3.1 we have tried to present the general characteristics of each component based on our analyses, observations, and on our knowledge of the culture as a members of the speech community, as well as on the insights gained from previous research, rather than describing specific cases as generally done in ethnographic research (eg. Nwoye, 1985; 13 Schiffrin,1994). In section 3.2, boundary markers for the ??z isteme event have been presented and illustrated through examples. Finally, as the main focus of the study, we have described the interactional patterns through a model which represents the potential generic move structures of the ??z isteme speech event. (table, 3.3.1) and illustrated it on a sample text. 3.1 Communicative Components of k?z isteme speech event. (S) Setting: K?z isteme visits generally take place in the girl’s house, during the evening; however, in some western parts of the country, other locations, such as the village guest house, may be used, especially for larger gatherings. The physical appearance of the household is also significant in that, it reflects the positive or negative attitudes of the family towards the event. (P)Participants: While in some households a larger body of company, such as, close relatives and friends, is expected along with the family members during the occasion, some events (especially in modern settings) may be realized only with the presence of the families on both sides. It is also, common, however, to have some third parties in the company, such as, a senior in the family or a reputable friend. These people may undertake the role of the ‘spokesperson’, to speak on family’s behalf while making the request for the girl’s hand during the occasion. This role is otherwise, performed by the father of the boy, or by an uncle, an elder brother or the mother in the absence of the father. (E)Ends: The purpose of the ??z isteme event is to take the consent of the girl’s family for the marriage i.e., to persuade the girl’s family to this end. (A) Act-sequence: -SPB:* asks for the consent of the girl’s family for the hand - SPG replies (signaling consent or suspending the proposal ) -Families make projections for a future gathering. * SPB: Spokeperson on the boy’s side SPG: Spokeperson on the girl’s side (K) Key: The tone of the discourse is politic and serious. However, speakers occasionally make use of humor to lower the tension in the atmosphere. 14 (I) Instrumentalities: The language employed by the participants during the event is highly formal and ritualistic. For example, using the formulaic expression, “allah?n emri ve peygamberin kavliyle” (with the command of the God and the word of the prophet) is an unmarked linguistic component of the request to be made for the girl’s hand. Non-verbal conduct is also effectively used as indirectness is often a prevalent feature of the communication. For instance, a cup of tasteless coffee served by the girl may convey a negative message. (N) Norms for interpretation: Although certain verbal and non-verbal codes have been conventionalized, details of the ??z isteme procedures may vary across different households, e.g., the young couple may freely participate in the ongoing discourse in love (couple-initiated) marriages, and the communication is relatively less indirect whereas in traditional contexts more strict conventions are at work. For instance, it is not customary for the girl’s family to express overt consent for the marriage during the first visit, even if they are willing to. The consent is usually suspended through some accounts such as the ones about the need for negotiating the matter within the family. Negative response, on the other hand, is made through excuses e.g., related to the young age of the girl or to her education etc. 3.2 Marking the boundaries Since communicative events are bounded entities, recognizing the boundaries of an event is essential for their identification. Saville-Troike (1997:141) suggests that the beginning and ending of a communicative event is signaled by means of verbal and non-verbal cues or by totally physical occurrences; just as a telephone conversation is bounded by a ring of the telephone and hanging up the receiver as a close. Some other examples identified by Saville-Troike on natural data are, phrasal expressions (e.g. ‘once upon a time’ and ‘happily ever after’ for a story), changes in gaze, bodily positions, and in linguistic code or style. Similar boundary markers cueing the opening and the closing of the ??z isteme event have been detected in this study, as presented below: 15 3.2.1. Boundary Markers for the beginning The beginning of the ??z isteme event in our data was distinguished from the preceding discourse via the following signals. a) non-verbal signals: Recognizable silence, gaze direction towards the spokesperson of the girl’s family, change in posture –alert and slightly bent as a sign of modesty, body movements: (e.g. sitting, standing, using gestures) and extra-linguistic signals such as changes in the pace and volume of the speech ( the pace slows down and the volume becomes higher). b) verbal signals: Alongside with the non linguistic signals, the event boundaries are marked with discourse-markers (Schiffrin,1994) such as attention getters, (eg. ?imdi ‘now’ ), and gap fillers, such as ????, ‘eer’, tabi ‘well’. In our data, direct address forms (i.e, names) preceded by honorifics such as say?n ‘distinguished’ or titles such as profösör ‘professor’ were also employed by the spokespersons while initiating the talk. Another prominent boundary marker identified in the data was topic markers securing a clear change in the topic. We have also observed that multiple cues were used simultaneously as shown in the excerpts below. To illustrate, in excerpt (1) the spokesperson remains silent for two seconds before he initiates the talk following a joke that he has been telling. This causes a sound and relatively tense silence in the room since it is probably interpreted by everyone as a signal for the expected ‘request’. Some changes in the posture (erect) and gaze of the SP (towards the girl’s grandfather) were also observed. Finally, the verbal cue, which is a direct address to the girl’s family in a respectful manner, followed by the discourse marker, ’now’ clearly signals the beginning of the event. Also see excerpts (2), (3) and (4). (1) (silence, smile, gaze towards the senior of the girl’s family, posture erect) SP: (..) ?imdi efendim. “Now sir,” [summoner + term of addres ] ( 2) (recognizable silence, body slightly bent forward, hands on the knee) SP: eveet (.. ) esas konuya gelirsek.. “well (...) as for the main topic...” [gap filler+topic marker] 16 (3) (standing, louder tone of voice) SP: Say?n Nur han’fendi, say?n Seyhan bey “Honorable lady (Nur) and sir (Seyhan)” (honorifics+ term of adress) (4) (posture slightly bent, slow pace of speech, gaze) SP: ??? Pa?am, ziyaret sebebimiz “eer, colonal, the very reason for our visit is...” [gap filler+ term of adress +topic marker] Apart from these linguistic and non linguistic cues, the shift into the new speech event is characterized by considerable changes in manners/attitudes of the other participants, their role relationships, the psychological atmosphere and the characteristics of the language used etc.(discussed in section 3.1) For example, the light-hearted tone in the preceding discourse may turn into a more serious one, while the speaker’s role changes into a SP from that of a close friend. Also, the rights and obligations of speakers to hold the floor seem to be directly determined by the conventions of the event. 3.2.2 Terminal Boundary Markers In our data, we have observed that the k?z isteme event is terminated by an exchange following the response of the SP from the girl’s side to the request made by the boy’s family. Provided that the response has a positive implication, the conclusion contains a high degree of appreciation and respect from the boy’s side. In cases where the response is suspended in a relatively negative manner, the SP uses aggravating expressions while repeating the request and expressions of hope for a positive answer (see excerpts 5 and 6). (5) SPB: ?lerde iyi ili?kilerimiz olaca???? gençlerle de dost olaca?????? umuyoruz “We wish that we have good relations in the future, the children and we become good friends”. SPG: In?allah “we wish the same”* 17 *Even though the idiomatic expression ‘in?allah’ meaning ‘with the will of the God’ has a religious connotation, it is deliberately used in casual discourse to express positive intentions for a future act. (6) SPB: Art?k sizden haber bekliycez, en k?sa zamanda in?allah görü?meyi diliyoruz. “we will look forward to hearing from you, we hope to meet again soon” SPG: Tabii ki, in?allah. Bizim için de çok önemli bir konu. “Yes indeed. It is a vital topic for us too” 3.3 The interactional patterns As the main focus of this study, we have further analyzed the interactional structure of the sözkesme speech event which has been initially presented in Section 3. We have observed two typical acts crucial to the realization of the event: 1) SP on boy’s side makes an explicit request for the hand of the girl, 2) the SP of the girl’s family suspends the response for the request by normally signaling positive cues or by providing an indirect negative answer. However, we have also observed that speakers perform a series of other acts, some of which are optional, to ensure a more socially appropriate interaction, or for some other rhetorical purposes. Shortly, we have identified the moves realized by the acts in this speech event, through an inductive method, rather than checking the predetermined categories on the data (Suarez, 2005), and we have proposed a five-step move model, which we suggest to represent the potential generic structure of the ??z isteme speech event, as shown in 3.3.1 below: 18 Table 3.3.1 The Move Sequence of ??z isteme Event ________________________________ Preparatory move Head move (Supportive move)* Head move 2 (Supportive move)* Complementary move ________________________________ *optional As the first component of the model, the Preparatory move performed by the spokesperson on the boy’s side, is composed of at least one main act which is the ‘expression of the intention’, that is usually enveloped by some potential acts such as ‘appraisals’ and ‘an account’ of the events leading to that very occasion (Table 3.3.2). Table 3.3.2 Preparatory move Acts Participants SPG SPB Exchanging compliments + + Giving account + Expressing intention + In this section, the acts detected in this study have been illustrated on excerpts from the data. Participants have been coded based on their roles during the speech event rather than as particular persons and expressed in abbreviated forms in the following manner: Spokesperson on the girl’s side (SPG), boy’s father (BF), boy’s mother (BM), girl’s father (GF), girl’s mother (GM). (see excerpts7,8,9) . 19 (7) Complimenting: SPB: Efendim, sizlerle olmak, ??, sizleri tan?maktan çok mutluyuz. “We are very happy to know you and to be here with you today” [complimenting] SP G: .?eref verdiniz . Bizim için de öyle Oh, it’s our honor., too. [complimenting ] (8) Expressing the intention (of the visit): SPB : Bugün sizlerle akraba olma niyetimizi ifade etmek için burday?z “We are here to express our intention to be relatives with you”. [expressing intention ] (9) Giving an account: Gençler kendi aralar?nda anla????lar evlili?e niyet etmi?ler.Okul ??llar?nda ba?layan bir arkada???klar? var. Biz de bunu resmile?tirmek istiyoruz. “The youngsters have already an agreement on this matter. They have been friends since the school years. And we’d like to legitimize it”. [giving an account] Table 3.3.3 Head move 1 Act(s) Participant SPB Asking for the hand of the girl + The Head Move I is realized by a single act which involves ‘asking for the consent for the girl’s hand’ (Table3.3.3). The formulaic discourse is used while making the request as seen in the excerpt below: (10) Asking for the hand of the girl 20 SP: E?im ve amcamla birlikte, geleneksel ifademizle, allah?n emir ve peygamberim kavliyle sevgili k?????z Zeynep’i o?lumuz Kaan’a istiyoruz.. Sizlerin izninizle ve Allah yazd?ysa. “Together with my wife, the uncle and the friends, having visited you and let us put it in traditional words we ask your consent for the hand of your daughter Zeynep, for our son Kaan, with the command of God and the words of the Prophet, if the God would allow us and you give permission”. [asking for the hand of the girl] Table 3.3.4 Supportive move Acts Participants SPG SPB Praising the couple + Expressing hopes and good + + wishes for the future An old family friend (‘uncle’ as the young ones call him) in Excerpt 11, tries to reinforce the request that has been just issued by the SPB by means of compliments, and expressesions of hopes for a good future, and with a final restatement of the request. The request, however, is kindly and shortly uptaken by the girl’s mother and by the SPG with no further comment in excerpts, 12 and 13. (11) Uncle: Zeynep’i çocuklu?undan beri tan???m. Halil bey ve ailesi de çok be?endi?imiz ve sayg? duydu?umuz insanlard?r. Bu evlili?in olmas??? çok gönülden arzu ederiz. “I have known Zeynep since her childhood. She is like our daughter. And Halil Bey’s family (the groom’s family) are also people we like and respect. I wish from the bottom of my heart that this marriage be realized, we’d be very happy.” [compliments] 21 (12) GM: Tabi bizler de “Sure, we would (feel the same), too”. [compliments] (13) SPB: ?lerde iyi ili?kilerimiz olaca???? gençlerle de dost olaca?????? umuyoruz “We hope that we will have good relations in the future, the children and we become good friends”. [expressing hope] (14) SPG: In?allah “we wish the same” [expressing hope] Table 3.3.5 Head Move II Acts Participant Suspending the response SPB a) by signalling consent b) by signalling disapproval Our investigation has shown that the response of the girl’s family may take two possible directions; one is toward a positive result where they signal consent. The second option is to imply that their response will be negative. In Excerpt 15 the implicature of the SP’s talk is that the family’s attitude on the girl’s side will be positive about the proposal, however they need some time to think over it. The other party will take this delayed response as a pre- accept, as a part of their knowledge of the cultural norms. (15) SPG: Sa?olun. Sa?olun. Ee, biz de allah?n izniye, olumlu bakaca??z. Yaln?z, takdir edersiniz aile büyüklerimizle konuyu bir görü?üp dan??mam?z gerekiyor. “Thank you. Thank you. We, eer, as how the god allows it to be. We, we will consider it positively (.) as you would appreciate we 22 have some elderly in the family that we should inform of this visit and consult” [delaying the response/ implying consent] Table 3.3.6 Supportive move 2* Acts Participants SPG SPB Praising the boy /girl + + Restating the request + Expressing hopes and good + wishes for the future *optional The supportive move shown in Table 3.3.6. is an optional category since the groom’s family tends to provide a series of appraisals, hopes and wishes etc. when they sense the message of pre-accept, or hesitations (Excerpts 16,17). However, in the case of receiving rejection signals from the girl’s side, the supportive move is neglected, while shifting into the complementary (closing) acts. (16) SPG: Bize umut verdiniz. Hanfendi, hay?rl? olsun diyebilir miyiz? Oh, this, eer, you have given us hope. (addressing the girls mother) Can we, madam, say “hay?rl? olsun” (an idiomatic expression that bear good wishes on starting something new or on giving a decision). [restating the request/insist] GM: Ee, ne diyelim, hay?rl??? olsun. “Eer, well, what should I say. I wish the best of whatever may happen.” [delaying the response/ pre-accept/implying consent] GF: Asl?nda biz de Ahmet’i o?lumuz gibi seviyoruz. Ama konuyu ailemizle görü?mek istiyoruz. 23 “In fact, we know Ahmet (the groom) well, like him as our son. Yet, you understand, we have to talk with the elderly” [delaying the response/ pre-accept/implying consent] Table 3.3.7 Complementary move Acts Participants SPG SPB Exchanging compliments + + Expressing intention for future visit + + ( Ratifying the future visit) + (Suspending the future visit) + In the the complementary closing, we have always noticed compliments, wishes of well being and often mentioning of the boy’s family a next visit to the girl’s home, as illustrated in the excerpt below: (17) GF: Tabii, tabii, gelenektendir. Sayg? duyar?z. ?imdi konuyu burada ??rak?yoruz, sizden gelecek iyi haberi bekliyoruz. ???allah. O zaman sizi tekrar ziyaret ederiz. “sure, sure, it’s the customs. We respect that (pause). Eer, we leave the matter here with our hopes and good wishes and wait for your informing us. Insallah (if god wills), we visit you once more” [expressing hopes for the next meeting] GF: In?allah [expressing hopes for the next meeting] GM: In?allah “we wish the same” 4. Conclusion In this study we have tried to explore the structure of ??z isteme ceremony as a part of the traditional marriage–arrangement discourse in Turkish culture. Treating the ??z 24 isteme ceremony as an individual speech event, we have described the nonverbal components of the context, i.e, how language use is constrained by cultural norms like any other type of human behavior. It was also interesting to observe how the language might become almost formulaic to the extent that the event is conventionalized. The close analysis of the four conversational texts has revealed that the sözkesme speech event is realized through a small number of acts that are tightly patterned. This feature of the event is reflected in the generic move-sequence model proposed in the study. The highly formal and ritualized language use we have come across in this study supports the findings reported in similar studies in some other cultures, (eg.Nwoye ,1985; Tannen & Saville-Troike,1985). Additionaly, we have observed that the ??z isteme event shares some common characteristics with the betrothal event investigated in Çubukçu (in press) regarding some aspects of the contextual features and the types of acts performed, since these events constitute two different stages of the same process. Also, the formulaic expression “allah?n emri ve peygamberin kavliyle” takes place in both events, in that the accomplishment of the betrothal necessitates the explicit and formal request made for the girl’s hand. However, the basic difference between these speech events lies in their purpose which is regarded as the main distinguishing feature between genres (Swales, 1990). To specify, whereas the betrothal ensures the formal agreement between the families, ??z isteme event is a part of the process of persuading the girl’s family. Therefore, given that the betrothal ceremony involves the announcement of the happy occasion (with special desserts, drinks and gifts, etc.), finalizing (sometimes) a long lasting process it naturally effects some changes in the physical and psychological features of the context (e.g, the key, instrumentalities, participants etc) as well as in the act sequence. Shortly, we will call them as twin speech events, but which are, by no means, identical. That is, taking into consideration the limited scope of our data and also the existence of the large variety of possible practices within the culture, it should be noted that the model proposed in this study is a tentative one, as implied by its title. We suggest that similar studies on the speech community’s practices be carried out for stronger conclusions about the ritualistic events, and more importantly, for a hope 25 towards better uncovering the far more complicated structures of the simple everyday talk. References Alagözlü, N. 2007. Bir söylem türü olarak kahve fal?na yönelik bir inceleme: kahvefal? ideolojisi. Edebiyat Fakülte Dergisi.V24,1,(Haziran), 1-22. Applebaum, K.D. 1995. Marriage With the Proper Stranger: Arranged Marriage in Metropolitan Japan. Ethnology, 34:1, 167-198. Bhatia, A. 2006. A critical discourse analysis of press conferences: Discourse and society. Sage Publications Bakhtin, M.M. 2005. The problem of speech genres. In Discourse Reader A. Coupland, & N. Jaworsky (eds), 121-133. New York: Routledge. Bauman, R. & Sherzer,J. (eds.) 1974. Explorations in th Ethnography of Speaking. London: Cambridge University Press. Chilton, P. &Schaffer, C.1997. Discourse and Politics. Sage Publications. Coulthard, M.1988. An introduction to discourse Analysis. London: Longman. Çubukçu, H. 2007. The discourse of betrothal in Turkish culture. Hacettepe University Journal of Faculty of Letters, December 2007, 45-62. De Carvalho, G. 2001. Rhetorical patterns of academic books reviews written in Portuguese and in English. In Iglesis L., Rabade & S.M. Doval Suarez (Eds.). Studies in contrastive Linguistics, 261-268. Demircan,Ö. 1990. Yabanc?-Dil Ö?retim Yöntemleri. ?stanbul: Ekin E?itim Duben,A.& Behar,C. 1991. Istanbul Households: Marriage, Family and Fertility: 1880-1940. Cambridge: Cambridge University Press. Greenblatt, S.L. 1979. Chinese Fortune-Telling in Action: Humanist, 39:3, 21-28. Gumperz, J. & Hymes, D. 1986. Directions in Sociolinguistics. Oxford: Blackwell. Hasan, R. 1978. Texts in systematic functional model. In: Current Trends in Textlinguistics. W.V. Dressler (ed.). Berlin: De Gruyter. Holtgraves, T. &Lasky,B. 1999. Linguistic Power and Persuasion. Journal of Language and Social Psychology. 18:2, 196-205. Hortacsu, N. 1997. Family and Couple Initiated Marriages in Turkey. Genetic, Social And General Psychology Monographs, 123:3, 325-354. Hill, J.H & Hill,K.C. 1986. Speaking Mexicano. Tuscon: University of Arizona Press. Hymes, D. 1962. The ethnography of speaking. In Glandunt, T. & Sturtevant, W. (eds.) Anthropology And Human Behavior, Washington, D.C.: The Anthropology Society of Washington, 13-53. Hymes, D. 1974. Foundations in Sociolinguistics: An Ethnographic approach. Philadelphia: University of Pennsylvania Press. Hymes, D. H. 2000. 21. Toward Kawaiisu Poetics. In Functional Approaches to Language, Culture and Cognition. David G., Peter H. Fries and James E. Copeland (eds.), Lockwood, 369-389. Milroy, L .1987. Language and society networks. Second Edition. Oxford: Basil Blackwell. 26 Motta-Roth, D.1998. Discourse analysis and academic book reviews: a study of text and disciplinary cultures. In Genre studies in English for academic purposes (29-59). Fortanet Gómez I., S. Postaguillo, J.C.Palmer & J.F Coll (Eds.) Castelló de la Plana: Publicacions de la Universitat Jaume I. Nwoye, G. O. 1985. The eloquent silence among the Igbo of Nigeria. In: Perpectives on silence. D. Tannen & M.Saville-Troike (eds.), 185-190. Norwood, NJ: Ablex. Nwoye, G. O. 1993. An Ethnographic Analysis of Igbo Greeting. Afr?can Languages And Cultures, Vol. 6, 1:37-48. Philipsen, G. & Coutu, L. M. 2005.The ethnography of speaking. In Handbook of Research on Language ad Social Interaction, R. Sanders & K. Fitch (Eds.), 355-379. Mahwah, NJ: Lawrence Erlbaum. Saville-Troike,M.1997. The analysis of communicative events. In Sociolinguistics: A Reader. A. Coupland & N. Jaworsky (eds), 126-145. New York: St. Martin’s Press. Saville-Troike,M. 2003. Ethnography of communication: An Introduction. 3rd edition. Oxford: Blackwell Publishers. Schiffrin, D. 1994. Approaches to Discourse. Cambridge, Massachusetts: Basil Blackwell. Sherzer, J. 1983. Kuna ways of speaking. Austin: University of Texas Press. Stirling, P. 1961. Turkish Village. The Nature Of Human Society. Centre for Social Anthropology and Computing: Kent University Publications. Suarez-Tejerina, L. 2003. Is evaluation structure bound in English and Spanish book reviews? In Strategies in academis discourse. E. Tognini-Bonelli & G.D.L. Camiotti. (116-132). Amsterdam: Benjamin’s. Suter, E. A., 2000. Focus Groups in Ethnography of Communication:Expanding Topics of Inquiry Beyond Participant Observation. The Qualitative Report, V 5, N 1 & 2, (May), 18-36. Tannen, D. & Saville-Troike,M. (eds.)1985. Perspectives on Silence. Norwood, N.H: Ablex Publishing. Timur, S. (1972). Türkiyede Aile Yap???. Ankara: Hacettepe Üniversitesi Yay?nlar?. D-15. Ventola, E. 1979. The structure of casual conversation in English. Journal of Pragmatics,V3, 3-4 Aug, 267-298. Wardough, R. (1997). An introduction to sociolinguistics. Oxford: Blackwell Publishers. 27 APPENDIX The SPEAKING Grid (Wardough, 1997). The SPEAKING Grid which is used as the tool for analysis in this study helps to uncover potential verbal and nonverbal features of distinct speech events, and also to understand how these features are related to one another. Each letter in the acronym S-P-E-A-K-I-N-G stands for one possible component of the communication to be analyzed (Wardaugh,1997: 245-247): (S) Setting/scene: While setting refers to physical circumstances eg. time, place in which the event takes place, scene refers to psychological/ abstract settings. For instance, the scene of the president’s New Year message would largely differ from the scene of a message following a disaster that hits the country. (P) Participants involve various combinations of speaker-listener, sender-receiver, addresser-addressee, etc., and they usually fill socially specified roles. E.g. a prayer makes a deity participant. Or, in a classroom setting, the teacher’s question followed by a student’s answer does not only involve the two interactants, but it assigns the role of ‘attentive listeners’ to the rest of the class. (E) Ends refers to the conventionally recognized and expected outcomes or purposes of the event but it also includes specific goals of the interactants. E.g. a trial in a courtroom has a recognizable social end but different participants (eg. defendants or prosecutors) also have their own goals. (A) Act-sequence refers to the actual form and content of what is said. This part is the basic aspect of speech events with which linguists interest themselves (as in this study). (K) Key refers to the tone and manner or spirit in which a particular message is conveyed. Some examples may be light-hearted, serious, sarcastic etc. (I) Instrumentalities has to do with the channel (verbal/non-verbal), physical forms of speech, written or oral etc. A single event may involve using multiple instrumentalities; e.g., oral and written modes, or switches across languages. (N) Norms of interaction and interpretation refers to specific properties attached to speaking concerning cultural beliefs, attitudes and styles by the speech community . (G) Genre refers to clearly categorized textual categories, such as, poems, jokes, sermons, lectures. 28 Expressing social distance in Polish discourse - the use of the grammatical category of person Miros?awa Kaczmarek (Tokyo University of Foreign Studies) 1. Introduction The problem of meaning in linguistics can be approached from two different perspectives. In cognitive linguistics it is usually associated with the idea of some external object or sign that language users have in mind. This position entails a pre-given natural world, which is represented to the mind as concepts or mental representations. In pragmatics and systemic-functional linguistics on the other hand, the problem of meaning is approached from a different position. It is claimed that meaning does not correspond to the extra-linguistic conditions, but it is situated in the internal context of words. In other words, it is created either through the existing oppositions between neighboring linguistic forms in a sentence, or in the processes of an utterance exchange conducted by speakers in the domain of discourse. This study adopts the second perspective on meaning. The definition of meaning is taken from Halliday (1994) and systemic-functional linguistics (Eggins 2004), who state: (…) language is structured to make three main kinds of meanings simultaneously. This semantic complexity, which allows ideational, interpersonal and textual meanings to be fused together in linguistic units, is possible because language is a semiotic system, a conventionalized coding system, organized as a set of choices. The distinctive feature of semiotic systems is that each choice in the system acquires its meanings against the background of the other choices, which could have been made. This semiotic interpretation of the system of language allows us to consider the appropriacy or inappropriacy of different linguistic choices in relation to their context of use, and to view language as a resource, which we use by choosing to make meanings in contexts. (Eggins 2004: 3) 29 The objective of this paper is to reconsider linguistic resources that are used for the expression and regulation of social distance in Polish. The focus of the analysis lies in combining the use of the first and second person verb morphemes with such contextual parameters of interaction as the opposition between formal (teacher-student interaction) and informal (acquainted female conversation) types of speech. Brown&Levinson (1987) claim that politeness principles as for instance, the regulation of socio-psychological distance between speech participants are inevitably encoded in language. They suggest moreover that these principles should be possible to reconstruct from certain recurring patterns of language use, like for instance certain recurring forms of the use of grammar. This paper suggests that the use of the first and second person verb forms in Polish discourse belongs to such recurring patterns of language use and can therefore reflect the interdependence of the extra-linguistic context of interaction, as for instance the distinction between formal and informal types of speech. 2. Social distance and the grammatical category of person The systemic-functional approach to language originated and was developed on the foundation of work done by the social semiotic linguist Michael Halliday (1978). One of the things about language that Halliday was mostly interested in was, how people use language with each other in order to accomplish everyday social life. This interest led him and his colleagues to advance four main theoretical claims about language, that language use is functional, semantic, contextual and semiotic. Halliday (1994) understood the functional character of language use as a meaning – making device. He also claimed that language meanings are influenced by the social and cultural context in which they are exchanged. He stated moreover that the process of using language is a semiotic process, a process of making meanings by choosing. This study aspires to reconsider the above four claims about language raised by Halliday (1994) by applying them to the use of the first and second 30 person verb morphemes in the Polish interaction data. The use of these morphemes is analyzed from two perspectives. A sentence-level perspective is defined as either the absence or presence of personal verb morphemes within the boundary of one single utterance. The speaker therefore has three choices. He can choose second person verb form. He can choose no second, only first person verb form or he can choose neither second nor first person verb form. A discourse-level perspective on the other hand focuses on two neighboring utterances uttered by different interlocutors that contain either first or second person verb morphemes. 2.1. Expressing meaning of social distance from the perspective of systemic functional linguistics This study presumes that the expression of meaning of social distance takes place in the context of words or morphemes that are frequently exchanged by speech participants. It takes the perspective that interpersonal reality is constructed through the oppositions encoded in the semiotic system of a natural language. 2.1.1. Language as a meaning-making device Considering the notion of meaning, systemic-functional linguists (Eggins 2004) argue that language is structured to make three main kinds of meanings simultaneously. Ideational meanings are meanings about how the experience is represented in a language. Interpersonal meanings are meanings about role relationships with other people and people’s attitudes to each other. Textual meanings on the other hand constitute meanings about cohesion and coherence in texts, i.e. how what people are saying hangs together and relates to what was said before and to the surrounding context. This study presumes that the meanings created through the use of the first and second person verb morphemes in Polish conversations are both of interpersonal and textual kind. They are of an interpersonal nature, because they point to individual speakers and hearers. They are also of a textual nature, because they are indispensable in everyday conversation contributing both to cohesion 31 (relation between text and internal context) and coherence (relation between text and external context) of texts. 2.1.2. Language as a semiotic system Going back to the four claims stated about language use, it is worthwhile to note that the three main kinds of meaning, i.e. ideational, interpersonal and textual are created simultaneously in language. This semantic feature of texts is possible because language is a semiotic system organized as a set of choices. A potential set of choices constitutes a paradigm, which can be defined after Cysouw (2003: 9) as a “closed set of alternative options”. According to Cysouw (2003) individual elements in a paradigm do not arrive at their referential value intrinsically, but in mutual delimitation vis-à-vis the other elements. Therefore, actual choices of linguistic elements acquire their meanings against the background of potential choices of linguistic elements, which could have been made. For instance, when the potential of choices for a greeting formula between unacquainted people is Good morning (more formal), Hello (less formal) and Hi (rather casual), the choice of a Good morning will acquire a meaning of speaker’s intention to confirm a rather formal atmosphere to the encounter. This study analyzes the potential of choices speakers and hearers have to their disposal when using the first and second person verb morphemes in Polish conversation. It is presumed that speakers by choosing first person verb forms over second person verb forms will contribute to the creation of socio- psychological distance between interlocutors. The choice of a second person verb form over a first person verb form on the other hand, will acquire a meaning or reveal the speaker’s intention to reduce the socio-psychological distance between interlocutors. 2.1.3. Language use as a reflection of context The human ability to deduce context from text, to predict when and how language use will vary, and the ambiguity of language removed from its context 32 provide evidence that when speaking about language use one must focus on language use in context. However, since not every aspect of the context (for example the color of the hearer’s hair) has an impact on language use, systemic-functional linguists inquire in the aspects of language use that may be affected by the particular dimensions of context. For example, when one contrasts texts where speakers are good friends with texts where the speakers are not acquainted with each other, how can one tell where in the language they use this contextual difference is expressed? One of the mostly discussed levels of the linguistic context is the level of register. Register describes the impact of the immediate context of situation of a language event on the way language is used. Three register variables (Halliday 1978) such as mode (the adopted channel of communication), tenor (role relations of power and solidarity between interlocutors) and field (topic or focus of activity) are believed to have a significant and predictable impact on language use. This study has focused its attention on the variable of tenor. It is presumed that the patterns of exchanging the first and second person verb morphemes by speakers in Polish conversations have a great influence on the regulation of the socio-psychological distance between interlocutors. 2.2. The grammatical category of person in Polish It’s generally taken for granted that language is an interactional device with two speech act participants who organize their speech in a turn-taking sequence. The speaker/addressee roles in the turn-taking system are prototypically reflected in the structure of language, but recent research calls into question the idea that all languages possess grammatical category of person and abide to the principles of language use that it imposes on speech participants. Polish belongs to the Indo-European family of languages. One of the distinctive features that the Indo-European languages such as English, German, French, Italian, Polish, Czech and Russian have in common is that they possess grammatical category of person. The grammatical category of person can 33 manifest itself in the use of personal pronouns and verbal morphemes. Its use is obligatory in everyday conversation and constitutes a paradigm. According to Cysouw (2003) a paradigm can be interpreted as a set of alternative options that a language speaker has at his/her disposal when using language. In Polish for instance, a speaker has the following set of personal verb form options that constitute a paradigm while talking about present, future or past affairs. It is worthwhile to note here that depending on the speaker’s gender the paradigm will slightly differ when talking about past affairs. Table 1: Personal verb form paradigm in Polish (present tense) Singular Plural 1. (ja) jestem - I am 1. (my) jeste?my - We are 2. (ty) jeste? - You are 2. (wy) jeste?cie - You are 3. on jest - He is ona jest - She is ono jest - It is 3. oni s? - They are (male; male&female) one s? - They are (female) Table 2: Personal verb form paradigm in Polish (future tense) Singular Plural 1. (ja) b??? - I will be 1. (my) b?dziemy - We will be 2. (ty) b?dziesz - You will be 2. (wy) b?dziecie - You will be 3. on b?dzie - He will be ona b?dzie - She will be ono b?dzie - It will be 3. oni b??? - They will be (male; male&female) one b??? - They will be (female) Table 3: Personal verb form paradigm in Polish (past tense) Singular Plural 1. (ja) by?em - I was (male) (ja) by?am - I was (female) 1. (my) byli?my - We were (male; male&female) (my) by???my - We were (female) 2. (ty) by??? - You were (male) (ty) by??? - You were (female) 2. (wy) byli?cie - You were (male; male&female) (wy) by???cie - You were (female) 34 3. on by? - He was ona by?a - She was ono by?o - It was 3. oni byli - They were (male; male&female) one by?y - They were (female) 2.2.1. Benveniste (1971): The three persons Benveniste (1971) discerns three persons that form a separate grammatical category in Indo-European languages. His interest however, lies not in the lexical differences that distinguish the first, second and third person, because these differences do not indicate the necessity of the category. He is interested in the principles that the opposition between the three persons is based on. He points to the fact that there is a clear disparity between the first and second persons and the third. He maintains that concerning the first two persons there are both a person and a discourse (speech context) related to that person. He makes the following statement about the first and second persons: “I” designates the one who speaks and at the same time implies an utterance about “I”; in saying “I”, I cannot not be speaking of myself. In the second person “you” is necessarily designated by “I” and cannot be thought of outside a situation set up by starting with “I”; and at the same time “I” states something as the predicate of “you”. (Benveniste 1971: 197) The first and second persons are therefore closely interrelated. He states that “you” is more dependent on “I” than “I” is dependent on “you”. This perspective is also adopted in the present study while stating that when the first and second person verb morphems appear together in one utterance, only the second person will be coded, because his/her presence is regarded as more marked than the first person. Regarding the third person, Benveniste (1971) claims the following: But in the third person a predicate is really stated, only it is outside “I-you”; this form is thus an exception to the relationship by which “I” and “you” are specified. Consequently, the legitimacy of this form as a “person” is to be questioned. (Benveniste 1971: 197) 35 In other words, as Benveniste (1971) states later, the third person is not a person. It is called a non-person, because it lies outside of the relationship designated by “I” and “you”. All utterances containing neither first nor second person verb morphemes in the present study have been coded for this reason as non-marked (NM). 2.2.2. Jespersen (1922) & Jakobson (1984): “I” and “you” as shifters The first scholar who discovered that the personal pronouns like “I” and “you” belong to the class of shifters was Jespersen (1922). He noticed that they are very difficult to acquire for small children. Consequently, Jakobson (1984) in recalling the study of Burks on Peirce’s classification of signs into symbols, indices and icons, stressed the fact that shifters combine the functions of both indices and symbols and therefore belong to the class of indexical symbols. One of the distinctive features of indexical symbols is that they cannot represent an object without being in existential relation (function of an index) with it by a conventional rule (function of a symbol). In Polish, for instance, the first person pronoun “ja” plays a function of a conventional symbol and on the other hand by designating the speaker and being existentially related to his/her utterance plays a function of an index. Jakobson (1984) points out the following facts: In language and in the use of language, duplicity plays a cardinal role. In particular, the classification of grammatical, and especially verbal categories requires a consistent discrimination of shifters. (Jakobson 1984: 44) (…) one of the two mutually opposite grammatical categories is “marked” while the other is “unmarked”. The general meaning of a marked category states the presence of a certain (whether positive or negative) property A; the general meaning of the corresponding unmarked category states nothing about the presence of A, and is used chiefly, but not exclusively to indicate the absence of A. (Jakobson 1984: 47) In the present study, which is based on the assumption that interpersonal meanings of social distance arise through oppositions, the first type of opposition 36 or as formulated above the first type of duplicity is the distinction between those utterances that contain shifters and those utterances that do not contain shifters. This corresponds exactly to the distinction suggested by Benveniste (1971) between utterances that contain the “I-you” dyad (shifters) and utterances that do not contain the “I-you” dyad (non-shifters), i.e. those utterances that either contain or do not contain the first or second person markers. The second type of duplicity is related to utterances containing the “I-you” dyad, where the first person verb form represents the unmarked type of utterances while the second person verb form represents the marked type of utterances. Accordingly, this paper states that interpersonal meanings expressing social distance in Polish arise in oppositions between those linguistic forms that belong to the class of shifters, i.e. the first and second person verb morphems. This statement is based on the reasoning that the indexical character of a shifter has direct impact on the construction of interpersonal meanings between interlocutors. 2.2.3. Siewierska (2004): Discourse roles of the speaker and addressee Siewierska (2004) referring to person as a grammatical category makes a critical comment concerning the notions of the first and second person. She states that in the case of the first and second person, the grammatical category of person does not simply express the speaker and addressee, but rather the participant or discourse roles of speaker and addressee. She claims moreover that such an understanding of the grammatical category of person draws on the origin of the term person, i.e. mask. In other words, she implies that the use of the first and second person forms in conversation does not directly indicate the presence of a speaker or addressee, but points to the socio-psychological relations that exist between them, i.e. their discourse or social roles. This study aspires to reveal the relationship that is presumed to exist between the frequency of using the first and second person verb morphemes in real speech and the process of creating the extra-linguistic context in conversation. One may also put the same problem the other way and inquire about the influence the extra-linguistic context, such as for instance the formal and informal type of 37 speech has on the frequency of using the first and second person verb morphems by speech participants. 3. Hypothesis Brown&Levinson (1987) presume that every social encounter is threatening to the participants’ face and speech participants need to undertake certain measures in order to protect face. Acts threatening to speakers’ face are called Face Threatening Acts (FTAs). Brown&Levinson(1987) argue that the assessment of the seriousness of an FTA involves the following three factors. (i) the ‘social distance’ (D) of S and H (a symmetric relation) (ii) the relative ‘power’ (P) of S and H (an asymmetric relation) (iii) the absolute ranking (R) of impositions in the particular culture. (Brown&Levinson1987: 74) This paper adopts the first factor (i) for the analysis of Polish first and second person verb morphemes in natural conversations and aims to illustrate how the recurring patterns of language use at local level, i.e. the level of the two neighboring utterances influence the extra-linguistic context at global level, i.e. the level of a one text unit. It is assumed that the psycho-sociological parameter of social distance (D) between speakers (S) and hearers (H) can be associated with the following two patterns of personal verb form shifts occurring between neighboring utterances uttered by different interlocutors and containing personal verb forms. [FR/FR]: the maximal social distance between the speaker and the hearer. (personal verb form shift occurring between the preceding utterance that contains no second, only first person verb form [FR] and the following utterance that also contains no second, only first person verb form [FR], but is uttered by a different speaker) [SC/SC]: the minimal social distance between the speaker and the hearer. (personal verb form shift occurring between the preceding utterance that contains second person verb form [SC] and the following utterance that also contains second person verb form [SC], but is uttered by a different speaker) 38 4. Methodology In order to illustrate the above hypothesis, i.e. the existence of a relationship between recurring patterns of language use and the extra-linguistic context of conversation, quantitative discourse analysis has been chosen as a method for the present research. 4.1. Linguistic material In order to observe the relationship between the frequency of the above mentioned two patterns of shifting behaviors occurring between personal verb forms and the socio-psychological distance between speech participants, two types of interaction models that differ in respect to whether the social- psychological distance between speech participants is considered large or small have been chosen for the present study. Interaction model where the socio-psychological distance between interlocutors is considered large are for instance interviews between academic teachers and students. Eight encounters recorded between academic teachers and students have been used in the present study. Their topic was the writing of the student’s graduate dissertation. Interaction model where the socio-psychological distance between interlocutors is considered relatively small are for instance conversations between acquainted female speakers. Eight encounters recorded between 20-year old acquainted female speakers have been used in the present study. Their topic was the spending of free time and study plans. Interviews between academic teachers and students were recorded at Casmir the Great Academy, Bydgoszcz, Poland in March 2003. Conversations between acquainted female speakers were recorded in Warsaw, Poland in June 2005. The recorded material was transcribed into a database of computer software Excel 2000. Transcription process followed transcription conventions proposed for Polish by Kaczmarek (2006). 39 4.2. Analysis of the linguistic data Linguistic data is analyzed according to two topics. First topic is confined to analyzing single utterances. It distinguishes three choices on the part of the speaker that belong to a paradigm called personal verb form. Second topic on the other hand, makes the object of its analysis two neighboring utterances that contain either first or second person verb forms. It discerns two kinds of choices on the part of the hearer displayed against two kinds of choices previously made by the speaker. The second topic therefore, is of a different nature than the first one. The first topic of analysis confined to the choices made by single speakers in the boundaries of one single utterance, does not take into account the surrounding context of other utterances and other speakers. The second topic however, being related to the choices made by both the speaker and the hearer, has a more context sensitive nature and aims at describing not only linguistic forms which are language specific (as is the case of the first topic) but also and first of all the processes of exchanging these linguistic forms between speakers. Although both topics can be defined as instances of language use, because when making a choice one has to take some action, the first topic reflecting the action of only one speaker is more of a static kind than the second one. The second topic, reflecting the action of both the speaker and the hearer, has a more dynamic nature and can therefore be characterized as the type of language use in its proper name. This paper claims that linguistic functions as understood by Halliday and his associates are more applicable in the previously mentioned second topic of analysis, i.e. in the processes of exchanging the linguistic forms by speakers rather than in the first topic comprising single utterances uttered by individual speakers. 4.2.1. First topic of analysis: personal verb forms As stated above, the analysis of the linguistic material in the first topic is confined to single utterances. All utterances are divided into three types that correspond to the three kinds of choices on the part of the speaker. They include 40 utterances containing second person verb forms, utterances containing no second, only first person verb forms and all the rest of the utterances that contain neither second nor first person verb forms. They are coded in the following way: [SC]: SECOND person verb forms (utterances that contain second person verb forms) [FR]: FIRST person verb forms (utterances that contain no second, only first person verb forms) [NM]: NON-MARKED (utterances that contain neither second nor first person verb forms) The coding process of the first topic with the example of two conversation data transcripts is presented below. Line Nr Utter. Nr Utter. End Speaker Content of conversation Personal verb form 214 181-1 PWF05 oj, fajne czasy, no zobacz,, oj: oh: EXCLAMATION, fajne: great: ADJECTIVE, czasy: time: NOUN, no: EMPHASIZING PARTICLE, zobacz: look: 2nd PERSON VERB FORM, IMPERATIVE, great time, wasn’t it,, / 215 182 * PWF01 no. no: EMPHASIZING PARTICLE yeh. NM 216 181-2 * PWF05 ?e tam studiowa??? {<}. ?e: that: CONNECTING PARTICLE, tam: there: INDICATIVE PRONOUN, studiowa???: study: 2nd PERSON VERB FORM, PAST, FEMININE, for you to study there. SC 217 183 * PWF01 {>} lubi?am tam studiowa?. nie: EXCLAMATION, fajnie: great: ADVERB, lubi?am: like: 1st PERSON VERB FORM, PAST, FEMININE, tam: there: INDICATIVE PRONOUN, studiowa?: study: INFINITIVE, it really was a great time, I loved to study there. FR 41 218 184 * PWF05 a co ci? tu przywia?o potem?. a: CONNECTING PARTICLE, co: what: QUOTATION PARTICLE, ci?: you: 2nd. PERSON PRONOUN, ACCUSATIVE, tu: here: INDICATIVE PRONOUN, przywia?o: blow over: 3rd.PERSON VERB FORM, PAST, NEUTRAL, potem: afterwards, why did you come here afterwards, in that case? NM 219 185 * PWF01 znaczy wiesz co ja nie zamierza?am zosta? w [nazwa miasta] (mhm), poza tym dosta?am si? na aplikacj? tam w (mhm) u siebie i- jako? tak wiesz. znaczy: mean: 3rd. PERSON VERB FORM, wiesz: know: 2nd PERSON VERB FORM, co: what: EMPHATIC PARTICLE, ja: I: 1st. PERSON PRONOUN, nie: no: NEGATION PARTICLE, zamierza?am: intend: 1st. PERSON VERB FORM, PAST, FEMININE, zosta?: stay: INFINITIVE, w: in: PREPOSITION, [name of the city], (mhm: EMPHATIC PARTICLE), poza tym: besides, dosta?am si?: get accepted: 1st. PERSON VERB FORM, PAST, FEMININE, si?: REFLEXIVE PRONOUN, na: for: PREPOSITION, aplikacj?: apprenticeship, tam: there: INDICATIVE PRONOUN, w: in: PREPOSITION, (mhm: EMPHATIC PARTICLE), u siebie: at my place, i: and: CONNECTING PARTICLE, jako? tak: somehow, wiesz: know: 2nd. PERSON VERB FORM, I mean, you know I wasn’t going to stay in [city name] and besides, I was accepted for an apprenticeship in my place, and- somehow you know. SC 220 186 * PWF05 a na jak??. a: CONNECTING PARTICLE, na: for: PREPOSITION, jak?: which: QUOTATION PARTICLE, what kind of apprenticeship?. NM 221 187 * PWF01 ??dow?. judical. NM 42 222 188 * PWF05 a to je, znaczy jeste? teraz {<}. a: well: EMPHATIC PARTICLE, to: this: INDICATIVE PRONOUN, je(st): is: 3rd. PERSON VERB FORM, znaczy: mean: 3rd.PERSON VERB FORM, jeste?: are: 2nd. PERSON VERB FORM, teraz: now, czy: or: CONNECTING PARTICLE, ju?: already, sko?czy???: finish: 2nd.PERSON VERB FORM, PAST, FEMININE, well, it means you still are taking classes or have you finished already?. SC 223 189 * PWF01 {>} sko?czy?am s?dow?, ale dosz?am do wniosku ?e to nie, nie jest zawód dla mnie [lekki ?miech]. nie: no: NEGATION PARTICLE, sko?czy?am: finish: 1st. PERSON VERB FORM, PAST, FEMININE, ??dow?: judicial: ADJECTIVE, ale: but: CONNECTING PARTICLE, dosz?am: reach: 1st. PERSON VERB FORM, PAST, FEMININE, do: to: PREPOSITION, wniosku: conclusion, ?e: that: PARTICLE, to: this: INDICATIVE PRONOUN, jest: is: 3rd. PERSON VERB FORM, zawód: job: NOUN, dla: for: PREPOSITION, mnie: me: 1st. PERSON PRONOUN, GENITIVE, no, I finished a judical apprenticeship, I have finished a judical, but have also decided it wasn’t a job for me [laughing slightly]. FR 224 190-1 PWF05 na prawd?? {<},, really?, oh and ,, / 225 191 * PWF01 {>}. yes, really. NM 226 190-2 * PWF05 wycofa??? si? z tego. wycofa???: abandon: 2nd. PERSON VERB FORM, PAST, FEMININE, si?: REFLEXIVE PRONOUN, z: from: PREPOSITION, tego: this: INDICATIVE PRONOUN, MASCULINE, GENITIVE, you abandoned it. SC 227 192 * PWF01 znaczy sko?czy?am, po egzaminie jestem wszystko. znaczy: mean: 3rd.PERSON VERB FORM, sko?czy?am: finish: 1st.PERSON VERB FORM, PAST, FEMININE, po: after: PREPOSITION, egzaminie: exam: NOUN, jestem: I am: 1st.PERSON VERB FORM, wszystko: everything, FR 43 well, I did finish it, I took final exams, that’s all. 228 193 * PWF05 ale nie ten, {<}. ale: but: CONNECTING PARTICLE, nie: no: NEGATION PARTICLE, ten: this: EMPHATIC PARTICLE, masz: have: 2nd.PERSON VERB FORM, asesury: assessor qualifications: NOUN, GENITIVE, but you are not a qualified assessor, are you? . SC 229 194 * PWF01 {>} nie bo po prostu powiedzia?am? ?e to nie dla mnie i koniec. nie: no: NEGATION PARTICLE, bo: because: CONNECTING PARTICLE, po prostu: simply: ADVERB, powiedzia?am: say: 1st. PERSON VERB FORM, PAST, FEMININE, ?e: that: CONNECTING PARTICLE, to: this: INDICATIVE PRONOUN, dla: for: PREPOSITION, mnie: me: 1st.PERSON PRONOUN, GENITIVE, i: and: CONNECTING PARTICLE, koniec: end, no, no, I am not, I simply told myself it wasn’t for me and that’s the end. FR 230 195 * PWF05 [wybuch ?miechu]. [burst of laughter]. NM Transcript 1: Example of the first topic of analysis: [SC], [FR] - personal verb form (above). Line Nr Utter. Nr Utter. End Speaker Content of conversation Personal verb form 19 16-1 PTF02 {>} nie chodzi o to, ?eby pani dok?adnie tam wszystko, ale od czasu do czasu mo?na by,, oczywi?cie: of course, nie: no: NEGATION PARTICLE, chodzi: GO: 3rd.PERSON VERB FORM, o: about: PREPOSITION, to: this: INDICATIVE PRONOUN, ?eby: in order to: CONJUNCTION, pani: you: HONORIFIC PRONOUN, FEMININE, dok?adnie: exactly: ADVERB, tam: there: EMPHATIC PARTICLE, wszystko: everything, ale: but: CONJUNCTION, od czasu do czasu: from time to time, mo?na by: one could: SUBJUNCTIVE, IMPERSONAL, HONORIFIC, of course, it’s not that you should be doing everything thoroughly, but from time to time you could,, / 44 20 17 * PSF02 {<}. si?gn??: reach out: INFINITIVE, reach out for. NM 21 16-2 PTF02 {>} a nu? pani trafi tutaj jak?? (mhm), no,, mo?e: maybe: a nu?: suppose (you): PARTICLE, pani: you: HONORIFIC PRONOUN, FEMININE, trafi: find: 2nd.PERSON VERB FORM, HONORIFIC, FUTURE, tutaj: here: INDICATIVE PRONOUN, jak??: some, mhm: NODDING SIGN, no: EMPHASIZING PARTICLE, maybe, maybe you happen to find some (yes) a-,, / 22 18 * PSF02 {<}. sugesti?: suggestion: NOUN, ACCUSATIVE, suggestion. NM 23 16-3 * PTF02 {>}, które by mog?a pani tam jeszcze (mhm), przypis mo?na zawsze doda?, prawda. trafne: suitable: ADJECTIVE, spostrze?enie: remark: NOUN, które: which: CONNECTING PARTICLE, by mog?a: could: SUBJUNCTIVE, HONORIFIC, FEMININE, 2nd.PERSON VERB FORM, pani: you: HONORIFIC 2nd.PERSON PRONOUN, FEMININE, tam: EMPHASIZING PARTICLE, jeszcze: still, mhm: NODDING SIGN, przypis: note: NOUN, mo?na: one can: MODAL VERB, IMPERSONAL, zawsze: always, doda?: add: INFINITIVE, prawda: right, isn’t it?: SENTENCE- END PARTICLE, suitable remark that you could still (yes), you can always add it in the notes, can’t you?. SC 24 19 * PSF02 my?li pani doktor, ?e ta y-, ksi??ka pani jest na przyk?ad do wypo?yczenia na [nazwa ulicy] ?. my?li: think: 2nd.PERSON VERB FORM, HONORIFIC, pani: you: HONORIFIC 2nd. PERSON PRONOUN, FEMININE, doktor: doctor: HONORIFIC TITLE, ?e: that: CONECTING PARTICLE, ta: this: INDICATIVE PRONOUN, FEMININE, y-:WONDERING SIGN, ksi??ka: book: NOUN, pani: your: POSSESIVE PRONOUN, HONORIFIC, FEMININE, jest: is: 3rd.PERSON VERB FORM, na przyk?ad: for instance, do: in order to: CONNECTIVE, wypo?yczenia: borrow: NOMINALIZED VERB, na: at: PREPOSITION, [nazwa ulicy]: [NAME OF SC 45 THE STREET]. do you think, doctor that your book is available at, that I could borrow it at [name of the street]?. 25 20-1 PTF02 znaczy ja pani to,, znaczy: mean: 3rd.PERSON VERB FORM, ja: I: 1st.PERSON PRONOUN, pani: you: HONORIFIC PERSONAL PRONOUN, FEMININE, DATIVE, to: this: INDICATIVE PRONOUN, I mean, I will,, / 26 21 * PSF02 aha {<}. aha: EXCLAMATION PARTICLE, pani: you: 2nd PERSON PRONOUN, HONORIFIC, FEMININE, doctor: doctor: HONORIFIC TITLE, to: EMPHASIZING PARTICLE, mo?e: maybe: MODAL EXPRESSION, oh, in that case Doctor, maybe. NM 27 20-2 PTF02 {>},, poniewa?: because, w tej chwili: in this moment, po?ycz?: lend: 1st.PERSON VERB FORM, FUTURE, I can lend it now to you,, / 28 22 * PSF02 {<}. aha: EXCLAMATION PARTICLE, to: this: INDICATIVE PRONOUN, bardzo: very, dobrze: good: ADVERB, oh, that’s very good. NM 29 20-3 PTF02 {>}ale, ale...,, tylko: only, to: this: INDICATIVE PRONOUN, nie: not: NEGATION PARTICLE, jest: is: 3rd.PERSON VERB FORM, moja: my: POSSESIVE PRONOUN, FEMININE, SINGULAR, ??asno??: property: NOUN, ale: but: CONNECTING PARTICLE, only this isn’t my property, but, but,, / 30 23 * PSF02 mhm. mhm: NODDING SIGN, yes. NM 46 31 20-4 * PTF02 no, pani mi odda j?, prawda?, {<} ??miech>. no: EMPHASIZING PARTICLE, pani: you: 2nd.PERSON PRONOUN, HONORIFIC, FEMININE, mi: me: 1st. PERSON PRONOUN, DATIVE, odda: return: 2nd.PERSON VERB FORM, HONORIFIC 2nd.PERSON VERB FORM, j?: it: 3rd.PERSON PRONOUN, ACCUSATIVE, prawda: right, isn’t it?: SENTENCE-END PARTICLE, oczywi?cie: of course, you will return it back to me, won’t you, of course . SC 32 24 * PSF02 {>}. oczywi?cie: of course I will, of course. NM Transcript 2: Example of the first topic of analysis: [SC], [FR] - personal verb form (above). 4.2.2. Second topic of analysis: personal verb form shifts Analysis of the linguistic material in the second topic is based on the results of the first topic and takes into account only those utterances that contain either second person verb forms [SC] or contain no second, only first person verb forms [FR]. They must be a pair of the two neighboring utterances uttered by different interlocutors and need to belong to the same type. They are coded in the following way: [SC/SC]: SECOND person vs. SECOND person (personal verb form shift occurring between utterances uttered by different interlocutors that contain second person verb forms [SC], i.e. the preceding utterance contains the second person verb form [SC] and the following utterance also contains the second person verb form [SC]) [FR/FR]: FIRST person vs. FIRST person (personal verb form shift occurring between utterances uttered by different interlocutors that contain no second, only first person verb forms [FR], i.e. the preceding utterance contains no second, only first person verb form [FR] and the following utterance also contains no second, only first person verb form [FR]) 47 The coding process of the second topic with the example of the two conversation data transcripts is presented below. Line Nr Utter. Nr Utter. E nd Speaker Content of conversation Personal verb form Personal v verb form Shift 1 1 * PBF03 /od 18 sek./ no to porozmawiajmy Alu co tam u ciebie y n, n studiach. no to: EMPHATIC PARTICLES, porozmawiajmy: talk: IMPERATIVE, 1st. PERSON VERB FORM, PLURAL, Alu: Al: FEMALE NAME, VOCATIVE, co: what: QUOTATION PRONOUN, tam: there: EMPHATIC PARTICLE, u ciebie: at your place, studiach: studies. come on Al, let’s talk about your matters, how are your studies going?. FR 2 2 * PBF01 ale co chcesz wiedzie?, jak sesja?, {<}. ale: but: CONNECTING PARTICLE, co: what: QUOTATION PRONOUN, chcesz: want: 2nd. PERSON VERB FORM, wiedzie?: know: INFINITIVE, jak: how: QUOTATION PRONOUN, sesja: exam session, czy: QUOTATION PARTICLE, but what do you want to know, about the exam session or what?. SC 3 3 * PBF03 {>}, tak, jak tam egzaminy, czy ju? masz wszystko zaliczone. jak: how: QUOTATION PRONOUN, sesja: exam session, tak: yes: EMPHATIC PARTICLE, jak: how: QUOTATION PRONOUN, tam: there: EMPHATIC PARTICLE, egzaminy: exams: NOUN, czy: QUOTATION PARTICLE, ju?: already, masz: have: 2nd.PERSON VERB FORM, wszystko: everything, zaliczone: pass: PASSIVE VOICE, about the exam session, yes, about the exams, have you passed them all. SC SC/SC 48 4 4 * PBF01 nie, nie, mam trzy egzaminy zdane, jeden praktyczny te?, i jeden jeszcze jest do zdania. nie:no: NEGATION PARTICLE, mam: have: 1st. PERSON VERB FORM, trzy: three: NUMERAL, egzaminy: exams: NOUN, PLURAL, zdane: pass: PASSIVE VOICE, jeden: one: NUMERAL, praktyczny: practical: ADJECTIVE, te?: too, i: and: CONJUNCTION, jeszcze: still, jest: is: 3rd.PERSON VERB FORM, do: in order to: PREPOSITION, zdania: pass: NOMINALIZED VERB, PASSIVE VOICE, no, no, I have passed three exams, one of them was practical and there is still one exam I need to take. FR 5 5 * PBF03 ale teoretyczny te? czy praktyczny?. ale: but: CONNECTING PARTICLE, teoretyczny: theoretical: ADJECTIVE, te?: too, czy: QUOTATION PARTICLE, praktyczny: practical: ADJECTIVE, but is it too a practical one or a theoretical one? . NM 6 6 * PBF01 teoretyczny, ale b?dzie ustny. teoretyczny: theoretical: ADJECTIVE, ale: but: CONNECTING PARTICLE, ??dzie: will be: 3rd.PERSON VERB FORM, FUTURE, ustny: oral: ADJECTIVE, a theoretical one but it is going to be an oral one. NM 7 7 * PBF03 ustny, o to trudno. ustny: oral: ADJECTIVE, o: EXCLAM. PARTICLE, to: this: INDICATIVE PRONOUN, trudno: tough: ADVERB, an oral one, oh, sounds tough. NM Transcript 3: Example of the second topic of analysis: [SC/SC] - personal verb form shift (above). Line Nr Utter. Nr Utter. End Speaker Content of conversation Personal verb form Personal verb form shift 16 13 * PWF05 =a w ogóle uprawiasz jakie? sporty?. a: CONNECTING PARTICLE, w ogóle: at all, uprawiasz: do: 2nd.PERSON VERB FORM, jakie?: any, sporty: sports, do you do any sports, anyway?. SC 49 17 14 * PWF04 znaczy ja bardzo na rowerze lubi? je?dzi? i ale to jest mniej m?cz?ce o wiele. znaczy: mean: 3rd. PERSON VERB FORM, ja: I: FIRST PERSON PRONOUN, bardzo: very, na rowerze: by bike, lubi?: like: 1st.PERSON VERB FORM, je?dzi?: ride: INFINITIVE, i: and: CONNECTING PARTICLE, ale: but: CONNECTING PARTICLE, to: this: INDICATIVE PRONOUN, jest: is: 3rd.PERSON VERB FORM, mniej: less, ??cz?ce: tiring: ADJECTIVE, o wiele: much (less), I mean, I like bicycle riding very much and, but this is much less tiring. FR 18 15 * PWF05 no, no, te? je?dzi?am, no nie, ja bie, ja biegam ale to tak sporadycznie i tylko dlatego ?eby tam ko?ci rozprostowa? i fitness lubi? te? co powiedzia?am. no: EMPHATIC PARTICLE, te?: too, je?dzi?am: used to ride: PAST, 1st. PERSON VERB FORM, FEMININE, IMPERFECT, no nie: EMPHATIC PARTICLES, ja: I: 1st PERSON PRONOUN, biegam: run: 1st. PERSON VERB FORM, ale: but: CONNECTING PARTICLE, to: this: INDICATIVE PRONOUN, tak: EMPHATIC PARTICLE, sporadycznie: rarely: ADVERB, i: and: CONNECTING PARTICLE, tylko: only, dlatego: because, ?eby: in order to, tam: EMPHATIC PARTICLE, ko?ci: bones: NOUN, ACCUSATIVE, rozprostowa?: straighten: INFINITIVE, i: and: CONNECTING PARTICLE, fitness: fitness: NOUN, ACCUSATIVE, lubi?: like: 1st.PERSON VERB FORM., te?: too, co: what (as I said): PARTICLE, powiedzia?am: say: 1st.PERSON VERB FORM, PAST, FEMININE, right, right, I also used to ride a bicycle, but I do some running instead and I do it only from time to time and I do it only in order to straighten my bones, and I also like fitness as I said before. FR FR/FR 50 19 16 * PWF04 to te?, no, te? lubi? tylko ?e jako? tak nigdy si? nie mog? zabra?? ?eby na to chodzi?, znaczy ci?gle co? tak by?o ?e, ?e w???nie nie chodzi?am. to: this: INDICATIVE PRONOUN, te?: too, no: EMPHASIZING PARTICLE, lubi?: like: 1st.PERSON VERB FORM, tylko: only, ?e: that: CONNECTING PARTICLE, jako?: somehow, tak: EMPHATIC PARTICLE, nigdy: never, si?: REFLEXIVE PRONOUN, nie: NEGATION PARTICLE, mog?: can: 1st.PERSON VERB FORM, zabra?: take: INFINITIVE, ?eby: in order to: CONNECTIVE, na: PREPOSITION, to: this: INDICATIVE PRONOUN, chodzi?: take part: INFINITIVE, znaczy: mean: 3rd.PERSON VERB FORM, ci?gle: all the time, co?: something, tak: EMPHATIC PARTICLE, by?o: was: PAST BE, 3rd.PERSON VERB FORM, ?e: that: CONNECTING PARTICLE, ????nie: just: EMPHASIZING EXPRESSION, nie: no: NEGATION PARTICLE, chodzi?am: go: 1st. PERSON VERB FORM, PAST, FEMININE, IMPERFECT, me too, I like it too, but somehow I can never get myself to participate in it, I mean, there was always something that was keeping me away from taking part in it. FR FR/FR Transcript 4: Example of the second topic of analysis: [FR/FR] - personal verb form shift (above). 5. Results and Discussion Results of the data analysis are presented in three sections. The first section provides general information regarding the linguistic material and the transcribed utterances. The second section highlights the results of the first topic of analysis, i.e. the distribution of utterances containing the second [SC] person verb forms, no second, only first [FR] person verb forms, as well as all other utterances that contain neither second nor first person verb forms [NM] in all encounters between academic teachers and students as well as in all conversations conducted between 51 20-year old acquainted females. The third section highlights the results of the second topic of analysis, i.e. the frequency of the second to second [SC/SC] personal verb form shifts, as well as the first to first [FR/FR] personal verb form shifts that occurred in all encounters between academic teachers and students. The results are then contrasted with the results of all conversations performed by the 20-year old acquainted females. 5.1. General information Information regarding the number of conversation data, total number of transcribed utterances [N], mean lenght of utterances in a single conversation [M], as well as the standard deviation from the mean lenght value [SD], is provided in the table below. Table 4: General information Interaction Type Number of conversation data Transcribed utterances N M SD Interviews between teachers and students 8 2074 259.25 70.7 Conversations between acquainted female speakers 8 1332 166.5 30.9 Total: 16 3406 212.9 28.1 5.2. Results of the first topic of analysis: utterances containing personal verb forms In the total of 8 interviews performed between teachers and students, the following distribution of utterances containing second person verb forms [SC], utterances containing no second, only first person verb forms [FR] as well as all other utterances that contain neither second nor first person verb forms [NM] has been calculated. 52 Table 5: Distribution of utterances containing personal verb forms in the interviews between academic teachers and students [SC] [FR] [NM]Interview Teacher Student Teacher Student Teacher Student Total: 1 29 3 29 36 72 80 249 2 31 0 19 27 57 45 179 3 33 4 15 25 39 93 209 4 23 4 24 28 60 94 233 5 47 3 31 47 52 65 245 6 28 2 17 44 47 85 223 7 47 5 37 70 99 114 372 8 53 0 35 74 78 124 364 291 21 207 351 504 700 Total: 312 (15%) 558 (27%) 1204 (58%) 2074 (100%) According to the above results, utterances containing second person verb forms [SC] cover 15% of the total number of utterances, while utterances containing no second, only first person verb forms [FR] cover 27% of the total number of utterances. The greater percentage of utterances containing no second, only first person verb forms [FR] can be interpreted as a reflection of the formal atmosphere of the setting. It can also function as an indicator of a considerable socio-psychological distance between interlocutors. Moreover, looking at a number of utterances uttered by individual speakers for instance, a considerable difference is visible between utterances containing second person verb forms [SC] uttered by teachers [T] and uttered by students [S]. This big gap between teachers and students can be interpreted as a reflection of their role relationship. It can also stand as evidence to the assumption that the extra-linguistic context of speech situation may influence the language use of individual speakers. In the above situation for example, the student in order to display a subordinate social status towards his supervisor has to constrain himself from using second person verb forms. 53 In the total of 8 conversations performed between 20-year old acquainted female speakers, the following distribution of utterances containing second person verb forms [SC], utterances containing no second, only first person verb forms [FR] and all other utterances containing neither second nor first person verb forms [NM] has been calculated. Table 6: Distribution of utterances containing personal verb forms in the conversations between 20-year old acquainted female speakers [SC] [FR] [NM]Conver sation Female 1 Female 2 Female 1 Female 2 Female 1 Female 2 Total: 1. 29 16 17 20 31 36 149 2. 18 21 7 8 34 26 114 3. 37 15 18 20 43 58 191 4. 16 26 15 10 38 50 155 5. 15 36 9 17 53 21 151 6. 15 28 20 24 55 37 179 7. 21 26 13 17 82 56 215 8. 23 22 18 13 48 54 178 174 190 117 129 384 338 Total: 364 (27.3%) 246 (18.5%) 722 (54.2%) 1332 (100%) The above results show that in conversations between acquainted female speakers, utterances containing second person verb forms [SC] cover 27.3% in the total number of utterances, while utterances containing no second, only first person verb forms [FR] cover 18.5% in the total number of utterances. The greater percentage of utterances containing second person verb forms [SC] can be interpreted as a reflection of the casual atmosphere of the setting and the existence of a minimal socio-psychological distance between interlocutors. Figure 1 below contrasts the percentage of utterances containing second person verb forms [SC] with the percentage of utterances containing no second, 54 only first person verb forms [FR] in conversations recorded between acquainted female speakers as well as in interviews performed between teachers and students. The results show that in conversations performed between acquainted female speakers, the percentage of utterances containing second person verb forms [SC] exceeds the percentage of utterances containing no second, only first person verb forms [FR]. In interviews performed between academic teachers and students however, the percentage of utterances containing second person verb forms [SC] does not exceed and is much lower than the percentage of utterances containing no second, only first person verb forms [FR]. If the difference in the percentage of the above type of utterances is accounted for as a reflection of the difference in the perception of the socio- psychological distance between speech interlocutors, therefore in conversations performed between acquainted female speakers, the higher percentage of utterances containing second person verb forms [SC] could be interpreted as a tendency to diminish the socio-psychological distance between speakers. In interviews performed between teachers and students on the other hand, the higher percentage of utterances containing no second, only first person verb forms [FR] could stand for as a tendency to increase or keep up a considerable socio- psychological distance between teachers and students. Figure 1: Distribution of utterances containing second person verb forms [SC] and no second, only first person verb forms [FR] in all conversations recorded between acquainted female speakers as well as in all interviews recorded between teachers and students SC (15%) SC (27,3%) FR (27%) FR (18,5%) 0% 5% 10% 15% 20% 25% 30% Teachers and students Acquainted Females 55 5.3. Results of the second topic of analysis: personal verb form shifts In the total of 8 interviews performed between teachers and students, the following frequency of the first to first [FR/FR] as well as the second to second [SC/SC] personal verb form shifts between neighboring utterances uttered by different interlocutors has been calculated. Table 7: Frequency of the second to second [SC/SC] and the first to first [FR/FR] personal verb form shifts in the interviews between teachers and students [SC/SC] [FR/FR]Conver sation Teacher Student Teacher Student Total: 1 1 0 4 5 10 2 0 0 6 2 8 3 1 2 2 2 7 4 1 0 6 3 10 5 0 1 9 11 21 6 0 1 4 5 10 7 0 2 10 7 19 8 0 0 15 14 29 3 6 56 49 Total: 9 (7.9%) 105 (92.1%) 114 (100%) The above results reveal a very low frequency of the second to second personal verb form shifts [SC/SC], compared to a very high frequency of the first to first personal verb forms shifts [FR/FR] in the interviews performed between teachers and students. This big gap indicating 7.9% for the second to second [SC/SC] and 92.1% for the first to first [FR/FR] personal verb form shifts between interlocutors may be interpreted in the following way. Keeping in mind that the overall atmosphere in the interviews between academic teachers and students is usually very formal, especially when the interview takes place at the university campus or at the teacher’s office, the high percentage of the first to first personal verb form shifts [FR/FR] between interlocutors may be accounted for as an indicator of a considerable socio-psychological distance between teachers and 56 students. To put it in a different way, in order to maintain the expected socio- psychological distance, teachers and students tend to perform a high number of the first to first personal verb form shifts [FR/FR] from interlocutors. Meanwhile, in the total of 8 conversations performed between 20-year old acquainted female speakers, the following frequency of the first to first [FR/FR] and the second to second [SC/SC] personal verb form shifts conducted between different interlocutors has been calculated. Table 8: Frequency of the second to second [SC/SC] and the first to first [FR/FR] personal verb form shifts in the conversations between acquainted females [SC/SC] [FR/FR]Conver sation Female 1 Female 2 Female 1 Female 2 Total: 1. 4 6 5 5 20 2. 5 6 0 1 12 3. 7 2 6 5 20 4. 3 3 1 2 9 5. 7 6 5 2 20 6. 2 1 6 1 10 7. 4 4 1 2 11 8. 6 5 2 1 14 38 33 26 19 Total: 71 (61.2%) 45 (38.8%) 116 (100%) The above results indicate that the frequency of the second to second [SC/SC] personal verb form shifts was higher than the frequency of the first to first [FR/FR] personal verb form shifts in conversations conducted by 20-year old acquainted female speakers. The 61.2% of the second to second [SC/SC] personal verb forms shifts compared to the 38.8% of the first to first [FR/FR] personal verb form shifts can be interpreted in the following way. 20-year old acquainted female speakers have a tendency to diminish the socio-psychological distance that is inevitably present between them due to the relatively high percentage of the first to first [FR/FR] personal verb form shifts. This behavior can be interpreted as an 57 inclination on the part of both interlocutors to conform to a rather informal atmosphere of the setting. Figure 2 below contrasts the percentage of the second to second [SC/SC] personal verb form shifts, with the percentage of the first to first [FR/FR] personal verb form shifts from interlocutors in conversations performed between acquainted female speakers as well as in interviews performed between teachers and students. The results show that in conversations performed between 20-year old acquainted female speakers, the percentage of the second to second [SC/SC] personal verb form shifts is higher than the percentage of the first to first [FR/FR] personal verb form shifts from interlocutors. In interviews between teachers and students on the other hand, the percentage of the second to second [SC/SC] personal verb form shifts is much lower than the percentage of the first to first [FR/FR] personal verb form shifts from interlocutors. The above difference in the percentage of the second to second [SC/SC] and the first to first [FR/FR] personal verb form shifts can be accounted for as a perception difference of the socio-psychological distance between speech participants. In conversations between acquainted female speakers therefore, the SC/SC 7,9% 0% 20% 40% 60% 80% 100% Teachers and students Acquainted Females Figure 2: Percentage of the second to second [SC/SC] compared to the first to first [FR/FR] person verb form shifts in all conversations recorded between acquainted female speakers as well as in all interviews recorded between teachers and students S /SC 7,9% SC/SC 61,2% FR/FR 38,8% FR/FR 92,1% 58 higher percentage of the second to second [SC/SC] personal verb form shifts can be interpreted as functioning in order to diminish the socio-psychological distance between speakers. In interviews between teachers and students on the other hand, the higher percentage of the first to first [FR/FR] personal verb form shifts can be interpreted as functioning in order to increase or keep up a considerable socio- psychological distance between teachers and students. 6. Conclusion The aim of this study was to illustrate the relationship between the extra-linguistic context of conversation, i.e. the socio-psychological distance between speech participants and the recurring patterns of grammar in the linguistic behaviors of interlocutors. Two patterns in the shifting of personal verb forms were distinguished and each pattern was attributed a meaning. Both patterns were identified as actively contributing to the regulation of the socio-psychological distance between speakers. One pattern was recognized as functioning in order to diminish the socio-psychological distance, while the other one acquired the meaning of functioning in order to increase or maintain the existing socio- psychological distance between interlocutors. Results of the conversation data analysis confirmed that in interviews performed between teachers and students, the most frequent type of the personal verb form shifting behavior was the shift from first to first [FR/FR] personal verb morphemes. It has been therefore attributed a meaning as functioning in order to increase or maintain the existing socio-psychological distance between interlocutors and contributing to the creation of a formal atmosphere of the setting. In conversations performed by acquainted female speakers on the other hand, the most frequent type of the personal shifting behavior was the shift from second to second [SC/SC] personal verb morphemes. It has been therefore attributed a meaning as functioning in order to diminish the socio-psychological distance between interlocutors and contributing to the creation of a casual atmosphere of the setting. 59 Further research would inquire into personal verb form shifts occurring between utterances that are of the opposite type, i.e. first to second [FR/SC] and second to first [SC/FR] personal verb form shifts from interlocutors. It would be worthwhile to find out to what extent these patterns contribute to the creation of superior and subordinate social roles between speakers. References Benveniste, E. 1971. Problems in General Linguistics. Florida: University of Miami Press. Brown, P. and Levinson, S.C. 1987. Politeness, Some universals in language usage. Cambridge University Press. Cysouw, M. 2003. The Paradigmatic Structure of Person Marking. Oxford University Press. Eggins, S. 2004. An Introduction to Systemic Functional Linguistics(2nd Ed.). New York-London: Continuum. Halliday, M.A.K. 1978. Language as Social Semiotic. London: Edward Arnold. Halliday, M.A.K. 1994. An Introduction to Functional Grammar (2nd Ed.).London: Edward Arnold. Jakobson, R. 1984. “Shifters, verbal categories and the Russian verb” In Russian and Slavic Grammar Studies 1931-1981, L.R.Waugh and M.Halle (eds), 40- 58. Berlin-New York-Amsterdam. Mouton Publishers. Jespersen, O. 1922. Language, its nature, development and origin. New York: The Norton Library. Kaczmarek, M. 2006. Polski tekst mówiony jako przedmiot zainteresowa? lingwisty –propozycja systemu transkrypcji oraz metod analizy tekstu– [Polish spoken text as a subject of interest for the linguist –a proposal for transcription system and methods of text analysis]. In Slavia Occidentalis Iaponica 9: 81-116. Siewierska, A. 2004. Person. Cambridge University Press. 60 Linguistic Survival Strategies in Geoffrey Chaucer’s Clerk’s Tale, William Shakespeare’s Othello and Elizabeth Cary’s The Tragedy of Mariam Nely Keinänen and Mari Pakkala-Weckström, University of Helsinki This paper combines linguistic and literary techniques in order to study the relationship between gender, power and politeness in heterosexual relationships in the medieval to the early modern period, focusing on three fictional heroines facing the threat of marital violence. Given that women are typically both physically weaker than men, and institutionally subordinate to them, when threatened with marital violence, women must resort to linguistic survival strategies in order to cope. In other words, they must find something to say and a way of saying it which will defuse the current situation. Although wife-abuse is not a common theme in literature, a number of literary texts do treat the issue, perhaps most famously Geoffrey Chaucer’s depiction of Patient Griselda,1 the heroine of the Clerk’s Tale, who suffers extreme emotional violence at the hands of her husband Walter, and William Shakespeare’s Othello. Less well known is Elizabeth Cary’s The Tragedy of Mariam, which was written at the same time as Othello, telling a similar story about a jealous husband who has his wife killed when he is tricked into thinking her unfaithful. All three works present extreme views of a wife’s vulnerability to her husband, and in this sense provide the most useful material for studying depictions of wives under duress. Our aim in this essay is two-fold: as linguists, we hope to complicate the idea that women mainly use negative politeness strategies, as a result of their institutional subordination to men, while men use more positive strategies. As literary scholars, we hope to apply linguistic tools to deepen feminist critique of these characters, showing that 1 Another one of the Canterbury Tales to treat – albeit briefly – domestic violence is the Wife of Bath’s Prologue. 61 while these female characters lack power, their flexible use of linguistic survival strategies shows that they are not passive. Since the links between language, gender and power are so complex, let us start with a brief consideration of the many faces of power. Power can be roughly divided into two categories: institutional and private power.2 Institutional power is exercised for example in courtrooms, classrooms and other public arenas. The private side of power belongs to the sphere of family, home, and friendships. These two categories are interrelated and sometimes overlapping: in this study, we are particularly interested in the interplay between institutional power in a private setting, i.e. the gender-related power which nominally gave husbands the right to control their wives up to the point of killing them, and private power, which is not only more evenly distributed between partners, but is also prone to more fluctuation given its strong emotional component. These fluctuations are often visible in dialogue; they are manifested through the introduction of various linguistic strategies, which include features commonly associated with politeness, such as terms of endearment, respectful address forms, and apologies, combined with impolite or rude expressions etc. Previous research has demonstrated that women are generally more polite than men, and that women tend to have a more co-operative conversational style as opposed to men, whose style is more competitive.3 Our data challenges these ideas. Of our three heroines, Griselda faces the most linguistic constraints, as she has essentially promised to always be polite to Walter, to obey his every whim. Yet within these constraints, she manages to turn Walter from a distant, despotic husband into a loving one, mainly through a switch from negative to positive politeness strategies, judicious topic selection and increasing control of the floor. Griselda’s shift from negative to positive politeness is reversed in Desdemona’s exchanges with her husband, where mainly positive politeness gives way to 2 A similar division can be found in e.g. David Hart (2000), who has studied power and powerlessness in four of Shakespeare’s plays. Hart makes a distinction between ‘authority’ and ‘relative power’ in which the latter means the ability to persuade or manipulate superiors. Another division is made by Ervin-Tripp et al., who, in their article “Language and Power in the Family” (1984) talk about “effective power” and “esteem”. 3 See e.g. Holmes 1995 and Herman 1996: 260. 62 negative politeness, then to no politeness when she is arguing for her life. Our analysis also shows, however, how much these categories seem to overlap, especially when interpreting forms of address. Desdemona’s linguistic strategies also correlate with her efforts to control the floor: the more she attempts to do so, the more deferential and apologetic she becomes, even as she is vigorously arguing to reinstate Cassio or save her life. Finally, in order to demonstrate the fluidity of the links between gender and politeness, we conclude this discussion with an examination of Mariam, who makes little attempt to diminish the force of her face threatening acts with conventional politeness, and can indeed be said to use rudeness as a verbal weapon. The key difference here is that, unlike Griselda (whose feelings for Walter are strategically ignored) and Desdemona (who loves her husband dearly to the end of her play), Mariam does not love her husband. In this case, her private power over her husband reverses the traditional roles, leaving Herod to try a number of ineffectual politeness strategies before abandoning them and resorting to his institutional power over his wife. Part I: Theories of Politeness One of the most basic social skills is politeness, which “represents a social norm that can be observed empirically in language and reliably analyzed by means of language” (Held 1992: 133). Politeness is discussed perhaps most influentually by Penelope Brown and Stephen Levinson (1987), who note that what they call Model Persons will cooperate to maintain face in ordinary social interaction. Their concept of face includes two aspects, positive, which essentially means a person’s need to feel appreciated; and negative, essentially a person’s need to act unimpededly.4 Given this mutual need to maintain face, there are a number of strategies for doing what they call Face Threatening Acts, ranging from not doing the act at all, to doing it off record (indirectly), or doing it on record with or without redressive action – this last option is referred to as bald on record. Redressive action is broken down into two categories, negative or positive politeness. Brown and Levinson propose that a person’s choice of politeness 4 For an interesting discussion on the concept of face, see Bargiela-Chiappini 2003. 63 strategy depends on the relative power, distance and the imposition of a face- threatening act. Later scholars have noted that politeness strategies are mainly used by acquaintances, rather than by people in intimate relationships at one end, or complete strangers at the other, who can sometimes be comfortable enough, or unconcerned enough, to adopt balder strategies (see e.g. Wolfson 1988). Brown and Levinson’s negative politeness comprises strategies which are considered ‘polite’ also in laymen’s terms: for example indirectness, respectful terms of address, apologies, hedging, and avoiding the use of personal pronouns. Positive politeness is more akin to what is generally considered friendly behavior: e.g., the use of nicknames and terms of endearment, complimenting, joking, presupposing or fulfilling the hearer’s wants or needs, and including the hearer in the activity by for example using ‘we’ instead of ‘you’. These strategies are chosen by the speaker to fit each speaking situation, so that the speaker can best achieve his or her purpose. Distinguishing between positive and negative politeness is not always straightforward. For example, when applying Brown and Levinson’s work to address formulae in early English correspondence, Nevalainen and Raumolin- Brunberg suggest thinking of politeness as a continuum between the two poles, since a single term can mean any number of things depending on the context (see also Nevala 2004). One of the weaknesses of the Brown and Levinson theory is that it does not differentiate between what we can call conventional politeness, and politeness used for egotistical reasons.5 Richard Watts, who has criticized the model on this account, makes the following division: “Politic behaviour is that behaviour, linguistic and non-linguistic, which the participants construct as being appropriate to the ongoing social interaction (2003: 20)”, while “polite behaviour . . . is behaviour beyond what is perceived to be appropriate to the ongoing social interaction” (2003: 21). According to this division, for example most address 5 In fact, Brown and Levinson avoid this distribution, stating that: “We do not mean to imply that what we dub ’strategies’ are necessarily conscious . . . but the general unconscious nature of such strategies raises methodological problems that we simply skirt” (1987: 85). 64 forms would simply fall into the category of politic behavior, being merely appropriate. However, as we shall demonstrate, the use of various address forms often seems to be influenced by strong emotions. There are also significant differences between the male and female characters’ use of such forms, which might suggest that they either go beyond what is merely appropriate, or that the somewhat ambiguous concept of appropriateness is not the same for the two genders (see also Pakkala-Weckström 2005). Within Shakespearean studies, Brown and Levinson have also been criticized for leaving out significant variables, including affect (See Brown and Gilman 1989: 159-212; and Kopytko 1995: 515-40). Furthermore, Brown and Levinson’s theory of politeness does not really stress the role of the addressee, even though any redressive action also has to be perceived as such by the hearer in order to be successful. Miriam Locher makes a clear-cut distinction between the role of the speaker and the hearer in her definition of politeness: Politeness for the speaker: A polite utterance is a speaker’s intended, marked and appropriate behavior which displays face concern; the motivation for it lies in the possibly, but not necessarily, egocentric desire of the speaker to show positive concern for the addressees and/or to respect the addressees’ and the speaker’s own need for independence. Politeness for the addressee: Addressees will interpret an utterance as polite when it is perceived as appropriate and marked; the reason for this is understood as the speaker’s intention to show positive concern for the addressees’ face and/or the speaker’s intention to protect his or her own face needs. (2004: 91). The role of the addressee is, of course, extremely important in our material: these female characters turn to linguistic strategies because they have no other means of dealing with their husbands, who are in all cases physically and institutionally superior, and in the case of Walter and Herod, also politically superior. That Desdemona and Mariam are killed speaks to the overwhelming significance of these forces. The flipside of politeness is impoliteness. Jonathan Culpeper has experimented with a categorisation of impoliteness strategies similar to Brown 65 and Levinson’s categories of positive and negative politeness strategies. The aim of impoliteness strategies is to attack face and cause social disruption (1995: 350). Culpeper’s positive impoliteness strategies include ignoring and excluding the hearer, being disinterested, using inappropriate address terms and seeking disagreement by selecting sensitive topics; negative impoliteness strategies include frightening, ridiculing or scorning the hearer and personalizing (1995: 357-358). These strategies function as opposites of Brown and Levinson’s positive and negative politeness strategies, and they seek to attack positive and negative face respectively. Interestingly, sometimes utterances can be both impolite and polite at the same time; below, we shall discuss some instances where an intended attack on the hearer’s face is at the same time garnished with politeness strategies (see also Nevala 2004, esp. chapters 7 and 8). Several scholars have also considered the relationships between politeness and gender. In Women, Men and Politeness (1995), Janet Holmes comes to the qualified conclusion that modern New Zealand women are more “polite” than New Zealand men, which means in practice that women more often adopt strategies which take into account the listener. Sara Mills, in her Gender and Politeness, considers politeness – regardless of the speakers’ gender – as “a set of practices or strategies which communities of practice develop, affirm, and contest, and which individuals within these communities engage with in order to come to an assessment of their own and others’ behavior and position within the group” (2003: 9). Interestingly, Mills goes on to argue that some speakers “use seemingly feminine strategies strategically to achieve their ends rather than to assert their subordination” (op. cit. 235). This notion of strategic use of politeness seems to come up in our data as well. Some scholars are also studying whether results from contemporary oral sources also fit historical written data. For example, Minna Palander-Collin has found that 17th century English husbands and wives used different strategies when writing letters to each other, and that wives “showed a particular concern for [their husbands]. . . with their appraisal of his actions, decisions and qualities” 66 (2002: 121). In another study, she found that wives were more likely than husbands to use “I think” as a politeness strategy (Palander-Collin 1999). Linguists have also examined Chaucer and Shakespeare for examples of politeness discourse, though in most of these studies they have for the most part been less concerned with possible gender differences in strategy use.6 In addition to Brown’s and Gilman’s, as well as Kopytko’s more descriptive and quantitative accounts of politeness in Shakespeare (1993; 1995), which were directed mainly at linguists, Lynne Magnusson (1992) has also studied politeness in Henry VIII from a more literary perspective. She provides nuanced readings of especially Katherine’s and Wolsey’s speeches, arguing that politeness strategies can be seen to help make up the personalities of the characters, and are not completely tied to individual social interactions. Her reading of Wolsey is especially impressive in the way it teases out the contradictions between his almost excessive use of negative politeness strategies and his insistence that things be done his way right now, even in the same sentence, while her reading of Katherine shows how women incorporate even in their speech the subservient subject positions allowed to them by their society, though she also shows how Katherine finds a voice through adopting the strategies of religious piety. Our work contributes to this earlier work by providing a more nuanced examination of both politeness and impoliteness strategies, along with topic selection and control of the floor, employed by these fictional heroines. Since to a very large extent the endings of all three works are determined by genre, we do not discuss the effectiveness of these linguistic strategies per se, but only focus on the types of strategies used. Part II: The Clerk’s Tale The story of Patient Griselda is often read as a medieval allegory of man’s obedience to God. Her suffering at the hands of her cruel husband is seen as an example as how people should submit to the will of God without questioning. 6 Pakkala-Weckström 2005 however has considered gender as one of the main variables in politeness use in Chaucer’s dialogue. 67 However, in Chaucer’s version of the story in particular,7 Griselda shows that she is a skilled negotiator, acting within the boundaries of her original promise of total obedience. She endures the supposed killing of her daughter and then her son with relative calm, but when Walter announces, in front of an audience, that he will replace her with a younger, more noble bride, she makes a long and emotional farewell speech, in which she addresses him in both negatively (“my lord”, “lord”) and positively polite terms (“myn owene lord”, “myn owene lord so deere”). It is significant that the positively polite forms only appear in Griselda’s speeches at this moment, when her emotions are clearly running high. This suggests that at least these address forms cannot be categorised into Watts’s politic behavior; their use would appear to go beyond merely ‘appropriate.’ Griselda also introduces intimate, uncomfortable (i.e. sensitive as suggested by Culpeper) topics such as giving him her virginity and bearing his children.8 The mention of the children, who Walter is believed to have secretly murdered, is a particularly clever move on Griselda’s part, since he is about to send her back to her father with “thilke dowere that ye broghten me”, ‘with the dowry that you brought to me’, which in Griselda’s case was absolutely nothing: she was stripped of her peasant’s clothes, and thus, in effect, came to his house naked. She openly reminds Walter of this, and begs to be allowed at least a shirt to cover “thilke wombe in which youre children leye”. Her request is enhanced by a presupposition strategy (Brown and Levinson 1987: 125), “ye koude nat doon so dishonest a thing”, meaning that he couldn’t possibly want her to walk naked in front of all these people – which of course had been his intention all along. Thus she not only exposes his cruelty about taking along the non-existent dowry, but also reminds him, and their audience, about the fate of their children. However, all this is done in a polite and friendly manner, and therefore within the parameters of Griselda’s promise; she never questions his judgment or tries to change his mind about sending her away. Yet, at the same time, on the level of topics, she is being extremely rude in response to her husband’s mocking cruelty, and this strategy 7 Although Chaucer follows his sources faithfully, he has made significant additions to Griselda’s speeches in particular (cf. Winny 1966). 8 For a detailed analysis of this scene, see Pakkala-Weckström 2005: 29-33. 68 pays off: Walter really has no choice but to allow her to keep the shirt. While Walter does not change his mind about sending his wife back to her father, it becomes clear that he is moved by her emotional speech; his answer is reported to be given with great difficulty, and he leaves the scene immediately thereafter. Thus Griselda’s combination of an impoliteness strategy with politeness pays off, at least to a degree. Griselda uses a similar technique at the emotional climax of the scene: when forced to comment on the young bride’s beauty, she also warns Walter not to be as cruel to his new bride as he has been to his first one:9 (1) O thyng biseke I yow, and warne also, That ye ne prikke with no tormentynge This tendre mayden, as ye han doon mo (1037-39) This entire scene is, of course, Walter’s final test on Griselda’s patience and humility: he eventually rewards her by revealing that her children have not actually been killed, nor has he had any intention of divorcing her, and his new ‘bride’ is actually the long-lost daughter, who has been fostered by Walter’s relatives. It is hard to know the narrator’s attitude towards Griselda’s obedience, whether we are supposed to understand these linguistic strategies ironically, whether this level of politeness and obedience is really being offered up as an example of a woman to be emulated.10 Towards the end of the tale, the narrator comments: (2) This storie is seyd nat for that wyves sholde Folwen Grisilde as in humylittee, For it were inportable [intolerable], though they wolde (1142-44) Rather, readers are encouraged to accept adversity as well as Griselda does. In any case, in Griselda we already see politeness features which later writers could develop and exploit: both negative and positive politeness strategies, coupled with 9 The word ‘mo’ is somewhat ambiguous, usually taken to mean ‘others’ – the French source text says ‘l’autre’ (Farrell and Goodwin 2002: 165); by not naming herself directly, Griselda softens her FTA considerably (Pakkala-Weckström 2005: 227). 10 See Pearsall 1985: 274-274 for a discussion on different interpretations on the narrator’s position. 69 other strategies, such as control of the floor, or subtle impoliteness, which provide women with a modicum of power in a marital relationship. Part III: Desdemona, Othello and Politeness Just as Chaucer develops Griselda’s use of politeness features compared to his source, so does Shakespeare regarding Desdemona’s use of politeness features when compared to the character Disdemona in his source text, Cinthio’s Gli Hecatommithi. In the prose source, there is relatively little reported speech of any kind, and Disdemona’s politeness is mainly indicated through narrator description, in phrases like “The lady, all courtesy and modesty, replied” (245) or “she said humbly” (245). The most significant change Shakespeare makes to his source is in forms of address: Shakespeare uses more of them, and uses them to emphasize power and status differences between his heroine and her male interlocutors. Cinthio’s Disdemona addresses the Moor as “my Moor” (242) and “husband” (243), terms emphasizing either nationality or family relationship, while Desdemona’s forms of address continually emphasize her subordinate position, usually variations on “my lord.” Desdemona varies her use of politeness strategies somewhat depending on whom she is speaking to, and also on her emotional state: she mainly uses negative politeness strategies when unaroused, but when angry or frightened she tends to favor bald on record strategies and also take more speaking turns. In order to establish a context in which to examine Desdemona’s conversations with her husband, let us first examine the linguistic strategies she uses with other characters. With her father, Desdemona mainly uses negative politeness strategies. Her first words in the play, when called upon to address the Venetian court and defend her marriage, are addressed to him, whom she calls her “noble father” and her “lord of duty” (1.3.180, 184). With her father, she uses other strategies as well to show deference: she uses the polite form “you” (182, 184), she depersonalizes herself, “My life and education learn me” (183), and also emphasizes her father’s power over her. Her only direct challenge to her father is softened with a modal, “So much I challenge that I may profess / Due to the Moor 70 my lord” (1.3.188-9). This negative politeness seems connected to Desdemona’s wish to placate her father, now that she has secretly married. She has been subservient to him, but now must also be subservient to her husband, and if forced to choose between them, she will choose Othello. Her negative politeness here functions to make this transition as easy as possible for her father. Desdemona’s position vis a vis the Duke is even more complex: nominally he is her ruler, but his bond is purely legal. Interestingly, when the power dynamic is governed by legal, rather than familial bonds, Desdemona’s politeness strategies shift slightly. In addition to negatively politeness, Desdemona also uses positive politeness, as well as upwards compliments (see Holmes 1995: 119). When asking to be allowed to follow Othello to the war, she begins with a positive terms of address, “Most gracious Duke,” which she follows with a compliment about his “prosperous ear” (245), as well as a further compliment that his voice has “charter” which can assist her “simpleness” (24-7). In the famous speech begging the Duke to allow her to accompany Othello to the war, she pauses in the middle for a positive term of address, “dear lords” (256), thus softening her tone and preparing for her request. Desdemona’s relationship with Othello, as manifested by her use of different linguistic strategies, varies a great deal from the beginning to the end of the play. When they are reunited in Cyprus, she mainly plays the role of dutiful, and silent, wife. Othello’s relative power in this scene can be seen in the way he dominates the interaction: of 18 total lines, he speaks all but 2.5. Desdemona mainly uses positive politeness strategies in this first scene. Both address the other using positive terms of address: he says “O my fair warrior!” and “O my soul’s joy!”, while she calls him her “dear Othello!” (2.1.180, 82). Both also include the other in their talk, using “our” constructions. Desdemona says, “The heavens forbid / But that our loves and comforts should increase / Even as our days do grow (191-93), which Othello responds to with a comment about “our hearts” (197). This power structure is reversed in their next major exchange, where Desdemona begs to have Cassio reinstated in 3.3 (quite a FTA from Othello’s 71 point of view).11 Of these 49 lines of interaction, Desdemona speaks 39.5 lines, leaving Othello with 9.5. This is the only exchange where Desdemona so clearly both introduces the topic and controls the floor, though here she also spends a lot of energy to soften the verbal force of her demands. She addresses her husband with positive forms such as “my lord,” “good my lord,” or “good my love” and deprecates herself, saying for example that she “has no judgment in an honest face,” thus implying that Othello has good judgment; this behavior can be interpreted as another instance of upwards complimenting. She softens her request that Othello meet with Cassio by saying “I prithee,” and similarly appeals to in- group solidarity by referring to “our common reason” which could judge Cassio’s fault. Desdemona is indeed quite self-aware of using such politeness strategies to soften her requests, as she tells Othello: (3) Nay, when I have a suit Wherein I mean to touch your love indeed It shall be full of poise and difficult weight And fearful to be granted. (3.3.80-83) As we might expect, at some critical moments Desdemona chooses not to redress her FTAs, as in their fight just after Othello has been summoned to return to Venice. Desdemona says that she’s glad that Othello’s been called back, leaving Cassio in charge. After a short fight, Othello yells “devil,” and according to the stage direction strikes her. Desdemona replies, with no effort at redress, “I have not deserved this” (4.1.240), but then she begins to weep, prompting Othello’s remark about crocodile tears. In the climactic scene where Othello directly accuses her of being a whore (4.2), two changes are evident, although in this scene as previously, Desdemona still uses a number of negative politeness strategies, most notably forms of address such as “my lord,” which is repeated several times. First, both characters resort to distancing strategies, but as emotional tension builds, they both (predictably) make less effort to redress their FTAs. For example, when Othello tells Desdemona to swear that she is honest, she says “Heaven doth truly know it” 11 For another analysis of this scene, see Vimala Herman 1995: 232. 72 (4.2.39) rather than a simple declarative sentence such as “I am honest.” Later, after further accusations on Othello’s part, she moves even farther away from declaring her own chastity, saying “I hope my noble lord estems me honest” (66). Two aspects are important here: the most obvious is that Desdemona knows that declaring her own honesty is irrelevant as long as her husband doesn’t share her opinion – a mere speech act on her part cannot make her “honest” if her husband doesn’t believe she is. A second difference is that her most typical strategy of negative politeness, calling Othello her “noble lord” is here shifted to a third- person statement, rather than a direct address. A similar distancing effect can be seen in Othello’s speech a few lines later, where she becomes a “book” where he will write the word “whore” (73). Politeness strategies in general seem to be used to protect the person, so when communication breaks down, bodies themselves get more and more removed from the discussion. At the emotional climax of the scene, Desdemona makes no effort to redress her responses to Othello’s accusations: (4) Othello: Are you not a strumpet? Desdemona: No, as I am a Christian! . . . . Othello: What, not a whore? Desdemona: No, as I shall be saved! (4.2.83-88) Desdemona cannot even bring herself to name the thing which Othello accuses her of, as she says to Iago: “Am I that name, Iago?” (4.2.120).12 Their final scene includes many gestures familiar from earlier in the play, and is also interesting in that it is the one interaction in which Desdemona and Othello most equally share the floor: of a total of 60 lines of interaction, Desdemona speaks 29, while Othello speaks 31. The character of the sequence is also quite different from earlier ones, with very few long speeches, and many half lines shared between speakers. As in the previous scene, Desdemona starts with negative politeness, twice addressing her husband as “my lord.” But as the 12 See also Jardine 1996: 19-34 73 argument gains fury, and she begins to understand that Othello means to kill her, she drops all politeness, responding baldly to Othello’s accusations: (5) Othello: O perjured woman, thou dost stone my heart And makest me call what I intend to do A murder, which I thought a sacrifice! I saw the handkerchief. Desdemona: He found it then. I never gave it him. Send for him hither, Let him confess a truth (5.2.63-68). At these highly emotional moments, ordinary social skills do not come into play, even for Desdemona who seems to have internalized particularly negative politeness to a very high degree: here she speaks bald on record (Brown & Levinson 1987: 94-101), in simple and short declarative sentences, leaving no room for doubt about her innocence. Yet, we can nevertheless see the extent of Desdemona’s internalization of politeness strategies in her final lines, when she replies to Emilia about who has done this to her: “Nobody. I myself. Farewell. / Commend me to my kind lord -- O, farewell” (5.2.122-23). In her dying moments, she defends her husband, whom she now calls a “kind lord,” echoing her extensive use of “good lord” or “my lord” throughout the play. Moreover, she virtually takes a formal leave-taking of him through Emilia, a leave-taking which becomes almost ironic in comparison with similar leave-takings in The Tragedy of Mariam, where Salome tells her husband to “fare ill” instead of “farewell.” In short, probably due to her undying devotion to her husband, Desdemona’s speeches to Othello are consistently marked with elements of both positive and negative politeness, though no strategy can save her from her husband. Part IV: The Tragedy of Mariam Like Desdemona, Mariam is aware of her power to shape her language to deal with her husband, though she mainly chooses not to soften her words with politeness strategies. Early in The Tragedy of Mariam, in a conversation with Sohemus, who has just revealed to Mariam that Herod, in a fit of jealousy, has ordered that if he should die Mariam should be killed straightaway so nobody else 74 could ever marry her, Mariam explicitly calls attention to her ability to shape Herod's responses to her through her looks, behavior and language: (6) I know I could enchain him with a smile: And lead him captive with a gentle word (3.163-4) But now, having heard of Herod's cruel intentions, which come on top of the fact that he has had her grandfather and brother murdered, she vows in effect to stop playing the politeness game with her husband: (7) I scorn my look should ever man beguile, Or other speech than meaning to afford (3.165-6). In other words, she will only say what she means, without regard for the feelings of the listener, and so decides that when she meets Herod, she will not pretend a love she no longer feels. The meeting of Mariam and Herod provides fascinating insight into the politeness strategies used when one person holds absolute power over the other (as Herod does over Mariam as her husband and political ruler), while the so- called weaker member wields emotional power over the other, for Mariam in this scene angrily announces that she is going to withhold the one thing Herod really wants, her love. Herod’s power over Mariam can be seen both in the way he controls the floor (of the 122 total lines of their interaction, Herod speaks 103, or 84.4%) and in the way he initiates dialogue and selects topic. Much of their exchange is shaped around questions which he asks and she answers, and the single time Mariam can be said to introduce a topic is when she responds to a question about why she is so angry by saying that she cannot love her husband because he has killed both her brother and grandfather (4.112-16), which provokes a twenty-line response. Despite the obvious disparity in length of speaking turns, Mariam controls this scene rather as men have been found to control marital interaction through non-cooperation in talk, in her case by giving what Herod perceives as inadequate responses13 and by refusing to respond to his profuse positive politeness even with 13 See Coates 2004: 120-124. 75 negative politeness. Herod’s use of especially positive politeness features is especially striking. Throughout the scene, he continually uses terms of endearment when addressing his wife, such as “my dear,” “my best and dearest half,” “dearest Mariam,” and “Jewery’s queen, and Herod’s, too,” even after it's become clear that Mariam is intent only on chastizing him and he begins to threaten her. Other positive strategies include expressing concern for her wants, stressing their mutual bond, promising her gifts (including her own kingdom, which he can then conquer), and complimenting his wife. Mariam almost grudgingly calls him “my lord” at the beginning of her first speech to him, but otherwise she does not use any other politeness strategies (by contrast, she refers to herself as “wretched Mariam”). In addition to politeness, Herod uses other linguistic features as well to ingratiate himself with Mariam. For example, Herod uses the familiar thou form at the beginning of the scene, switching to you only after he is already thoroughly angered and begins to threaten his wife: “I will not speak, unless to be believ’d / This forward humour will not do you good: It hath too much already Herod griev'd, To think that you on terms of hate have stood.” Note, however, that even these threats are undercut by Herod’s admission that he truly loves his wife. Mariam, by contrast, answers Herod’s “thous” with “you,” thus emphasizing the distance she is seeking to create between them. Also, even in his darkest moments, Herod uses a modal to soften his final warning: “Nought is so fix’d, but peevishness may move.” Mariam, by contrast, criticizes him baldly, without any hedging, saying for example that she won’t build on the “unstable ground” of his love. He also tempers his threats with hedges and explanations, softens commands with various linguistic devices, offers extended apologies for his behaviors, and finally declares his love quite frankly, thus making himself even more vulnerable to her. He does this in almost a fugue of good and bad will, forgiving her and then becoming resolute in his anger, torn between love and anguish. 76 Part V: Conclusion We hope these examples suggest some of the complexities involved in analyzing the relationship between gender, language, power and violence in these texts. One aim of this analysis has been to highlight the distinction between powerlessness and passivity in these characters: while Griselda and Desdmona, in particular, are often held up as supreme examples of female passivity, we hope this analysis demonstrates that while they lack power, both physically and culturally, they are not passive in any simple sense. Both Griselda and Desdemona make use of a wide range of linguistic strategies, especially politeness, in their efforts to appease their husbands, and indeed they abandon strategies which are not working, and try out new strategies when necessary. Within the confines of gendered power relations, they are linguistically creative, flexible and determined.14 Griselda’s use of such strategies shows remarkable insight into the psychological workings of her husband, while Mariam courageously decides not to use such strategies, knowing full well the psychological implications of this choice on her husband. Another important issue is the role of the emotions in shaping linguistic politeness strategies. In Chaucer’s tale, Griselda’s emotions are revealed precisely by her switch from negative to positive politeness, as feelings long held in check come closer to the surface. Crucially, at these moments of highest passion, Chaucer mainly focuses on her love for her children rather than her feelings for her husband, which he, like Walter, has painstakingly suppressed. At the same time, Chaucer, again like Walter, seems to want his happy ending: no matter how completely the wife’s emotions are ignored (through the supposed deaths of her children, her own threatened loss of marital status), she is expected to recover quickly and forgive her husband. Being a dramatist, Shakespeare is more interested than Chaucer in exploring both sides of his marital conflict, and thus Desdemona is given more space to express her own feelings and defend herself. Like Chaucer, however, there seems to be an odd kind of male fantasy at work at the end of Shakespeare’s play, where Desdemona is shown forgiving her husband 14 Edward Pechter (2003) similarly argues that Desdemona is linguistically more active than she is usually given credit for, though he does not focus on specific linguistic strategies. 77 for killing her, reverting back to negative politeness strategies. This she does, we are led to believe, out of overwhelming love for him, the kind of devoted wifely adoration which helps to transform an angry, gullible husband into a tragic hero. Elizabeth Cary, by contrast, examines a wife who chooses rudeness over politeness, who chooses not to use linguistic survival strategies because they are somehow beneath her. Finally, this analysis suggests that we might more carefully explore the links between unresolved face needs, sexual jealousy leading to marital violence, and gendered politeness strategies. While patriarchal structures generally seem to focus on enabling men to act unimpededly (leading in modern society to a preference for negative politeness), marriage and sexual relationships also include a strong component of needing to feel appreciated, sexually desirable, loved. If sexual jealousy is primarily experienced as a threat to positive face, this might suggest that positive politeness would be the more effective strategy in dealing with difficult husbands. And indeed, in our comparison, only Griselda, who uses positive politeness at emotionally-powerful moments, succeeds in mollifying her husband (although Griselda’s “success” is as much a question of genre as effective choice of strategy: in a tragedy, no linguistic strategy is going to be effective). Sexual jealousy, however, more probably arises from a powerful conflict between the need to feel appreciated and the need to control the other, such as we see more visibly in Othello and The Tragedy of Mariam. This would call for a more flexible, nuanced approach on the part of the female heroines, and indeed linguistic strategies might not always be the best defense, since there is a rather large risk that the wrong strategy would exacerbate rather than improve the situation (e.g. if the woman emphasizes subservience at a point where the better emotional response is to emphasize solidarity and affection). In any case, we hope that by paying special attention to those places in literature where characters wrestle with powerfully conflicting feelings of love and hate, we can perhaps learn more about the dynamics of marital violence, and work towards new solutions to this very old social problem. 78 References Bargiela-Chiappini, F. 2003. “Face and politeness: new (insights) for old (concepts).” Journal of Pragmatics 35. 1453-1469. Benson, L. D. 1987. The Riverside Chaucer. Oxford: Oxford University Press. Brown, P. and S. Levinson. 1987. Politeness. Some Universals in Language Usage. Cambridge: Cambridge University Press. Brown, R and A. Gilman, 1989. “Politeness theory and Shakespeare’s four major tragedies.” Lang. Soc. 18: 159-212. Cary, E. 1994. The Tragedy of Mariam. B. Weller and M. W. Ferguson (eds). Berkeley, CA: University of California Press. Cinthio, G. 1973. “Gli Hecatommithi.” In Narrative and Dramatic Sources of Shakespeare, volume 7. Geoffrey Bullough (ed), 239-252. London: Routledge. Coates, J. 2004. Women, Men and Language: A Sociolinguistic Account of Gender Differences in Language. 3rd edition. Harlow, England: Pearson. Culpeper, J. 1996. “Towards an anatomy of impoliteness.” Journal of Pragmatics 25. 349-367. Ervin-Tripp, S, M. C. O’Connor and J. Rosenberg. 1984. “Language and power in the family.” In: C. Kramarae, M. Schulz and W. M. O’Barr (eds.): Language and Power. Beverly Hills, London & New Delhi: Sage Publications. 116-135. Farrell, T. J. and A. Goodwin. 2002. “The Clerk’s Tale”. In Correale, R. M. and M.Hamel (eds.). Sources and Analogues of the Canterbury Tales. Cambridge: D.S. Brewer. 101-167. Hart, D. 2000. “Power and Powerlessness in Shakespearean Dialogues. In Gabriella Di Martino and Maria Lima (eds.). English Diachronic Pragmatics. Napoli: CUEN. 209-226. Held, G. 1992. “Politeness in linguistic research” In Watts, R J., S. Ide and K. Ehlich (eds.). Politeness in Language. Studies in its History, Theory and Practice. Berlin and New York: Mouton de Gruyter. 131-155. Herman, V. 1995. Dramatic Discourse: Dialogue as Interaction in Plays. London: Routledge. Holmes, J. 1995. Women, Men and Politeness. London: Longman. Jardine, L. 1996. “’Why should he call her whore’: Defamation and Desdemona’s Case.” In Reading Shakespeare Historically, 19-34. London: Routledge. Kopytko, R. 1993. Polite Discourse in Shakespeare’s English. Poznan: Adam Mickiewicz University Press. Kopytko, R. 1995. “Linguistic Politeness Strategies in Shakespeare’s Plays.” In Historical Pragmatics: Pragmatic Developments in the History of English. Andreas H. Jucker (ed), 515-40. Amsterdam: John Benjamins. Locher, M. A. 2004. Power and Politeness in Action. Disagreements in Oral Communication. Berlin and New York: Mouton de Gruyter. Magnusson, L. 1992. “The Rhetoric of Politeness and Henry VIII.” Shakespeare Quarterly 43: 391-409. Mills, S. 2003. Gender and Politeness. Cambridge: Cambridge University Press. 79 Nevala, M. 2004. Address in Early English Correspondence. Helsinki: Sociéte Néophilologique. Nevalainen, T. and H. Raumolin-Brunberg. 1995. “Constraints on Politeness: The Pragmatics of Address Formulae in Early English Correspondence.” In Historical Pragmatics: Pragmatic Developments in the History of English. Andreas H. Jucker (ed), 541-601. Amsterdam: John Benjamins. Pakkala-Weckström, M. 2005. The Dialogue of Love, Marriage and Maistrie in Chaucer’s Canterbury Tales. Helsinki: Sociéte Néophilologique. Palander-Collin, M. 1999. “Male and Female Styles in 17th Century Correspondence: I THINK.” Language Variation and Change 11 (1999): 123-141. Palander-Collin, M. 2002. “Tracing Patterns of Interaction in Historical Data.” In Variation Past and Present. 117-134. Helsinki: Société Néophilologique. Pearsall, D. 1985. The Canterbury Tales. London and New York: Routledge. Pechter, E. 2003. “’Too Much Violence:’ Murdering Wives in Othello.” In Women, Violence, and English Renaissance Literature, L. Woodbridge and S. Beehler (eds), 217-242. Tempe, Arizona: Arizona Center for Medieval and Renaissance Studies, volume 256. Shakespeare, W. 1997. Othello. E. A. J. Honigmann (ed). The Arden Shakespeare, 3rd edition. London: Thomas Nelson & Sons Ltd. Vasvári, L. “‘Buon cavallo a mal cavallo vuole sprone, e buona femina e mala femina vuol bastone.’ Medieval Cultural Fictions of Wife-Battering.” In A. Classen (ed.). Discourses on Love, Marriage, and Transgression in Medieval and Early Modern Literature. 313-336. Tempe, Arizona: Arizona Center for Medieval and Renaissance Studies. Watts, R. J. 2003. Politeness. Cambridge: Cambridge University Press. Winny, James (ed.). 1965. The Clerk’s Prologue and Tale. Cambridge: Cambridge University Press. Wolfson, N. 1988. “The bulge: A theory of speech behavior and social distance.” In Jonathan Fine (ed.). Second Language Discourse: A Textbook of Current Research. Norwood, N.J.:Ablex. 21-38. Woodbridge, L. 2003. “Introduction.” In Women, Violence, and English Renaissance Literature, L. Woodbridge and S. Beehler (eds), xi-xlix. Tempe, Arizona: Arizona Center for Medieval and Renaissance Studies, volume 256. 80 La dérision comme acte de langage de l’art belge francophone Jean-Yves Malherbe, Université de Jyväskylä Dans des domaines aussi divers que la géopolitique, la linguistique, les arts ou l’Histoire officielle de la Belgique, un acte de langage original transparaît. Un des caractères les plus troublants de cette nation réside en effet dans la capacité de ses habitants à décaler la réalité quotidienne afin de dévoiler un tabou, celui de la fragilité des piliers de la vie communautaire, par le biais de manifestations communicatives trop réitératives pour résulter d'une simple coïncidence. Leur objectif avoué ou inconscient sera de déstabiliser des convictions intimes comme celui de la vérité sémantique et d’outrepasser des normes cognitives inculquées pendant les années d’apprentissage de la vie en société. Il s’agit en quelque sorte de la formulation d’une dérision ontologique inhérente à ce pays et à ses coutumes. L’article qui suit, basé surtout sur le point de vue francophone, a une ambition, celle d’illustrer les propos qui précèdent à travers un aperçu rapide de l’histoire belge et par des exemples issus d’arts divers dont l’originalité est qu’ils allient le plus souvent l’écriture au graphisme, qu’il s’agisse de poésie, de bande dessinée ou de peinture. 1. Le dérisoire du passé belge Dans l’Histoire, la dérision sera basée sur l’irruption d’un terme appartenant au réalisme quotidien ou dialectal dans le registre traditionnellement emphatique des faits militaires.1 Comparée à l’Histoire de France refaçonnée par des images d’Epinal, l’histoire belge abonde en évènements tragicomiques. C’est ainsi que l’on mentionne une 1 Voir note 4 ci-dessous. Les remarques concernant l’histoire de la Belgique sont issues de sources variées, en particulier l’œuvre de Xavier Mabille (pour les trois derniers siècles) et les sites Internet cités en bibliographie. Une approche particulièrement riche est celle des grands mythes de l’histoire de Belgique (sous la direction d’Anne Morelli) qui classe un grand nombre d’évènements liés à la thématique de cet article. 81 guerre de la Vache, celle qui a éclaté en 1275 à propos d’un animal volé retrouvé peu après dans une foire ; ou bien des dénominations aussi éloquentes que celles présidant à des conflits qui opposèrent jadis des Awans à des Waroux ou bien des Chiroux à des Grignoux. La bande dessinée a fait connaître les Gaulois dans le monde entier, et donc les débuts officiels de l’Histoire de France telle qu’elle a été enseignée à partir de la Troisième République. La Belgique s’enorgueillit de son côté de ses Eburons et de ses Nerviens dont les chefs Ambiorix et Boduognat n’ont rien à envier à Vercingétorix dans la lutte à l’envahisseur romain. L’illustration de telles images légendaires aboutit à une vision amusée des prétentions historiques des grands voisins. Si l’occupant espagnol traite de « gueux » les révoltés du peuple belge, ceux-ci s’en enorgueilliront et se saisiront de ladite dénomination en en valorisant socialement la signification. 2 Même dans l’Histoire proche, le sentiment d’une fatalité tragicomique subsistera : dans quel autre pays européen y a-t-il eu une véritable révolution d’opérette comme celle de 1830 où de jeunes nationalistes sont sortis dans la rue au son de La Muette de Portici d’Auber ? Quel autre pays modeste européen héritera miraculeusement d’une énorme colonie africaine (le Congo) par le biais d’accords entre un « petit » roi (Léopold II) et un « grand » aventurier (Stanley) ? L'ironie est sans cesse présente dans la conscience exhibée de sa propre faiblesse et de résultats absurdes auxquels le pays semble confronté sans l'avoir vraiment désiré. Elle a également laissé des codes qui, au-delà du cas belge, construisent une dimension ontologique aux réalités universelles tant du passé que du présent. Il est clair qu'une histoire aussi excentrique a été une source importante de l’art visuel belge, en particulier dans la bande dessinée d’Hergé (Tintin) ou de Franquin (Spirou), et avant tout celle de Willy Vandersteen, un auteur flamand dont la série la plus connue, Suske en Wiske, a été traduite sous les titres français Bob et Bobette, suédois Finn och Fiffi et finnois Anu ja Antti. Dès le deuxième titre, Op het eiland Amoras (L’île d’Amphoria, 19473), dont l’action est située en 2 Cette valorisation est évidente dans l’œuvre fondatrice de la littérature belge qu’est La Légende d’Ulenspiegel de Charles de Coster. 3 La traduction finnoise Amforian saari (1987) correspond à une réédition de 1967. 82 1641, le héros se trouve mêlé à une lutte opposant les gros et les maigres, lutte qui rappelle du côté francophone la guerre opposant au XVIIe siècle les Chiroux de l’aristocratie, habillés de noir et de blanc comme des hirondelles4, et les Grignoux, autrement dit les Grincheux, représentant le peuple. D’autres titres font référence de manière irrévérencieuse à la culture, qu’il s’agisse de peinture comme dans le cas du tableau de Breughel De dulle griet (Margot-la-folle) ou celui des ateliers de l’école flamande (De raap van Rubens / Le rapin de Rubens) ; du soulèvement populaire contre le Duc d’Albe, qui est déjà à l’origine de Till Eulenspiegel, le roman-phare de la littérature belge, et que l’on retrouve dans Het spaanse spook (Le Fantôme espagnol) ou De zingende kaars (Le Flambeau chantant) ; de l’histoire des anciennes tribus belges (De nerveuze Nerviërs / Les Nerviens nerveux) ; de celle des clans (De stalen bloempot / Le Pot aux roses) ; voire enfin de curiosités touristiques célèbres (Het kregelige Ketje / Manneken Pis, l’irascible)5. L'humour basé sur l'humilité exhibée se retrouve également dans les titres de magazine. Un magazine intitulé « Petits Belges » a vu le jour en 1920 et l'affectivité apportée par l'adjectif sera reprise fréquemment, jusque dans les titres de recherche, comme Le petit Belge avait vu grand : une littérature coloniale de Pierre Halen (1990)6. Dans le fond, une position humble n'est-elle d'ailleurs pas un avantage pour se mettre en valeur face à des géants : c'est l'origine d'un certain humour tchèque par exemple, et la Finlande (petite elle aussi comme chacun le sait) n'est-elle pas arrivée bonne seconde derrière l'URSS pendant les tristes années de guerre ? Apparaissent ainsi des traits originaux qui tissent une trame particulière à l’histoire du peuple belge : l’absurdité de hauts faits historiques, un peuple épicurien et idéaliste, porté à la révolte contre les injustices, un peuple dominé mais jamais soumis dans le passé. Il en résulte aujourd’hui une entité complexe, divisée linguistiquement, culturellement, économiquement, dont les frontières 4 Les termes wallons jouent ici un grand rôle : « Chiroux » signifie « hirondelles » en wallon, « grignoux » comme indiqué correspond au français « grincheux » et « Spirou » est un écureuil. 5 Pour plus de précisions, voir le site internet néerlandophone « suskeenwiske » cité en bibliographie. 6 Le titre de l’ouvrage est commenté par Pierre Halen en introduction (1993: 29-30). 83 extérieures ont été multipliées par la création du Benelux et la position centrale de Bruxelles en politique internationale (UE et OTAN). Un tel concept, enclavé entre les cultures dominantes de grands pays comme l’Allemagne et la France, semble ne posséder aucune autonomie de fait si ce n’est la présence d’un roi, d’un gouvernement et d’éléments coagulants dans le domaine sportif ou culturel. La question de la réalité de son existence se pose sans cesse: l’importance de genres paralittéraires comme le fantastique ne sont pas un simple effet du hasard.7 2. Ludisme contre purisme La situation géopolitique et surtout linguistique de la Belgique a conduit à des réactions originales de la part de nombreux intellectuels et artistes. La première réaction sera celle de grammairiens, en particulier Maurice Grevisse qui publiera dès 1936 son célèbre Bon usage à l’intention de ceux qui hésitent sur la norme, un ouvrage sans cesse corrigé et réédité depuis cette date. Pour cet amoureux de la langue française, il ne s’agit pas tant de règles contraignantes que de conseils à suivre. Le point de vue est celui d’une langue idéale, et l’idée d’un exorcisme particulier à la conception négative du français parlé en Belgique vient naturellement à l’esprit. La langue paraîtra alors comme un objectif plutôt qu’un instrument, un objet débarrassé de ses imperfections, pur et incontournable. La seconde réaction, l’apparition du ludisme, fait l’objet de notre article. La tradition comique de l’enfant éternel et du bon vivant était présente dès l’origine de la littérature de fiction francophone de Belgique dans La Légende et les Aventures héroïques, joyeuses et glorieuses d’Ulenspiegel et de Lamme Goedzak au pays de Flandre et ailleurs du Flamand Charles de Coster (1858, ouvrage publié en 1867)8. Loin de la rigueur cartésienne, ce roman est un discours rappelant le carnavalesque que Bakhtine a analysé chez Rabelais, et dont la forme souvent poétique, onirique ou dialectale prône au fond la liberté artistique et le droit à la justice et à la différence. Cette œuvre inclassable appartient à tout un pan 7 L’article a été écrit avant la crise politique de 2007. 8 La légende de Thyl Eulenspiegel possède de nombreuses versions, en particulier germaniques, voir par exemple le début de la lecture de Jean-Marie Klinkenberg (1983). 84 populaire de la culture belge, celui du droit aux plaisirs offerts par la vie qui est si bien représenté par des artistes comme Breughel. Le moteur artistique en est le jeu, c’est-à-dire la mise en cause à la fois amusée et lucide d’un ordre établi. Ce n’est pas un hasard si Nele, l’héroïne, a pour mère une folle visionnaire. La folie comme le rêve font partie de l’évasion du quotidien, et aboutissent en réalité au droit à la liberté de l’expression, même si ces concepts restent liés à un besoin de non-conformisme. Folie visionnaire comme celle des tableaux de Hyeronimus Bosch ou épicurisme matériel à la Brueghel concourent à une philosophie de l'inattendu au moment présent. Nous aboutissons ainsi à un carrefour concrétisé par le schéma suivant. Son origine est belge, une conférence donnée à Jyväskylä en 1992 par l’écrivain et enseignant de philosophie Jacques Sojcher dans le cadre d’une exposition consacrée aux « Irréguliers du langage ». Schéma 1 : orientations linguistiques de l'art belge Les puristes de la langue (Bon Usage) (recherche de sécurité, langue normative unique) Littérature intime, vie courante L'attrait de Paris (tradition, modes) Attrait de l'irréalité (paralittérature : fantastique, féerie) Irréguliers du langage : la part riche et absurde de l'écriture L’art belge est ainsi à une croisée des chemins : faut-il suivre la tradition française imposée par l’attrait de Paris, la métropole des plus célèbres éditeurs francophones, ceux qui offrent des possibilités plus importantes de distribution, ou bien utiliser la couleur locale dans des œuvres plus intimes ou proches du quotidien, se servir de l’imaginaire et des portes ouvertes par le fantastique et la féerie? Faut-il enfin exprimer sa révolte envers la norme et déstabiliser des domaines comme la langue ou la convention (voir schéma 1)? En d’autres termes 85 plus directs, faut-il user de nuances ou au contraire chercher le sensationnel et l’exhibitionnisme? Une analyse plus détaillée dans divers domaines peut déjà fournir un certain nombre de réponses. Qu'il s'agisse de création lexicale ou métaphorique, comme dans les chansons de Jacques Brel (bruxelliser, un canal qui se pend dans le « plat pays »), qu'il s'agisse d’écoles artistiques dépassant le cadre des frontières dont l’effet a été parfois retardé en Belgique, comme un surréalisme vivace ou le dadaïsme de Clément Pansaers, qu’il s’agisse de la quête de l’irrégularité (par exemple l’introduction du texte dialectal wallon) chez André Blavier9, l’attrait de l’interdit semble souvent primordial dans le texte écrit. Il est également essentiel dans le texte mêlé à l’image, comme le montrent les difficultés lexicales des héros d’Hergé ou de Peyo dans la bande dessinée, et comme le montrent également les constructions picturales de personnalités aussi diverses que René Magritte et Christian Dotremont. Et que penser des textes créés à partir de néologismes et d’onomatopées comme « Le grand combat » ou « glu et gli » d’Henri Michaux dans le recueil Qui je fus de 1927? 3. Surréalisme, dadaïsme, wallonisme et onomatopées L’enthousiasme pour la modernité et par conséquent le désir de rejet du modèle ancien font partie des critères de l’histoire de la progression artistique. Dans le cas des intellectuels francophones belges, il faut toujours saisir l’importance qu'ils ont donnée aux nouvelles tendances françaises, et plus précisément parisiennes, pour mieux comprendre leur histoire littéraire depuis la moitié du XIXe siècle10. Une première remarque préliminaire concerne un phénomène temporel : il se produit parfois un décalage entre l’essor littéraire de certains mouvements en France et son écho en Belgique, comme si la réaction, positive mais réfléchie, se concrétisait lentement pour plus de durée. Le cas le plus net est celui du surréalisme. Un autre décalage, plus insidieux, aura lieu, et concernera les objectifs personnels. Là où le surréalisme français cherchera des images destinées à frapper les imaginations et les subconscients, les surréalistes belges les 9 Un exemple, donné par Noël Artaud dans Les très riches heures d’André Blavier, commence ainsi : Pus tèribe qui l’délûge, li feû lès porsûrè (page 107). 10 Voir à ce sujet l’histoire des avant-gardes littéraires de Belgique parues chez Labor en 1991. 86 saupoudreront d’humour et de dérision. On pourrait par conséquent ajouter dans leur cas une dimension sociologique et ludique, sans doute moins sophistiquée que dans des œuvres françaises des années 20, plus proches de la démonstration, comme Entracte de René Clair ou Le Paysan de Paris de Louis Aragon. Les traits oniriques qui rayonnent de ce film et de ce roman manquent peut-être d'un besoin exacerbé de communication et de promiscuité plus tangible dans les textes et les images belges. Si un groupe surréaliste belge se forme autour de Paul Nougé dès les années 20, son action subversive est multiforme et survivra à la décomposition de ses modèles français. Paul Nougé, René Magritte, Louis Scutenaire, Marcel Lecomte, Paul Colinet, plus tard rejoints par Marcel Mariën, utilisent des moyens fort variés pour déstabiliser la société : détournements de textes (pastiche, plagiat), supercheries, mise en scène publique de la poésie sous forme d’aphorismes proposés par des hommes-sandwiches dans les villes belges, film désacralisant le Christ, refus de la banalité par la valorisation de certaines classes sociales tenues à l’écart comme celles des mendiants, des souteneurs et des bagnards. Lié dès le début de son action au parti communiste belge, créé en 1919, le mouvement aura une action politique mais saura se démarquer par l'arme du rire. Dans les années 30 se crée le groupe politique « Rupture » autour d'Achille Chavée au moment où le mouvement français s'essouffle, confronté à des conflits internes qui ont abouti à de nombreuses exclusions (Antonin Artaud, Georges Bataille, Paul Eluard, Louis Aragon). Pendant l’occupation, Nougé donnera son appui à des expositions à Bruxelles (photographies de Raoul Ubac) comme en France (tableaux de Magritte) qui seront rapidement interdites. Sans pour autant se défaire des circonstances, les surréalistes belges prôneront une liberté proche de l’anarchisme tout en s’opposant à l'ordre contraignant, qu'il s'agisse du fascisme quotidien ou de 87 ce que tous ces révoltés considèrent comme asservissement linguistique et culturel.11 Si l’action surréaliste belge rejoint ainsi l’univers ludique et irrévérencieux envers la norme de nombreux artistes belges, elle a suivi également une autre école plus nihiliste, celle du dadaïsme représenté par Clément Pansaers. Pour mieux comprendre ce besoin de renouveau, au-delà des effets incontestables de la Grande Guerre sur la jeunesse sacrifiée d'une génération, il faut tenir compte de facteurs artistiques et d'écoles parfois outrancières dans leurs objectifs avoués. L'enthousiasme intellectuel tangible dans les trois décennies qui ont précédé la première invasion par l'armée allemande de la Belgique du populaire Albert 1er a donné lieu à des courants de pensée qui refoulaient l'intimité au profit du spectacle. Le symbolisme du futur nobelliste Maeterlinck s'opposait au courant expressionniste, la modernité positive et l'attrait de la mécanique nouvelle permettait une nouvelle victoire de l'homme sur son destin dans l'esprit futuriste. En contrepartie, les exemples étrangers de Huysmans et de Wilde donnaient une vision parfois morbide d'un univers en décadence, lié au passé et peu enclin à imprégner une Belgique éloignée de tout relent aristocrate. La Grande Guerre avait bouleversé les normes. Avant l’avènement du surréalisme, une vague de protestation née de l'absurdité des décisions prises par les politiques et les gradés a mené au refus de notions inculquées jusque là, celles par exemple de l’héroïsme patriotique et de la guerre en tant que rencontre traditionnelle entre clans bien établis. La Grande Guerre a causé la naissance de générations perdues sans illusions concernant leur futur malgré l’apport neuf de la psychanalyse et l’idéalisme créé par l’internationale socialiste. Quelle arme les intellectuels pouvaient-ils utiliser ? Ce fut encore la dérision, et ce par la manifestation du nihilisme et de la subversion. Né en 1916 à Zurich (Tsara, Hugo Ball, etc.), le mouvement Dada veut récupérer les débris de la civilisation 11 L’activité surréaliste belge a longtemps survécu à la seconde guerre mondiale. Voir à ce sujet le numéro 8 de la revue Textyles et le magazine Phantomas, né le 15 décembre 1953 et dont le dernier numéro a paru le 15 décembre 1980. 88 occidentale et détruire les normes esthétiques dont le langage fait partie. Le maître-mot est celui de régression, d'un retour à une langue infantile, d’un refus de toute marque intellectuelle par la juxtaposition de termes sans liens, par l'emploi de termes dévalorisants, par l'emploi de mots tabous et d'associations provocatrices. Picabia, Marcel Duchamp, Man Ray sont des artistes à la mode à Paris en 1920. Prenons deux extraits du poème Pan-pan de Clément Pansaers, seconde partie d’un ensemble intitulé Le pan-pan au cul du nu nègre (1919). Le pan-pan était alors une danse à la mode. En plein feu d’artifice nous dansons le Pan-pan Pan-pan — Pan-pan qu’un jour tu ailles en prison, tu jureras : à dix-huit mois mon père me le chanta : Pan-pan — Pan-pan qu’un jour tu ailles à l’hôpital, tu cracheras : à dix-huit mois mon père me le blasphéma : Pan-pan — Pan-pan qu’un jour l’on t’enferme en une maison d’aliénés tu chanteras : à dix-huit mois, je l’ai dansé : Pan-pan — Pan-pan Polyphonie — polyfolie Pan-pan Ma mère est une sainte ! Pan-pan Mon père est un café-chantant Pan-pan — Pan-pan Pan-pan-pan ! Zéro est de la petite vérole, Brûler un O dans le drapeau : Souffle le magnétisme des chiffres, consommation en cercles concentriques Rase cubique de la table ronde Pan-pan ! Fin-fin Pan-pan Finale ! Pan-pan O i u a 89 Pan-pan Da capo Bê Pan-pan — Pan-pan Pan-pan FIN ! Une analyse des textes de Clément Pansaers correspond à l'image générale donnée ci-dessus comme cadre pour le mouvement dadaïste. Ils suivent en général une structure éclatée, qui se retrouve dans le schéma suivant (2) : Schéma 2 : exemple d'analyse de poème dada (Panpan de Clément Pansaers) Le jeu sur la langue prend des formes variées selon les auteurs. Dans un texte intitulé en wallon M’nèfan d’chez nous, deux grands humoristes, les frères jumeaux Marcel et Gabriel Piqueray, proposent une liste de contre- vérités inscrites à la fois dans l’écriture, dans l’épitexte et dans le discours narratif : une épigramme posant la question de savoir si Napoléon n’était pas en réalité wallon, la présence d’un premier narrateur qui est un faux éditeur et d’un deuxième narrateur anglais qui aurait trouvé un poème en wallon de Gabriel Piqueray écrit en … 186312. 12 Cet exemple est compris dans une réédition de 1993 chez Labor. Le texte commence par l’épigramme : Bonaparte était-il wallon ? (pp. 37-42). 90 Le cas du wallonisme est très particulier et sert bien sûr à la fois à l’autodérision et à un regard amusé sur le « bon » français. Les irréguliers du langage s’en sont régalés. En-dehors des frères Piqueray et d’André Blavier (qui a recherché et classé des « fous » de la littérature universelle) mentionnés plus haut, ajoutons les grands auteurs dramatiques dont les pièces sont souvent truffées d’inventions lexicales, en particulier Fernand Crommelynck et Michel de Ghelderode. On trouve dans Mademoiselle Jaïre des références shakespeariennes adaptées en dialecte bruxellois13. Que le pseudonyme flamand Ghelderode (jaune et rouge) ne déconcerte pas le spectateur : il s’agit bien de francophonie, mais d’un sabir particulier à son auteur qui mêle un discours médiéval à une langue familière du XXe siècle dans un contexte imaginaire le plus souvent inattendu et d’apparence désordonnée, une sorte d’espace mental qui a pris la forme de tableaux vivants. Au-delà du wallon, le texte défiant la norme s'en prend également, comme on l'a noté dans le schéma 2, à la progression de l'apprentissage linguistique. En effet, l’onomatopée, si utilisée dans le texte dada, se retrouve en poésie moderne par exemple chez Jacques Sojcher, dans un texte en brefs vers libres intitulé « Apapa aphasie ». La lecture d'un tel texte sans connaître la biographie de l'auteur aboutit à un non-sens14. Ici, la matrice fondée sur une quête apparente de la langue zéro sert parfois d'exorcisme et de repli fœtal. Dérision et désespérance vont alors d'amble. La langue zéro devient une langue autre, porteuse de message intime et ouverte à l'analyse, voire l'auto-analyse. L'idée n'est plus tant chez le lecteur de tenter de comprendre un texte que de tenter de comprendre l'objet d'une communication. L'architecture d'ensemble devient prétexte à un discours ou lancinant ou exacerbé qui redessine le concept de dérision ontologique exprimé en introduction. 13 Dans la scène XVII de l’acte 1 apparaissent les trois Mariekes, parodiant les sorcières de Macbeth à partir d’une langue mi-bruxelloise, mi-inventée. 14 Un autre bref poème de « La mise en quarantaine » (qui inclut « apapa aphasie ») est tout à fait explicite : Déparler/ défaire la langue déterrer/l’enfance le père bégayer/le manque de premier geste balbutier l’apaparole fr tout/récit (Sojcher, Le rêve de ne pas parler, Labor, p.20). 91 4. Une lecture anarchisante de la bande dessinée L'exemple pris dans la bande dessinée fourni plus haut concernant l'histoire de Belgique, celui de Suske en Wiske, possédait déjà une marque d'individualisme anarchique dans son personnage le plus truculent, Monsieur Lambique (du nom d'une bière typique bruxelloise, la gueuse). Tout l'aspect ludique de cet épicurien bien ordinaire est constitué de gags et de jeux de mots à l'origine flamands. D'autres personnages qui ont gagné une renommée internationale appartiennent directement à l'univers qui nous intéresse ici : les créations d'Hergé, Quick et Flupke et surtout Tintin, et de Peyo, les célèbres Schtroumpfs. En janvier 1930, Hergé, responsable du supplément pour la jeunesse du journal Le XXe siècle, crée deux gamins bruxellois qu'il nomme Quick et Flupke et qu'il anime parallèlement à sa célèbre série Tintin qui paraît dans le même hebdomadaire. Cette série sera ensuite rééditée sous forme de gags en 11 albums, qui seront refondus pour une publication internationale. Celle-ci atteindra la Finlande à la fin des années 80 (6 albums traduits chez WSOY de 1986 à 1989 et des gags dans Ilta Sanomat en 1987). C'est dans cette série qu'Hergé sera le plus proche des jeux sur l'écriture en éliminant la calligraphie à l'intérieur de phylactères traditionnelle à la bande dessinée. Des mots échapperont occasionnellement à leur cadre et cela permettra d’introduire les textes manuscrits maladroits des enfants dans des cahiers ou des pancartes, voire des grafitis sur les murs, et parvenir à une dimension onirique quand il s'agira de supports invraisemblables comme les messages inscrits dans le ciel ou par l'intermédiaire d'une cohorte d'hommes-sandwichs exhibant chacun une lettre de l'alphabet.15 Les échappées purement graphiques qu'Hergé s'est permis de concrétiser en marge des phylactères montrent bien qu'il y a deux types d'écarts chez les irréguliers du langage belges, l'un basé sur la typographie traditionnelle, celle qui va de De 15 L'absence de texte fait partie de ce jeu : quand l'artiste peintre Flupke invite le tout-Bruxelles à son exposition « La lumière de la ville », le public en extase observe et commente des tableaux de dimensions diverses parfaitement noirs autour de leur auguste auteur (gag paru dans Ilta Sanomat en finnois le 17 avril 1987). 92 Coster à Jean-Pierre Verheggen, et un autre qui déforme l'unité graphique au profit de dérapages, ceux de l'écriture (pré)scolaire manuscrite de l'apprenti et celui des logogrammes qui concluront l'article. Les effets discursifs de l'univers de Tintin ont une autre origine, celle d'un idiolecte lié à un personnage au caractère fortement typé. Ces formes orales basées le plus souvent sur la méconnaissance sémantique exhibée ou sur divers troubles dans la communication orale renvoient à la prise de conscience d'une langue fautive. Si le capitaine Haddock emploie des termes pseudo-scientifiques comme jurons, si les Dupondt ne parviennent pas à conclure à deux des phrases relativement simples, si le professeur Tournesol détourne les informations reçues du fait de sa surdité, il résulte de la lecture des aventures de Tintin une autre dimension aventureuse, celle de la communication en français. Que ces défauts ne posent en général guère de problèmes insurmontables à la traduction littérale, sauf dans les cas d'adaptation de jeux de mots, ne peut occulter le fait qu'à l'origine la langue incriminée est bien le français. La présence d'étrangers francophones (Allan, Rastapopoulos, la Castafiore, Szut, le colonel Spontz, les généraux sud-américains par exemple), la clarté de la langue du héros Tintin et de sa contrepartie Milou jouant le rôle de contrepoint et le classicisme des histoires proposées permettent à la fois de lier ces irrégularités au récit et de les y relativiser. Les aventures de Tintin reprennent le flambeau laissé par Quick et Flupke dans les années 30 et en prolongent le pervertissement linguistique à l'égard de l'institution scolaire, voire même de la communication francophone. Le succès est tel que les enfants des années 40 et 50 surtout, au moment de la constitution de la grande famille du château de Moulinsart, ont appris par cœur ces déformations sans avoir le plus souvent la moindre notion de la signification première des bachi-bouzouks et autres pithécanthropes, termes rendus célèbres par la magie et l'impact de la bande dessinée. De telles irrégularités se limitaient pour la plupart à des cas personnels liés à des créations à forte personnalité et Tintin servait à la fois de point de comparaison et de garde-fou à ces écarts. La société n'en était pas pour autant globalement incriminée. Pour parvenir à ce stade, il fallut attendre la création de 93 toute une tribu de lutins bleus à bonnets blancs sous forme d'un mini livre paru dans un journal concurrent au magazine Tintin, le journal de Spirou. Les schtroumpfs, puisque tel est le nom de cette famille, ne parlent qu'en partie le français. Chaque terme de catégorie ouverte, substantif, adjectif qualificatif, verbe, se base sur un seul substitut, « schtroumpf », ce qui pourrait donner l'occasion de quiproquos, mais qui, par la magie de l'implicite et de l'intertextualité, aboutit à une langue tout fait compréhensible en dépit de ses blancs apparents. Il faut ajouter que, selon les schtroumpfs, la prononciation de ce seul vocable permet de différencier en réalité tous les signifiés. Que penser de cette déviation extrême de la langue française ? La première conséquence est sans doute qu'elle conforte l'idée d'une structure de la langue dans la mesure où le prédicat suit les règles traditionnelles de la morphologie avec le suffixe correspondant au temps du verbe : Qui schtroumpfe schtroumpfe ; Qui a schtroumpfé schtroumpfera et Qui schtroumpfera schtroumpfera devrait ainsi par la magie de la connaissance encyclopédique des proverbes aboutir en principe à des formes uniques Qui dort dîne ; Qui a bu boira et Qui vivra verra. Il est clair que de tels cas ne sont pas forcément légion. La deuxième conséquence, toujours en rapport avec l'existence d'une matrice sous- tendant la langue, est que ce langage conforte l’idée de la théorie distributionnelle selon laquelle les termes d'une phrase ont des liens internes qui expliquent leur position dans l'enchaînement linguistique : un mot remplacé par le substitut universel (à quelques variantes près) schtroumpf retrouve sa sémantique originelle dans la phrase et dans la chaîne morphosyntaxique. Il suffit alors d'employer à bon escient des lexèmes ayant pour origine le français classique : J'ai entendu schtroumpfer de lui pourrait se limiter à l'occurrence parler dans son interprétation. Cela demande à la fois une compétence linguistique suffisante qui ne nécessite d'ailleurs pas le recours à la « traduction » en français du terme rejeté dans le vocable schtroumpf. Il y a d'autre part une dimension clairement psycholinguistique à cette langue destinée en principe aux enfants, et en 94 particulier le fait que la langue schtroumpf autorise la prononciation orale de termes interdits dans la vie courante. Comme pour Tintin, l'écart linguistique dépasse les frontières de la langue française puisque la traduction ne posera pas d'énormes problèmes : il y aura peu d'ambiguïté en finnois dans une phrase normative comme Poika schtr[o]umpfi kiltisti kotiin ou bien le proverbe Se parhaiten schtrumpf joka schtumpfeksi schtrumpf.16 Cette dimension psycholinguistique ne suffira pas dans tous les cas. Comme il s'agit d'une communauté autonome, comportant ses éléments de cohésion apportés par les dangers causés par la nature (la recherche de nourriture) ou des personnages étrangers (un magicien et son chat), il entrera tout naturellement des problèmes sociaux internes au village où habitent les schtroumpfs. L'invitation au jeu linguistique ainsi fournie par les personnages se double d'une moralité sociale qui elle même se fonde en partie sur la langue, comme le montre le titre d'un album clairement centré sur la fameuse querelle linguistique qui divise les Belges depuis des générations, Schtroumpf vert et vert schtroumpf. Le titre lui-même est significatif puisqu’il renvoie à l’expression « Bonnet blanc et blanc bonnet », un couvre-chef universel chez les schtroumpfs. Doit-on dire « tire-schtroumpf » ou bien « schtroumpfe-bouchon » ? Une telle question est à la base de violentes dissensions parmi des êtres qui se sont révélé être jusqu’ici le plus souvent paisibles et inoffensifs. Les écarts concernant la francophonie débordent de leur sujet, puisqu'il s'agit d'une langue bâtarde à mi-chemin entre le « véritable » français parlé (ou écrit, puisque l'univers des schtroumpfs est farci d'allusions à la littérature) et un apport local caricatural qui se distingue par un seul son proche d'une onomatopée primaire. Au-delà d'une moralité que l'on peut considérer comme dérisoire ou enfantine dans une bande dessinée destinée en principe aux très jeunes, il y a bien une sonnette d'alarme non seulement politique mais aussi linguistique. Qu'il s'agisse clairement d'un défaut de communication entre deux communautés, sans doute entre Flamands et Wallons, ne fait pas de doute. Mais il 16 A noter toutefois l’influence pernicieuse du dessin animé américain qui a transformé le terme strumpf utilisé dans les premières traductions en finnois en smurf. 95 peut s'agir tout aussi bien d'une autodérision sur le bien parler dans la seule langue française et un autotest de compétence linguistique valable tout autant pour les enfants que pour les moins jeunes. Ce test démontre a posteriori combien la compréhension linguistique, en grande partie inconsciente, est dominée par les automatismes. 5. Au-delà du signe linguistique : calligramme et logogramme Les textes de fiction, les poésies n'ont pas l'apanage du droit à l'écriture. Cette écriture se trouve mêlée au graphisme comme dans le cas déjà analysé de la bande dessiné. Elle sera également mêlée à la peinture afin de jeter à nouveau le trouble sémantique dans les esprits. Il suffit de contempler des tableaux de Félicien Rops comme Le sacrifice, La Foire aux amours ou La Dame aux cochons pour se faire une première idée du besoin de provocation de l'artiste belge. Le jeu sur l’anticonformisme bourgeois se nourrit de satanisme, de masques, de nudité féminine et de présence concrète de la mort. Des oxymores graphiques allient des éléments pervers comme dans La Femme au cochon et un portait de femme nue mouchant un enfant sous l'œil intéressé d'un clown voyeur. Le cas de James Ensor et le scandale de l’entrée du Christ à Bruxelles (1888) a provoqué une levée des boucliers de la chrétienté en son temps. La censure n'a pu empêcher non plus la divulgation d'écrits mettant en cause des tabous sexuels au moment des procès intentés à Camille Lemonnier et Georges Eekhoud. Au temps du surréalisme belge, René Magritte va s'attaquer à nouveau à un autre tabou, celui de la vérité sémantique mise à mal tout au long du XXe siècle par les irréguliers du langage analysés ici. Chacun connaît la fameuse pipe qui n'en est pas une tout en étant clairement l'image d'une pipe. Le texte explicatif était pour le moins déroutant. La négation était alors tout le problème : un terme recopié recouvre normalement le concept générique. Magritte est allé plus loin dans cette déconstruction du signifiant et du signifié en peignant des séries d’objets et en leur attribuant des vocables normalement réservés à tout autre chose et en créant des métaphores mi- calligraphiques mi-graphiques : l’image d’un verre avec la définition écrite « L’orange », celle de la chaussure appelée « lune » ou bien celle du chapeau 96 intitulée « la neige ». Il reconstruit l’univers selon des architectures inattendues, en introduisant un nu féminin dans le texte manuscrit Je ne vois pas la [peinture de femme] cachée dans la forêt ; en assemblant des blocs de pierre à la manière des titres colossaux destinés dans les affiches et les génériques de péplums à attirer le public, et en composant les quatre lettres du mot RÊVE comme soubassement ; enfin en complétant un quadrilatère irrégulier à quatre cases avec les définitions ciel, corps humain (ou forêt), rideau et façade de maison. La recherche de l’élément perturbateur et les autoréférences par la répétition thématique aboutissent à un univers unique qui rejoint clairement l’obsession artistique belge de maltraiter la langue française. L’association entre poésie et image graphique a fourni certaines compositions mettant en cause l’ordre traditionnel de lecture, et en particulier les calligrammes. Les illustrations de poèmes sont parfois directement associées à l’écriture et la complètent, par exemple dans le cas de poèmes de Claire Lejeune basées sur le concept féminin de la matrice17. Le cas le plus extrême de la fusion entre le concept écrit et le concept graphique se retrouve chez plusieurs auteurs nés en Belgique, dont les plus renommés sont sans doute Henri Michaux et Christian Dotremont. Le plus célèbre des poèmes d’Henri Michaux, « Le grand combat » se base sur l’onomatopée et des néologismes suffisamment éloquents. Comme dans la langue schtroumpf, la morphosyntaxe y est en général respectée. Michaux est allé beaucoup plus loin dans son aventure concernant l’écriture. S’il n’a pas composé de longs romans à la manière du Français Georges Perec qui y introduisait des artifices (le lipogramme par exemple), il a tenté de lire des idéogrammes sans en connaître la signification première et sous forme de pensées séparées. Dans Affrontements, par exemple, il décrit ainsi une approche cognitive de l’interprétation face à une page de signes : 17 Dans Mémoire de rien (Editions Le Cormier, 1972), Claire Lejeune insère des peintures en noir et blanc introduites par le texte : L’écriture, c’est la mort de l’image. 97 « Idéogrammes sans évocation Caractères variés à n’en pas finir. La page qui les contient : un vide lacéré. Lacéré de multiples vies indéfinis. Il y eut pourtant une époque, où les signes étaient encore parlants, ou presque, allusifs déjà, montrant plutôt que choses, corps ou matières, montrant des groupes, des ensembles, exposant des situations. Il y eut une époque. Il y en eut d’autres. Sans chercher à simplifier, ni abréger, chacune à la tâche de dérouter pour son propre compte, se mit, brouillant les pistes, à manipuler les caractères de façon à les éloigner encore d’une nouvelle manière de la lisibilité primitive. »18 Chez Christian Dotremont, le point de départ n’est pas tant l’effort de compréhension que la fusion d’une image composée de signes noirs en tous sens et d’un titre qui est le point de départ de son illustration. Le matériau en est le graphème, mais un graphème métamorphosé par l’artiste, qui décompose les mots pour en recomposer une image globale sous forme de tableau. Il s’agit des « logogrammes ». Une telle invention n’a pas eu seulement pour support le papier. Christian Dotremont, fondateur du groupe COBRA (nom formé à partir des noms de ville COpenhague-BRuxelles-Amsterdam19), était un amoureux de la Finlande et en particulier d’Ivalo et de la Laponie. Il a élargi ses créations à des logoneiges qu’il dessinait à la main dans la nature avant de les photographier en raison de leur caractère éphémère. L’objectif est une nouvelle fois d’écarter la tradition sémantique au profit d’une interprétation personnelle transmise par le dessin. C’est en même temps une décomposition artificielle des mots qui remet en cause leur dimension cohésive. Dans quelle mesure en effet la double articulation du langage ou même les principes de la base de la phonologie sont-ils un simple effet d'équilibre effectué par l'effort cognitif ? Les logogrammes de Dotremont sont en premier lieu des ensembles qui se distinguent par une impression générale et la présence d'une énigme. Le simple fait de savoir qu'un tableau de Dotremont est 18 Affrontements, page 77. Comme Perec a joué avec les couleurs pour la grande aventure de la lettre e (La disparition), Michaux oppose les idéogrammes en rouge au texte français en noir. 19 Cette conjonction entre les trois villes du nord de l'Europe élimine clairement la zone d'influence de Paris, comme l'exprime par exemple Emmanuel Rubio (2005:201). 98 une entité composée des lettres de l'alphabet et que ces lettres appartiennent à un ensemble pouvant être lu selon les formes enseignées de gauche à droite et de haut en bas conduit à la sécurité apportée par la logique cohésive inscrite dans le titre. Il s'agira ensuite pour le spectateur/lecteur du logogramme de s'orienter sur un triple parcours pour répondre aux questions : quel lien rattache le titre au tableau ? Quel lien y a-t-il avec l'auteur et à quelle interprétation a-t-il tenté de nous familiariser ? Enfin, quel est mon propre rapport avec l'image et quelles interprétations suggère-t-elle ? Le logogramme (comme d’ailleurs tous les exemples fournis dans l’article) implique une herméneutique qui met en jeu de nombreuses frontières communicatives scripturales et graphiques. L’idée du logogramme et son exécution appartiennent à la logique du siècle : une symbiose entre l’écriture et le dessin a comme conséquence un effort demandé à l’interprétant qui aboutit à une remise en cause l’acte traditionnel de lecture. Le stylo a laissé sa place au pinceau chez Michaux et Dotremont, une vision globale de la phrase ou du signe précède l’interprétation, un retour à une métaphysique « primitive », éloignée de la logique occidentale chrétienne, offre une suite aux expériences dadaïstes et surréalistes tout en s’inscrivant dans une logique de l’irrespect. 6. Conclusion La destruction consciente ou inconsciente de la langue ne peut être le fait du hasard. Depuis les débuts de la littérature francophone de Belgique, c'est-à-dire les Aventures de Till Eulenspiegel, elle joue un rôle central. Qu'il s'agisse de la chanson, du théâtre, de la poésie, des histoires destinées aux enfants, de la bande dessinée, de la peinture ou de formes complexes comme la création d'idéogrammes et de logogrammes, la langue semble faire partie du subconscient de toute une nation qui de plus se pose la question de savoir dans quelle mesure la Belgique existe. Christian Dotremont et Henri Michaux par exemple ont offert des moyens d'évasion à cette question lancinante en éliminant d'une certaine mesure la 99 norme parisienne. Au-delà de ces écarts qui ne sont pas de simples excentricités mais l'expression d'un certain désarroi par rapport aux hasards d'une existence géopolitique sans cesse mise en question, l'aventure de la langue reste ouverte aujourd'hui avec des auteurs comme Jean-Pierre Verheggen. Les considérations pragmatiques sur la littérature et sur l’art mettent l’accent sur le jeu communicatif concernant la véracité implicite dans une représentation du monde. Une analyse sur l’énonciation des œuvres d’art aboutit pour sa part à la réalité chez l’artiste d’une manifestation consciente ou non de son attitude envers la langue (et donc le rôle de celle-ci) et l'acte de création (comprenant des marques exprimant par exemple des besoins, des confessions dissimulées ou des désirs d’exhibition). La motivation artistique qui pousse à l’acte d’écrire, de dessiner ou de peindre revient souvent à des généralités pratiques (renommée, fortune, concrétiser un rêve), psychologiques (les rapports troubles avec Eros et Thanatos), ou idéalistes (manifester le besoin d'un message inhérent à l'écriture, le besoin d'exister et de manifester cette existence à l'aide de manifestations artistiques). Le message belge sera : le français est une langue dont nous sommes conscients et que nous utilisons soit pour en extraire une méthode et un idéal, un exemple à suivre, soit en jouant avec lui pour montrer qu'il n'est pas limité à des formules académiques ou standard, car ce qui est standard est conservateur et tue la vivacité de la langue. Ce n'est plus une copie du français de France, mais une adaptation qui comprend le besoin de montrer son origine géographique. Il est vrai que les rénovateurs francophones de la langue sont nombreux : il suffit d’énumérer les membres de l'Oulipo (François Le Lyonnais, Raymond Queneau, Georges Perec)20 et du roman noir (Daniel Pennac), la présence des dialectes et des jeux linguistiques exhibant la différence chez les Canadiens, les Romands et les artistes et écrivains de Caraïbe et d'Afrique. Il s’agit d’un univers qui déborde clairement 20 L’Oulipo n’est bien sûr pas limité à la France, comme le démontre l’activité par exemple d’Italo Calvino. 100 le sujet de l’écriture et de l’image fixe, pour atteindre des arts périphériques. Ce sera le cas avec le monolinguisme qui florit en Belgique et le cinéma qui n’a pas dans ce pays la notoriété qu’il a atteint à une certaine époque en Suisse et surtout au Canada. Il résulte de ce message belge une littérature parallèle basée sur le jeu et l’absurde qui met en cause des structures et des écritures, un univers instable où le lexique sert à ouvrir des correspondances, un implicite enfoui dans des architectures et des structures, fréquemment ludiques et parfois tragiques, qui révèlent un cadre basculant vers des orientations anarchistes, nihilistes voire panthéistes au besoin. Une telle vision a une valeur politique certaine au-delà de son égocentrisme de base et son humour, bonhomme ou pervers, possède une force salutaire tout à fait revigorante qui perdure aujourd’hui en Belgique. Le message qui en découle est essentiel : la vérité passe par le déni de toute prétention, par la mise en cause du présupposé et de la norme enseignée. La dérision, qui en est l’outil, enclenche une déstabilisation qui atteint au-delà du cas belge une dimension ontologique universelle. Bibliographie Aron P. 1991. « Les groupes littéraires du surréalisme » in Textyles 8. Bruxelles. Aron P. et Soucy P.-Y. 1998. Les revues littéraires belges de langue française de 1830 à nos jours. Editions Labor. Bruxelles. Biron M. 1991. « Le refus de l’œuvre chez Paul Nougé » in Textyles 8. Bruxelles. Blavier A. 1982. Les fous littéraires. Veyrier. Blavier A. 1997. Le don d’ubuquité. Devillez. Dantine A. 1991. « Achille Chavée, le trafiquant de l’invisible » in Textyles 8. Bruxelles. Dotremont C. 1974. Logbook. Yves Rivière. Paris. Dotremont C. 1998. Cobraland . La Petite Pierre. Bruxelles. Ghelderode M. de 1950. Théâtre 1. Gallimard. Paris. Gobbers W. 1991. « Résonnances futuristes » in Les avant-gardes littéraires en Belgique (173-204). Editions Labor. Bruxelles. Halen, P. 1993. Le petit Belge avait vu grand. Labor. Bruxelles. Hergé. 1987. Quick et Flupke : Pardon Madame. Casterman. Tournai. 101 Klinkenberg J.M. 1991. « Le phénomène Jeune Belgique : un accident historique ? (153-167) in Les avant-gardes littéraires en Belgique (173-204). Editions Labor. Bruxelles. Klinkenberg J.M. 1983. Lecture in Charles de Coster. La légende d’Ulenspiegel/livre 1. Editions Labor. Bruxelles. Lisse M. 1991. « Franchir le mur de la langue » in Textyles 8. Bruxelles. Lits M. 1993. « Des fantastiqueurs belges ? » in Textyles 10. Bruxelles. Mabille X. 1986. Histoire politique de la Belgique. Facteurs et acteurs de changement. CRISP. Bruxelles. Michaux G. et Piret P. (dir.) 2000. Logique et écritures de la négation. Editions Kimé. Paris. Michaux H. 1944. L’espace du dedans (pages choisies). Gallimard. Paris. Michaux H. 1986. Affrontements. Gallimard. Paris. Morelli A. (dir.) 1995. Les grands mythes de l’Histoire de Belgique de Flandre et de Wallonie. Vie ouvrière. Bruxelles. Nougé P. 1983. Fragments. Espace nord 7. Editions Labor. Bruxelles. Pansaers C. 1986. Bar Nicador et autres textes dada. Editions Gérard Lebovici. Paris. Quaghebeur M. 1998. Balises pour l’histoire des lettres belges. Editions Labor. Bruxelles. Rosi R. 1991. « L’éternel retour du Pan-pan au cul » in Textyles 8. Bruxelles. Rubio E. 2005. « Christian Dotremont entre surréalismes belge et français » in Les surréalistes belges. Revue Europe no 912. Paris. Sojcher J. 1990. Le rêve de ne pas parler. Espace nord 59. Editions Labor. Bruxelles. Soumois F. 1987. Dossier Tintin. Editions Jacques Antoine. Bruxelles. Textyles. 1991. Surréalismes de Belgique. Numéro 8. Textyles-éditions. Bruxelles. [anonyme] 1980. Phantomas ultime. Argon. Paris. Sites Internet : http://suskeenwiske.ophetwww.net (3.11.2010) http://www.arquebusiers.be (3.11.2010) 102 Okay als Rückmeldesignal in deutscher und schwedischer Chat- Kommunikation Helena Nilsson (geb. Pettersson), Universität Göteborg 1. Einleitung Der Web-Chat bietet die Möglichkeit, im Internet mit anderen Personen in Realzeit schriftlich zu kommunizieren. Die Konversationen im Chat-Raum werden oft als geschriebene Gespräche bezeichnet, aber auch wenn Konversationen im Chat-Raum viele Übereinstimmungen mit dem Face-to-Face- Gespräch haben, weisen sie auch andere Eigenschaften auf, die für das mündliche Gespräch eher untypisch sind. Erstens ist die Chat-Kommunikation schriftlich statt mündlich, zweitens ist die Chat-Kommunikation im Unterschied zur gesprochenen Sprache nicht völlig synchron. Der Teilnehmer kann den Äußerungen der anderen Teilnehmer bei ihrer Produktion nicht folgen, die Mitteilung wird erst nach dem Abschicken an den Chat-Server für die anderen Teilnehmer im Chat-Raum sichtbar. Diese Asynchronität führt auch dazu, dass zwei Teilnehmer gleichzeitig eine Mitteilung abschicken können – die Sequenzierung wird dann technisch vom Chat-Server organisiert. Indem das Turn- taking nicht ausgehandelt werden muss, wie bei der gesprochenen Sprache, ist das Sprecherwechselsystem außer Kraft gesetzt. (Storrer 2001: 7). In diesem Beitrag werde ich die Funktion der Wortform okay in der deutschen und schwedischen Chat-Kommunikation näher untersuchen. Der Schwerpunkt liegt darauf, wie okay als Rückmeldesignal in der Organisation des Chat- Gesprächs eingesetzt wird. Bei mehreren exemplarischen Studien habe ich feststellen können, dass okay besonders im schwedischen Chat eine wichtige Rolle als Rückmeldesignal zu spielen scheint, indem mit okay die Wahrnehmung des Vorgängerbeitrags gezeigt wird. Im deutschen Chat scheint okay als deutlicher Ausdruck für Zustimmung verwendet zu werden. 103 Beim gesprochenen Gespräch kann im Unterschied zum Chat die Rückmeldung des Hörers parallel mit der Produktion des Sprechers gezeigt werden. Man spricht dann von Hörersignalen, mit denen der Hörer dem Sprecher seine Aufmerksamkeit versichert. Die Äußerungen des Hörers richten sich nicht auf die Übernahme der Sprecherrolle. In sprachlicher Hinsicht handelt es sich bei den Hörersignalen vor allem um Partikeln und Kurzäußerungen (Brinker/Sager 2001: 59). In der Chat-Kommunikation können die Hörersignale nicht von anderen Beiträgen unterschieden werden. Nachdem sämtliche Beiträge in der Chat-Kommunikation im selben Kanal produziert werden, da es keinen Hörerkanal gibt, sind alle Beiträge als turnwertig zu betrachten. Die Beiträge können daher nicht in Sprecher- oder Hörerbeiträge eingeteilt werden, sondern sämtliche Beiträge müssen als Turns angesehen werden. Die Rückmeldung muss demzufolge in der Chat-Kommunikation aufgrund technischer Voraussetzungen anders erfolgen als beim mündlichen Gespräch. Da die Beiträge erst nach der Fertigstellung für die anderen Teilnehmer im Chat- Raum präsentiert werden, gibt es keine Möglichkeit, den Sprecher parallel bei der Produktion zu unterstützen. In einer Untersuchung zur Gesprächsorganisation in der Chat-Kommunikation hebt Schönfeldt (2001: 42) die Notwendigkeit hervor, Aufmerksamkeit und Verständnis zu signalisieren, wenn der komplette Beitrag auf dem Bildschirm erscheint. Schönfeldt verwendet in diesem Zusammenhang die Bezeichnung „Rückmeldung im weiteren Sinne“ (Schönfeldt 2001: 42). Hierunter werden jene sprachlichen Mittel verstanden, die vom Empfänger eingesetzt werden, um dem Sender deutlich zu machen, dass ihn sein Beitrag erreicht hat. Rückmeldung kann auch mit Hilfe von so genannten Follow-up-Signalen ausgedrückt werden. Unter Follow-up wird der Kommentar nach einer Frage- Antwort-Sequenz verstanden, diese Erscheinungen wurden von Sinclair und Coulthard (1975) im Zusammenhang mit Diskussionen im Klassenzimmer wahrgenommen. Im Klassenzimmer besteht die Funktion von Follow-ups darin, dass bestätigt werden muss, ob der Schüler eine korrekte Antwort gegeben hat oder nicht (Sinclair/Coulthard 1975: 48). Außerhalb des Klassenzimmers hat der 104 Follow-up nicht diese Aufgabe, sondern es handelt sich dann um eine allgemeine Bestätigung, dass die Antwort wahrgenommen wurde. (Stenström 1983: 322). Stenström betrachtet Hörersignale (sie verwendet jedoch den Begriff Supports) und Follow-up-Signale als Varianten des gleichen Phänomens. Beide indizieren, dass die Information empfangen wurde. (Stenström 1983: 319 f.). In dieser Untersuchung wird die Definition von Rückmeldung von Schönfeldt (2001) und Stenström (1983) übernommen. Unter Rückmeldungssignal wird demzufolge ein Signal verstanden, das vom Empfänger eingesetzt wird, um den Empfang eines Beitrags zu bestätigen. Hier wird zusätzlich die Abgrenzung vorgenommen, nur Signale an erster Stelle im Beitrag als Rückmeldesignale zu betrachten. 1.1 Angaben zu okay in der Literatur Okay ist sowohl im Deutschen als auch im Schwedischen ein relativ neues Lexem, das aus dem Englischen entlehnt worden ist. Ins Deutsche wurde okay im 19. Jh. entlehnt, wobei die Herkunft aber ungeklärt ist (Etymologisches Wörterbuch des Deutschen 1997). Im Schwedischen ist okej seit 1938 belegt (Nationalencykolopedins ordbok 1995). 1.1.1 Funktionen von okay Bevor okay als Rückmeldesignal in der Chat-Kommunikation untersucht wird, soll eine Literaturstudie vorgenommen werden, um einen Einblick darin zu gewinnen, wie okay in den Grammatiken beschrieben wird. Hierfür werden die Grammatik der deutschen Sprache (GdS) (Zifonun et al 1997) und die Svenska Akademiens grammatik (SAG) (Teleman et al 1999) verwendet. In der GdS wird okay in drei Funktionen als Äußerungsvariante von ja beschrieben. Erstens wird okay als Äußerungsvariante vom Responsiv ja im turninternen Gebrauch mit der Funktion ‚positiver Bescheid’ erwähnt. Okay kann als Variante bei einer „Übernahme einer durch die Vorgängeräußerung (z.B. Bitte um Zusage, 105 Aufforderung zu einem Versprechen) angetragenen Handlungsverpflichtung“ (Zifonun et al 1997: 373) eingesetzt werden. Zweitens wird okay im turninternen Gebrauch in der Funktion ‚Kontinuitätssicherung’ genannt. Der Sprecher verwendet ja, um die Kontinuität seines Beitrags und die Aufrechthaltung seines Rederechts zu sichern (Zifonun et al 1997: 375). (a) ...und die spielen halt hauptsächlich so Sachen nach so ??.. ja ...Bruce Springsteeen und ich weiß... (Willkop, Gliederungspartikeln, 102 (Cafékollektiv)) Hier gliedert der Sprecher die eigene Äußerung mit Hilfe von okay. Eine andere Bezeichnung dieser Funktion wäre deshalb ‚Gliederungspartikel’. Unter Kontinuitätssicherung nimmt die GdS eine dritte Funktion auf, nach der ja äußerungseinleitend bei der Übernahme der Sprecherrolle, die es zu behaupten gilt, verwendet wird. Sie heben hervor, dass vor allem die Variante tja in dieser Funktion vorkommt. Tja findet sich dann an „Diskursstellen mit thematischem Abschluss des Vorgängerthemas oder handlungsstrukurellem Übergang; dies gilt auch für die funktionelle Variante okay“ (Zifonun et al 1997: 376). In der GdS wird leider kein Beispiel für tja oder okay gegeben. Viertens wird in der GdS okay in der Funktion „reorientierende Anforderung einer konvergenzherstellenden Hörerreaktion“ beschrieben. (Zifonun et al 1997: 374). Hier liegt die Funktion ‚Rückversicherungssignal’ vor. In der schwedischen Akademiegrammatik (SAG) (Teleman et al 1999) wird okej in vier Funktionen aufgenommen. Okej wird zum einen als bestätigendes Antwortsignal nach einer Frage beschrieben (Teleman et al 1999: 752). Als weitere Funktion nennt SAG, dass der Sprecher nach einer Aufforderung mit okej seine Absicht angibt, der Aufforderung zu folgen (Teleman et al 1999: 753). Bei diesen Funktionen handelt es sich um eine Entsprechung der Funktion ‚Responsiv’ in der GdS. Die dritte Verwendungsweise von okej, die in der SAG aufgenommen wird, ist die Angabe des Sprechers, dass er die Sprachhandlung wahrgenommen hat (Teleman et al 1999: 755). 106 Die vierte Funktion von okej ist, wenn der Sprecher angibt, dass er ein neues Thema oder Teilfrage einleiten möchte (Teleman et al 1999: 756). (b) – okej, vad ska vi säga om ventilationen då – okay, was sagen wir denn zur Ventilation Gemeinsam für die beiden Darstellungen ist, dass dem Lexem okay in den Grammatiken keine eigene Darstellung gewidmet wird. Okay wird stattdessen gemeinsam mit anderen Lexemen als eine alternative Realisierung genannt. Ein zusammenhängendes Bild von den verschiedenen Verwendungsweisen von okay wird deshalb nicht gegeben. In Abbildung 1 werden die Funktionen von okay in GdS und SAG noch mal dargelegt. Abb. 1: Beschreibung von okay in der GdS und SAG GdS (Zifonun et al 1997) SAG (Teleman et al 1999) Responsiv/Antwortsignal x x Themenübergang x x Wahrnehmung der Sprachhandlung x Kontinuitätssicherung/Gliederungspartikel x Rückversicherungssignal x Beim Vergleich der Beschreibungen in der deutschen und in der schwedischen Grammatik kann festgestellt werden, dass okay in beiden Grammatiken als Antwortsignal nach Frage oder Aufforderung wahrgenommen wird. Sowohl in der deutschen als auch in der schwedischen Darstellung wird festgestellt, dass okay in der Einleitung einer Äußerung bei einem Themenübergang vorzufinden ist. In der GdS steht jedoch „Abschluss eines Themas“; in der SAG dagegen „Einleitung eines Themas“. Die Funktion ‚Wahrnehmung’ wird nur in der SAG beschrieben, hier konnte keine Entsprechung in der GdS gefunden werden. Eine turninterne Funktion von okay als Gliederungspartikel wird dagegen nur in der GdS genannt. In der SAG wird die Funktion ‚Rückversicherungssignal’ nicht behandelt. 107 1.1.2 Bedeutungen von okay In den Grammatiken wurde die Funktion von okay im Gespräch beschrieben, bisher wurde jedoch nichts zur Bedeutung von okay gesagt. Im Duden Universalwörterbuch (2003) wird okay als Adverb und Adjektiv beschrieben: Duden Universalwörterbuch: I. (ugs.) abgemacht, einverstanden, du gehst vor, o.?; (verblasst) o. (also), gehen wir II. a) ugs. in Ordnung, gut: es ist alles o. ; das Mädchen ist wirklich o. (verhält sich kameradschaftlich); gestern ging es mir reichlich mies, aber heute bin ich wieder o.; b) (Flugw. Jargon) [geprüft u. daher] bestätigt: Ihr Flug nach Kairo ist, geht o. Die Bedeutung als Adjektiv wird in dieser Untersuchung nicht beachtet, da okay in diesem Fall immer in den Satz integriert ist und keine Funktion in der Gesprächsorganisation einnimmt. Die Bedeutung von okay als Adverb wird in Duden mit abgemacht, einverstanden angegeben. Mit okay wird demzufolge eine Zustimmung ausgedrückt. Auch im schwedischen Wörterbuch Svensk ordbok (1988) wird okay in der Bedeutung ‚Zustimmung’ beschrieben: Svensk ordbok: okay el okej interj. uttr. för accepterande ~ vi gör som du säger [okay od. okej interj. Ausdruck für Akzeptanz ~ wir machen, wie du sagst] Im schwedischen Wörterbuch Svenska Akademiens ordbok (SAOB ) gibt es keinen Eintrag zu okay. Die Buchstabenfolge N-Okörd wurde 1949 publiziert. Vermutlich war okay um diese Zeit noch nicht Teil der schwedischen Sprache. 2. Okay in der Chat-Kommunikation Beim Durchsehen von Teilen des Materials habe ich den Eindruck gewonnen, dass die Bedeutung ‚Zustimmung’ bei okay im schwedischen Chat-Raum verblasst ist. Okay wird eher nur dafür verwendet, um den Empfang eines Chat- Beitrags zu bestätigen und funktioniert dadurch als Wahrnehmungssignal für den Chat-Partner. Die Funktion „Der Sprecher gibt an, dass er die Sprachhandlung wahrgenommen hat“ wird auch nur in der schwedischen Grammatik beschrieben. 108 Im deutschen Chat scheint okay öfter als Antwortsignal mit der Bedeutung ‚Zustimmung' vorzukommen. In dieser Untersuchung möchte ich meine ersten Eindrücke mit Hilfe einer empirischen Untersuchung überprüfen. Welche Funktionen können bei okay in der Chat-Kommunikation belegt werden? In welcher Hinsicht wird okay als Rückmeldesignal verwendet? Wie unterscheidet sich die Verwendung zwischen einem deutschen und einem schwedischen Chat-Raum? Bei den Funktionsbeschreibungen dienen die Kategorien in den Grammatiken GdS und SAG (Siehe Abschnitt 1.1) als Ausgangspunkt. Die Darstellungen in den Grammatiken betreffen aber die gesprochene Sprache; das hier zu verwendende Korpus besteht aus Chat-Kommunikation, eine schriftliche Kommunikationsform. Differenzen zwischen Grammatikdarstellung und Verwendung können deshalb aufgrund der Modalitätsunterschiede zwischen gesprochener Sprache und Chat vorliegen. Inwiefern die Kategorien auch bei der Chat-Kommunikation zutreffen, wird in der empirischen Untersuchung nachgegangen. Als Material für diese Studie liegen die Logfiles eines schwedischen und eines deutschen Chat-Raums vor. Die Chat-Konversationen wurden während eines Zeitraums von zwei Wochen aus dem Internet heruntergeladen. Die Chat- Räume sind offene Räume, so genannte Plauder-Chats, in welchen über allgemeine Themen gechattet werden kann. Der schwedische Chat wurde im Herbst 2004 heruntergeladen und enthält ca. 190 000 Chat-Beiträge. Der deutsche Chat-Raum wurde im Frühjahr 2005 geloggt und enthält ca. 153 000 Chat- Beiträge. Mit Chat-Beitrag werden die Teilnehmeräußerungen gemeint, die auf dem Bildschirm aufgrund eines vorangehenden und nachfolgenden Absatzreturns isoliert sind (Beißwenger 2003: 208). In einer Vorstudie wurden die Frequenzen von Gesprächspartikeln in den Logfiles zusammengestellt. Die Lexeme wurden mit Hilfe von Frequenzlisten sämtlicher Types in den Logfiles von Hand herausgesucht, als Gesprächspartikel wurden sämtliche Types gezählt, die in der Funktion ‚Gesprächspartikel’ vorkommen können. Die quantitativen Angaben in Abbildung 2 müssen deshalb mit einiger Vorsicht behandelt werden. Die hohe Anzahl von ja im schwedischen 109 Chat-Raum kann zum Teil dadurch erklärt werden, dass die abgekürzte Form des schwedischen ich (jag) auch inkludiert ist. Doch kommt ebenfalls in anderen Funktionen als der einer Gesprächspartikel vor. Charakteristisch für die Chat- Kommunikation ist die Variation der Schreibweise. Es wird oft von den Normen der Rechtschreibung abgewichen. In Abbildung 2 sind die Schreibvarianten der verschiedenen Lexeme zusammengelegt worden. In der Untersuchung wurden jedoch nur Types mit mindestens 20 Vorkommen beachtet. Okay steht im deutschen Logfile an fünfter Stelle und im schwedischen Logfile an dritter Stelle unter den Gesprächspartikeln. Abb. 2 Die häufigsten Gesprächspartikeln in den untersuchten Logfiles Deutsch Anzahl Schwedisch Anzahl ja, jaa, jaa, jaaa, jaaaa, jaaaaa 8842 ja, jaa, jaaa, jaaaa, jaaaaa 9386 doch 5144 nä, nää, näää, nej, ne, näe 5571 na 3701 ok, okej, oki, okey, okeej, åkej, okay, okäjjj 5235 nein, nö, nöö, nööö, ne, nee, neee 3330 jo, joo, jooo, joooo 2643 ok, okay, oki, okai 1720 mm, mmm, mmmm, mmmmm, mmmmmm 2228 ach 1084 hm, hmm, hmmm, hmmmm, hmmmmm 1610 hm, hmm, hmmm, hmmmm 874 japp 1152 jo, jou 744 oj 938 gell, gelle 674 jasså, jaså 738 naja 649 jaha 736 Tokens insgesamt im Korpus: 759 094 Tokens insgesamt im Korpus: 1 022 253 Da im Web-Chat mit den Regeln der Orthographie großzügig umgegangen wird, kommt okay in verschiedenen Varianten vor. Gerade bei okay ist die Schreibweise auch weniger normiert als bei anderen Lexemen. Für Deutsch werden in Duden Universalwörterbuch (2003) die Schreibweisen okay und O.K. angegeben. Im schwedischen Wörterbuch SAOL (Svenska Akademiens ordlista) (2006) werden 110 die Alternativen okej, OK, O.K. und okay angegeben. In Abb. 3 sind die häufigsten Varianten von okay in den Chat-Logfiles aufgeführt. Die Lexeme wurden mit Hilfe von Frequenzlisten sämtlicher Types im Text gesucht, auch hier wurden ausschließlich Types mit mindestens 20 Vorkommen aufgenommen. Abb. 3: Varianten von okay in den Logfiles Deutsch Anzahl Schwedisch Anzahl ok 1209 ok 2975 okay 275 oki 1185 oki 160 okej 805 okai 76 okey 135 åkej 49 okeej 41 okay 24 okäjjj 21 SUMME: 1720 SUMME: 5235 Wie aus der Übersicht zu entnehmen ist, ist die Formenvariation im Schwedischen größer als im Deutschen. Wir sehen auch, dass die Schreibvariante ok in beiden Logfiles die häufigste ist. In dieser Analyse wird nur die Variante ok beachtet1. 2.1 Methodisches Vorgehen Diese Studie ist eine Kombination von Korpussuche und qualitativer Analyse. Für die Analyse wurde vorab eine Stichprobe zusammengestellt, für die 100 Beiträge mit ok aus dem deutschen und schwedischen Chat mit einer Zufallsmethode ausgewählt wurden. Hierfür wurde das Konkordanzprogramm Wordsmith verwendet. Da mich vor allem die Funktion als Rückmeldesignal interessiert, wurden nur Vorkommen an erster Stelle im Beitrag in der Zufallsstichprobe beachtet. Die Vorkommen in der Stichprobe wurden nach der Funktion qualitativ ausgewertet. Um festzustellen, welche Funktion bei ok vorliegt, reicht ein Beitrag 1 Ab jetzt wird in diesem Artikel die Schreibweise ‚ok’ durchgehend verwendet. 111 allein nicht aus, sondern die Funktion muss mit Hilfe des Kontexts interpretiert werden. Das Konkordanzprogramm bietet die Möglichkeit, einfach von dem einzelnen Beitrag in der Stichprobe zu dem aktuellen Abschnitt im Logfile zu gelangen. In der Wiedergabe der Konversationen aus den Chat-Räumen wurden die Nicknames anonymisiert, indem nur der erste Buchstabe im Namen behalten wurde. In den Beispielen sind auch nur die Beiträge, die Teil der jeweiligen Konversation sind, vorzufinden. Beiträge von anderen, parallel laufenden, Chat- Konversationen wurden aus Platzgründen gestrichen. 2.2 Funktionen von ok in der Chat-Kommunikation Bei den Vorkommen von ok als Beitragseinleitung können folgende Funktionen belegt werden: 2.2.1 Ok als Antwortsignal In dieser Funktion stimmt eine Person mit ok dem Inhalt im Vorgängerbeitrag zu. Hier geht es um die Übernahme einer Handlungsverpflichtung im Vorgängerbeitrag. Diese Funktion wird sowohl in der deutschen Grammatik GdS (Zifonun 1997: 367) als auch in der schwedischen Grammatik SAG (Teleman 1999: 752 f.) beschrieben. Kennzeichnend für die Verwendung als Antwortsignal ist, dass die Zustimmung gegen eine Ablehnung oder Missbilligung ausgetauscht werden kann. Die Handlungsverpflichtung im Vorgängerbeitrag kann jedoch unterschiedlich aussehen: Ok kann als Antwortsignal nach einer Frage stehen, wie in Beispiel (1). (1) (01:51:32) M => d: Vill du ha adressen då? (01:52:14) d => M: ok kan väl kika in kanske o se vad de är för nåt (01:51:32) M => d: Willst du die Adresse haben? (01:52:14) d => M: ok kann vielleicht reingucken um zu sehen was das ist Auch in Beispiel (2) handelt es sich um die Antwort einer Frage. 112 (2) (23:07:04) J: [zu b] Gläschen Wein rüberschieb (23:07:51) b: [zu J] was soll das - nur ein gläschen? (23:08:07) J: [zu b] heee... ich will auch noch was abhaben gg (23:08:24) b: [zu J] teilen wir? (23:08:53) J: [zu b] ok..... Fläschchen rüberreiche Ok kann auch als Antwortsignal nach einer Aufforderung stehen. (3) (15:57:21) J: [zu E] oweia... hab ich Dich ehrlich überlesen? schäm (15:57:26) (15:57:33) E: [zu J] du warst mal so frei ggg (15:57:50) E: [zu J] lass das . gibt komische eckige gesichter gg (15:57:51) J: [zu E] kann ich das je wieder gut machen? gg (15:58:04) J: [zu E] ok gg (15:58:07) E: [zu J] großes leckeres eis zb ;-) In Beispiel (3) bezieht sich ok wahrscheinlich auf die Aufforderung von E „lass dass. gibt komische eckige gesichter“. Mit ok stimmt J der Aufforderung zu. In der deutschen Stichprobe wird ok 31 Mal nach einer Frage oder Aufforderung als Antwortsignal verwendet. In der schwedischen Stichprobe kommt ok nach einer Frage oder Aufforderung zweimal vor. 2.2.2 Ok als Wahrnehmungssignal In dieser Funktion wird mit ok deutlich gemacht, dass der Beitrag beim Empfänger angekommen ist und wahrgenommen wurde. Mit ok wird in dieser Funktion im Unterschied zur Verwendung als Antwortsignal keine Stellungsnahme zum Inhalt des Vorgängerbeitrags gemacht. Das Wahrnehmungssignal erfolgt nach einer Aussage, die keine Stellungsnahme vom Empfänger erfordert. (4) (23:31:51) m => G: ska på konsert nästa lördag (23:31:56) G => m: ok, kul (23:31:51) m => G: will nächste Woche ins Konzert gehen (23:31:56) G => m: ok, schön In der Vorgängeräußerung geht es nicht um eine Frage oder Aufforderung, sondern um eine Mitteilung. In der Fortsetzung des Beitrags von G sehen wir 113 außerdem, dass G mit ok nicht bezweckt, der Aussage zuzustimmen. In diesem Beispiel geht es auch um einen Sachverhalt, der außerhalb des virtuellen Chat- Raumes gültig ist. G kann diesen Sachverhalt aus diesem Grund auch nicht bestreiten. (5) (00:33:08) D: [zu M] dann sag ich ganz leise gute nacht...werde mich in die heia verkriechen...müüüde bin (00:33:41) M: [zu D] ok ... schlaf gut und träume schön .... liebe grüße In diesem Beispiel (5) verabschiedet sich D, bevor sie sich ausloggt. Mit einem ok zeigt M, dass er die Verabschiedung wahrgenommen hat. Auch in diesem Beispiel geht es im Vorgängerbeitrag um eine Mitteilung, die vom Empfänger nicht bestritten werden kann. Ok als Wahrnehmungssignal kommt häufig nach einer Frage-Antwort- Sequenz vor und funktioniert deshalb als Follow-up-Signal. Mit ok wird gezeigt, dass die Antwort im Vorgängerbeitrag wahrgenommen wurde. In Beispiel (6) stellt mb eine Frage an m35. m35 beantwortet die Frage. Mit ok zeigt mb, dass die Antwort wahrgenommen wurde. (6) (13:35:53) mb=> m35: vad gör du idag då ? (13:36:13) m35 => mb: tja, har tvättat o donat lite, e hemma för vab (13:36:30) mb => m35: ok (13:35:53) mb=> m35: was machst du denn heute? (13:36:13) m35 => mb: tja, habe gewaschen und verschiedenes gemacht, bin zu Hause wegen kranken Kindes (13:36:30) mb => m35: ok In Beispiel (7) fragt B, wie es M geht. M antwortet, dass ihr der Rücken weh tut. B zeigt mit einem o weh (oj), dass er bestürzt ist. Er fragt, was passiert ist. M antwortet, dass sie keine Ahnung hat, aber dass es bestimmt bald besser wird. B zeigt mit einem ok, dass er die Antwort wahrgenommen hat. (7) (01:26:17) B => M: Hur (01:26:37) B => M: är läget i natten? (01:27:03) M => B: ont i ryggen, annars helt ok (01:27:18) B => M: oj vad har hänt? (01:27:52) M => B: ingen aning men det blir säkert bra snart 114 (01:28:08) B => M: ok vad gör du annars då (01:26:17) B => M: Wie (01:26:37) B => M: geht es in der Nacht? (01:27:03) M => B: Rücken tut weh, sonst ganz ok (01:27:18) B => M: oh weh, was ist passiert? (01:27:52) M => B: keine Ahnung, aber es wird bestimmt bald besser (01:28:08) B => M: ok, was machst du sonst so? In Beispiel (8) wird die Frage nicht explizit gestellt. Durch die Aussage „kann man auch woanders hin“ wird aber deutlich, dass H vorschlägt, den Chat-Raum zu wechseln, um privat chatten zu können. Zu diesem Vorschlag stellt sich M allerdings negativ. Mit ok zeigt H, dass er die Antwort als solche akzeptiert. (8) (22:18:35) H: [zu M] verliere den überblick zu voll (22:19:16) M: [zu H] das ist nicht gut (22:19:47) H: [zu M] kann man auch woanders hin (22:19:49) H: sfg (22:20:19) M: [zu H] möchte doch schon gern hier bleiben schmunzelt (22:20:38) H: [zu M] ok Ein ok als Wahrnehmungssignal wird oft eingesetzt, um die Kommunikation in Gang zu halten. Indem Wahrnehmung angezeigt wird, macht der Chatter auch deutlich, dass er daran interessiert ist, die Kommunikation fortzusetzen. Bei Konversationen, wo keine Wahrnehmung gegeben wird, hört die Kommunikation oft auf. Mein Eindruck ist, dass viele Chat-Konversationen auf diese Art und Weise im Sande verlaufen. Im folgenden Beispiel (9) antwortet der Chatter le nicht auf den letzen Beitrag von My - er verzichtet vermutlich auf das Wahrnehmungssignal, weil ihm die Konversation ohnehin nicht länger interessant erscheint, nachdem My geschrieben hat, dass sie nicht Single ist. (9) (10:26:14) le => My: ålder (10:26:17) le => My: (10:26:22) My => le: tillräckligt. (10:26:42) le => My: ok (10:26:44) le => My: bor (10:26:48) My => le: hemma... (10:27:10) le => My: ok (10:27:15) le => My: singel 115 (10:27:25) My => le: nej (10:26:14) le => My: alter (10:26:17) le => My: (10:26:22) My => le: genug. (10:26:42) le => My: ok (10:26:44) le => My: wohnst (10:26:48) My => le: Zu Hause... (10:27:10) le => My: ok (10:27:15) le => My: Single (10:27:25) My => le: nein Sehen wir uns die Beschreibungen in Grammatiken an, wird die Funktion von ok als Wahrnehmung einer Sprachhandlung in der schwedischen Akademiegrammatik (Teleman et al 1999: 755) besprochen. In Beschreibungen für das Deutsche kommt diese Funktion nicht vor. Ok als Signal für Wahrnehmung kommt häufiger im schwedischen Chat-Raum als im deutschen Chat-Raum vor. In der deutschen Stichprobe kommt die Funktion ‚Wahrnehmung’ 20 Mal vor, in der schwedischen Stichprobe 57 Mal. Bei mehreren der deutschen Vorkommen ist die Funktion weniger deutlich als in den schwedischen Vorkommen. In Beispiel (10) ist die Abgrenzung zwischen Antwortsignal und Wahrnehmungssignal nicht offenbar. (10) (17:46:55) h: [zu N] ich schick dir mal was rüber (17:47:12) N: [zu h] okai, immer herdamit (17:47:47) h: [zu N] mom, jags rüber (17:47:57) N: [zu h] ok (17:48:09) N: [zu h] hoffentlich nicht mit der schneckpost (17:48:59) h: [zu N] nein, eilpost (17:49:15) N: [zu h] ok, noch ist nichts hier In Beispiel (10) macht h dem Chat-Partner N einen Vorschlag. Mit okai zeigt N nicht nur, dass er den Vorschlag wahrgenommen hat, sondern auch dass er dem Vorschlag zustimmt. Auch im zweiten Vorkommen kann ok als Zustimmung eines Vorschlages verstanden werden. Im dritten Vorkommen liegt aber eindeutig die Funktion ‚Wahrnehmungssignal’ vor. N stellt mit dem Beitrag „hoffentlich nicht mit der Schneckpost“ eine Frage an h, h beantwortet diese Frage mit einem 116 Nein. N bestätigt mit einem ok, dass er die Antwort wahrgenommen hat. Ok ist im dritten Vorkommen also ein Follow-up-Signal. 2.2.3 Ok als Einleitungssignal In den bisher beschriebenen Fällen besitzt ok an sich eine Funktion, in anderen Vorkommen steht ok dagegen in der Einleitung eines Beitrags und macht mit der Fortsetzung des Beitrags gemeinsam die Funktion aus. Ok bezieht sich in diesen Fällen nicht unmittelbar auf den Vorgängerbeitrag, indem ein Antwortsignal gegeben oder eine Wahrnehmung ausgedrückt wird. Im Unterschied zu den schon behandelten Funktionen bezieht sich ok nicht rückwärts auf den Vorgängerbeitrag. Stattdessen richtet sich ok vorwärts in Bezug auf die Fortsetzung des Chat- Beitrags. a) Ok als Einleitung einer Akzeptanz Nach einer Einwendung kann ok zeigen, dass diese akzeptiert wurde: (11) (00:01:11) s: [zu M] schau nicht so viel lol (00:01:53) M: [zu s] ich kann mit geschlossenen augen so schlecht chatten.....gg (00:02:39) s: [zu M] Ok ab und zu mal schauen zählt ja nicht ggg Hier fordert der Teilnehmer s den Teilnehmer M auf, nicht so viel zu schauen. M antwortet darauf, dass er mit geschlossen Augen nicht chatten kann. s akzeptiert diese Einwendung mit einem ok und meint, dass „ab und zu mal schauen“ nicht zählt. Die Funktion Akzeptanz wird nicht vom Lexem ok allein getragen, sondern wird erst mit der Fortsetzung des Beitrags deutlich. Mit Hilfe der Fortsetzung kann die Bedeutung als „einverstanden“ interpretiert werden. Im Unterschied zu den Belegen, die als Wahrnehmungssignal klassifiziert wurden, folgt dem ok bei dieser Kategorie immer eine Äußerung von Akzeptanz. Würde ok in dem Beitrag allein vorkommen, hätte die Funktion ‚Wahrnehmungssignal’ angenommen werden müssen.. Auch in Beispiel (12) geht die Funktion ‚Akzeptanz der Einwendung’ erst aus der Fortsetzung des Beitrags hervor. Indem X die Frage stellt „was soll ich denn 117 sagen“ und damit zeigt, dass sie die Einwendung von L („ich mag keen neid) akzeptiert, wird deutlich, dass ok als Signal für Akzeptanz der Einwendung eingesetzt wird. (12) (15:21:48) (15:22:08) X: [zu L] heyyy sag nicht sowas ich werde ganz neidisch (15:22:24) L: [zu X] ich mag keen neid...ggg (15:22:40) X: [zu L] ok was soll ich denn sagen?...gg In den Grammatiken wird diese Funktion von ok nicht behandelt. Im Deutschen Chat-Raum wird ok 18 Mal als Signal für Akzeptanz eingesetzt, im Schwedischen Chat-Raum viermal. b) Ok als Einleitung einer Folgerung Bei dieser Kategorie besteht die Fortsetzung des Beitrags nach dem ok aus einer Folgerung des Inhalts im Vorgängerbeitrag. Der Verwender von ok geht auf den Inhalt im Vorgängerbeitrag ein und macht von ihm ausgehend eine Folgerung. In Beispiel (13) möchte sich N aus dem Chat-Raum verabschieden. M protestiert dagegen. N fällt dann ein, dass er noch nicht gehen kann, da er noch Erdbeeren zu essen hat. M kommt zu dem Schluss, dass er die Beeren langsam essen soll, damit er länger im Chat bleiben kann, diese Schlussfolgerung kündigt er mit einem ok an: (13) (23:18:59) N: [zu M] so .. noch ein Bierchen und dann ist aber schluss für heute (23:19:12) M: [zu N] ne ne, du kannst noch nicht gehen (23:19:34) N: [zu M] stimmt .. ich habe die Erdbeeren noch her stehen gg (23:19:47) M: [zu N] ok, jede beere langsam essen gelle In Beispiel (14) glaubt v, dass er auf der Ignore-Liste von c steht. C erklärt daraufhin, wie er zu der Ignore-Funktion steht. Ausgehend von dieser Auslegung kann v darauf schließen, dass er nicht auf der Ignore-Liste steht. (14) (23:35:52) v: [zu c] ach?... und ich dachte ich stehe bei dir auf ignore sfg (23:38:00) c: [zu v] ich bin mindestens 5 jahre im chat und auf 118 ignore stand bei mir noch niemand , allerdings gibt es drei leute , die bei mir auf geistigen ignore stehen , das heisst , sie sind mir nichtmal ein hi wert (23:38:29) v: [zu c] ok .. bei dem text bin ich keine der drei grins In der deutschen Stichprobe sind 17 Vorkommen als Signal für eine Folgerung klassifiziert, in der schwedischen Stichprobe sind es fünf Vorkommen. In vielen Fällen ist es schwierig, eine eindeutige Funktion für ok festzulegen. Bei vielen Belegen ist auch die Interpretation als Wahrnehmungssignal möglich. Im folgenden Beispiel (15) kann nicht eindeutig festgehalten werden, ob Ju ok einsetzt um jo mitzuteilen, dass sie den Beitrag wahrgenommen hat und dann unabhängig von ok noch einen Kommentar hinzufügt, oder ob ok ein Bestandteil der Folgerung ausmacht. (15) (15:20:24) jo [zu Ju] auch päpstinnen sind sicher dem zölibat unterworfen :o) (15:20:56) Ju: [zu jo] ok..päpstin is nix für mich In dieser Studie wurde die Wahl getroffen, alle Belege als Folgerung zu klassifizieren, bei denen die Interpretation ‚Folgerung’ möglich ist. . 2.2.4 Ok beim Übergang In vielen Belegen bezieht sich ok nicht auf einen spezifischen Beitrag vom Chat- Partner, sondern es wird auf die gesamte Konversation Bezug genommen. Ok leitet dann eine Folgerung aus der gesamten Konversation ein. Oft wird mit dem ok ein Übergang vorbereitet. Im folgenden Beispiel wird ein Themenübergang eingeleitet: (16) (00:48:53) U: [zu D] ich glaube darauf darfst du dir nichts einbilden ... (00:48:55) D: [zu U] ja .........? (00:49:04) U: [zu D] der kann nicht anders ... (00:49:19) U: [zu D] ... schön ..? (00:50:02) U: [zu D] ok ... - reden wir lieber über holländische maler ... In Beispiel (17) wird auf einen Wechsel der Sprache vorbereitet. Die Teilnehmer A und M diskutieren ihre Russischkenntnisse. Nachdem sie etwas diskutiert 119 haben, kommt M zu dem Schluss, dass sie das Gespräch auf Russisch weiterführen könnten, die Folgerung wird mit einem ok eingeleitet. (17) (14:58:38) A => M: stavar du bra på ryska då... (14:59:23) M => A: näääe (14:59:36) M => A: gör du???? (15:00:09) A => M: men pratar många språk flytande..... (15:00:25) A mumlar: Ibland..... (15:00:46) A mumlar: beror på hur många öl jag fått... (15:00:59) M => A: ok då kör vi lite ryska.? (14:58:38) A => M: schreibst du denn gut auf Russisch... (14:59:23) M => A: neeee (14:59:36) M => A: machst du???? (15:00:09) A => M: aber spreche viele Sprache fließend..... (15:00:25) A murmelt: manchmal..... (15:00:46) A murmelt: hängt davon ab, wie viel Bier ich getrunken habe.... (15:00:59) M => A: ok dann machen wir etwas Russisch.? In Beispiel (18) leitet J die Verabschiedung aus dem Chat-Raum mit einem ok ein. (18) (18:39:48) J: [zu S] Du lüchst.... schlimmer als Dein Sohn gg (18:39:52) S: [zu J] lach....wenn du es sagst... (18:40:05) S: [zu J] nu lass ma den kleenen in ruhe..das ist mein ganzer stolz... (18:40:28) J: [zu S] lach... ich sag ja schon nix mehr..... bevor ich mich schlagen lasse gg (18:40:55) S: [zu J] auf den kleinen..lass ich nix kommen...der is erst fümpf (18:43:12) (18:43:18) J: ah... geht doch gg (18:44:04) J: [zu S] ok... ich geh auch noch mal raus... schnell noch was einkaufen was ich mal wieder vergessen hab.... grrrr In der SAG wird unter der Funktion „Einleitung eines neuen Themas“ die vorwärts gerichtete Funktion von ok aufgenommen (Teleman et al 1999: 756). Auch in der GdS wird unter der Funktion ‚Kontinuitätsversicherung’ erwähnt, 120 dass okay an Diskursstellen mit thematischem oder handlungsstrukturellem Übergang zu finden ist (Zifonun et al 1997: 376). 2.2.5 Bezug unsicher In einigen Belegen in der Stichprobe kann nicht herausgefunden werden, worauf sich ok tatsächlich bezieht. Auch beim genaueren Studium von längeren Ausschnitten im Logfile kann keine Kategorie mit Sicherheit festgestellt werden. Möglicherweise sind bei diesen Fällen nicht sämtliche Beiträge im Chat-Raum publiziert, sondern teilweise privat übermittelt worden. Die Konversation im Chat-Raum schließt auch nicht aus, dass andere parallele Kommunikationswege verwendet werden, wie z.B. Telefon oder E-mail. Das Logfile vom Chat gibt dann nur einen Teil der Kommunikation wieder. (19) 12:52:42 m: >>> '-J-' kommt in den Raum ab40 geschlichen. 12:52:55 -J-: wieder da lach...geflogen 12:52:56 d: [zu -J-] ciao 12:53:00 N: [zu -J-] ups was los ? ggg auffang 12:53:12 -J-: [zu N] uiuiui...danke:-)) 12:53:35 N: [zu -J-] aber nicht jetzt dauernd das wiederholen gelle lol 12:53:37 -J-: [zu DC] doch habe dich gemeint...lach...wollte dir mal was gutes tun....gg 12:53:52 -J-: [zu N] wäre doch schöööööööön:-)) 12:54:16 DC: [zu -J-] OK, danke, ich kopier mir sofort den Log und druck den aus gg 12:54:23 -J-: muss raus...bis gleich 12:54:26 m: <<< '-Jette-' verläßt diesen Raum und den Chat Bei Beispiel (19) kann mit Hilfe dieses Ausschnittes nicht erkannt werden, worauf sich DC mit seinem Beitrag bezieht. Möglicherweise hat DC auf einem anderen Kommunikationsweg etwas erhalten, wofür er sich im Chat bedankt. 2.2.6 Versteckter Teilnehmer In vielen Fällen ist es deutlich, dass nur einer der Teilnehmer offen im Chat-Raum schreibt, während der andere Teilnehmer seine Beiträge privat nur zum Chat- Partner schickt. Der Anteil der Beiträge mit einem versteckten Teilnehmer ist im schwedischen Chat groß. Bei über einem Viertel der Beiträge in der Stichprobe ist 121 nur einer der Teilnehmer eines Dialogs im Chat-Raum sichtbar. Bei einem Durchsehen dieser Beiträge kann bei vielen die Funktion ‚Wahrnehmungssignal’ vermutet werden. Oft geht es um ok als Follow-up nach einer Frage. (20) (00:10:54) m => P: var är du ifrån då? (00:11:47) m => P: skåne (00:12:17) m => P: ok, jag är 39 (00:12:56) m => P: nää, nordöstra skåne (00:10:54) m => P: wo stammst du her denn? (00:11:47) m => P: schonen (00:12:17) m => P: ok, ich bin 39 (00:12:56) m => P: nöö, nordöstliches schonen In Beispiel (20) schreibt P privat an m. Ausgehend von den Beiträgen von m können wir erraten, dass sich ok auf eine Aussage bezüglich des Alters von P bezieht. Da die Beiträge von P nicht im Logfile erfasst sind, können wir dies aber nicht mit Sicherheit feststellen. Aufgrund der großen Unsicherheit sind diese Beiträge aus der Analyse ausgegrenzt worden. Diese Beiträge mit versteckten Teilnehmern werden als eine eigene Kategorie in der Statistik (Siehe Abschnitt 3) vorgelegt. 3. Zusammenfassung Die quantitativen Unterschiede zwischen der deutschen und schwedischen Stichprobe werden in Abbildung 4 angegeben: Abb. 4: Funktionen von ok in den Stichproben Deutsch Schwedisch Antwortsignal 31 2 Wahrnehmungssignal 20 57 Einleitungssignal a) Signal für Akzeptanz 18 4 b) Signal für Folgerung 17 5 Signal beim Übergang 5 3 Bezug unsicher 7 3 122 Versteckter Teilnehmer 1 26 Adjektiv 1 0 Vorkommen in der Stichprobe 100 100 Aus Abbildung 4 geht hervor, dass die Funktion ‚Wahrnehmungssignal’ in der schwedischen Stichprobe deutlich überwiegt, denn hier können nur einzelne Vorkommen der anderen Funktionen belegt werden. Der Anteil der Beiträge mit einem versteckten Teilnehmer ist im schwedischen Chat leider groß und macht ein Viertel aller Beiträge in der Stichprobe aus. In der deutschen Stichprobe ist die Kategorie ‚Antwortsignal’ am häufigsten. Die empirische Untersuchung bestätigt die Vermutung, dass ok im schwedischen Chat oft nur Wahrnehmung zeigt, während im deutschen Chat durch ok häufiger Zustimmung ausgedrückt wird. Wie die Untersuchung darlegt, kommt ok auch in anderen Funktionen vor. In der Einleitung eines längeren Beitrags kann ok auch Akzeptanz vermitteln oder eine Folgerung vorbereiten. Ok kann auch Übergänge bei Thema, Sprache oder Anwesenheit im Chat-Raum einleiten. Bei den Kategorien ‚Antwortsignal’, ‚Wahrnehmungssignal’ und ‚Einleitungssignal’ wird auf einen spezifischen Vorgängerbeitrag reagiert. Hier können wir von ‚Rückmeldung im weiteren Sinne’ sprechen. Bei der Kategorie ‚Signal beim Übergang’ wird jedoch auf die gesamte Konversation Bezug genommen, hier erscheint der Begriff Rückmeldung problematisch. In den Grammatiken Zifonun et al (1997) und Teleman et al (1999) sind diese Funktionen bereits zum Teil beschrieben worden. An erster Position im Beitrag wird ok in der schwedischen Grammatik als Antwortsignal, Wahrnehmungssignal und Signal für einen Themenübergang beschrieben. In der deutschen Grammatik wird die Funktion ‚Wahrnehmung’ nicht beschrieben, sondern nur als Antwortsignal oder beim Übergang im Gespräch. Diese Untersuchung zeigt jedoch, dass die Funktion ‚Wahrnehmung’ auch in der deutschen Chat- Kommunikation belegt werden kann. Aufgrund der Einschränkung, nur Vorkommen an erster Stelle im Beitrag in die Stichprobe zu inkludieren, konnten die Funktionen ‚Gliederungspartikel’ und ‚Rückversicherungssignal’ hier nicht 123 belegt werden. Bei der Funktion ‚Gliederungssignal’ ist okay in den Beitrag integriert, bei der Funktion ‚Rückversicherungssignal’ steht okay am Ende der Äußerung. Die Unvollständigkeit der Beschreibung von ok in den Grammatiken hängt wahrscheinlich zum Teil damit zusammen, dass ok nicht als eigenes Lexem behandelt wird, sondern als Alternative zu anderen Lexemen genannt wird. Eine größere empirische Untersuchung von ok in der gesprochenen Sprache hätte eventuell andere Funktionen berücksichtigt. Anzunehmen ist aber, dass die Funktion als Wahrnehmungssignal gerade in der Chat-Kommunikation besonders wichtig ist, da eine Rückmeldung parallel mit der Produktion technisch nicht möglich ist. Stattdessen wird Rückmeldung mit Hilfe von Wahrnehmungssignalen nach dem Präsentieren des Beitrags gegeben. Auf dem Bildschirm sind mehrere Dialoge ineinander verflochten2, wenn viele Teilnehmer eingeloggt sind, verschwindet der Text schnell von der Bildfläche. Es ist in solchen Fällen einfach, einen Beitrag zu übersehen. Aus diesem Grund ist es für die Konversationsteilnehmer wichtig zu zeigen, dass man den an sich selbst gerichteten Beitrag gesehen hat. Da die Chat-Teilnehmer sich nicht hören oder sehen können, wissen sie auch nicht, ob der Partner überhaupt noch vor dem Computer sitzt. Auch wenn man selbst nichts Inhaltliches zur Konversation beiträgt, ist es wichtig zu zeigen, dass man den Beitrag des Partners empfangen hat. Diese Untersuchung hat gezeigt, dass das Auftreten der Funktion ‚Wahrnehmungssignal’ bei ok sowohl im deutschen als auch im schwedischen Chat möglich ist. Die Dominanz dieser Funktion in der schwedischen Stichprobe kann auch erklären, warum das Lexem ok im schwedischen Chat-Raum viel häufiger vorkommt als im deutschen Chat-Raum. Literatur: Beißwenger, M. 2003. „Sprachhandlungskoordination im Chat“. In Zeitschrift für germanistische Linguistik 31 (2). 198-231. 2 In diesem Artikel sind Beiträge von anderen Teilnehmern aus den Beispielssegmenten aus Platzgründen gestrichen worden. 124 Brinker, K. und Sager, S. F. 1996. Linguistische Gesprächsanalyse. Eine Einführung. 2. durchgesehene und ergänzte Auflage. Berlin. Erich Schmidt. Duden. Deutsches Universalwörterbuch. 2003. 5., überarbeitete Auflage. Mannheim/Leipzig/ Wien/Zürich. Duden-Verlag. Etymologisches Wörterbuch des Deutschen. 1997. München: Deutscher Taschenbuch Verlag. Schönfeldt, J. 2001. ”Die Gesprächsorganisation in der Chat-Kommunikation”. In Beißwenger, M. (Hrsg.): Chat-Kommunikation. Sprache, Interaktion, Sozialität & Identität in synchroner computervermittelter Kommunikation. Perspektiven auf ein interdisziplinäres Forschungsfeld. Stuttgart: Ibidem Verlag. 25-53. Sinclair, J. M. und Coulthard, M. 1975. Toward an analysis of discourse: The English used by teachers and pupils. London: Oxford university press. Stenström, A-B. 1982. „Feedback“. In Enkvist, N.E. (Hrsg.) Impromptu speech: A symposium. Åbo: Åbo Akademi. 319-340. Storrer, A. 2001. „Sprachliche Besonderheiten getippter Gespräche. Sprecherwechsel und sprachliches Zeigen in der Chat-Kommunikation“. In Beißwenger, Michael (Hrsg.): Chat-Kommunikation. Sprache, Interaktion, Sozialität & Identität in synchroner computervermittelter Kommunikation. Perspektiven auf ein interdisziplinäres Forschungsfeld. Stuttgart: Ibidem Verlag. 3-24. Svensk ordbok .1999. Språkdata. Göteborgs universitet. Keuruu: Otava Svenska Akademiens ordbok. Svenska Akademien. http//: ww.saob.se (21.3.2007). Svenska akademiens ordlista över svenska språket. 2006. Stockholm: Svenska akademien: Norstedts akademiska förlag. Teleman, U., Andersson E. und Hellberg S. 1999. Svenska Akademiens grammatik (SAG). Stockholm: Norstedts. Willkop, E-M. 1988. Gliederungspartikeln im Dialog. München: Iudicum Verlag. Zifonun, G, Hoffmann L. und Strecker B.1997. Grammatik der deutschen Sprache (GdS). Berlin: Walter de Gruyter. 125 Oh as a discourse marker Mayumi Nishikawa, Setsunan University 1. Introduction The term discourse marker (henceforth DM) has been used mainly to refer to the words or expressions whose main functions are to connect two sentences (utterances) or phrases and to mark the relationship between them. It is not assumed, however, that there is necessarily a unified category “DM” that is strictly defined in theoretical terms. Thus linguists use this term in various ways, according to their own definitions (Schourup 1985 and 1999; Schiffrin 1987; Blakemore 1987 and 2002 etc.). In this paper I use the term “DM” to refer to items which are used intentionally by speakers to constrain the hearers’ inferential phase in interpreting the host-utterances, and which have several characteristics pointed out by Schourup (1999: 230-234): non-truth conditionality, connectivity, optionality, weak clause association, initiality, orality and multi-categoriality. The interjection oh is categorized as a DM by some linguists, such as Schourup (1985 and 1999) and Schiffrin (1987). Especially when oh attaches to its host utterances, it functions as DM in that it affects the hearers’ interpretation of the host utterances which follow. Consider the following examples (c.f. Schourup 1985): (1) [As soon as Mary meets Tom, she recognizes that she forgot to call him the previous night as she had promised. Mary says.] a. Oh! b. I didn’t call you last night. c. Oh, I didn’t call you last night. Oh in (1a) is a reaction of ‘surprise’ as Mary suddenly realizes that she had forgotten to call Tom the previous night as she had promised, and thus represents Mary’s mental state. (1b) is a full-fledged sentential utterance, wherein Mary says the truth that she did not call Tom the previous night. (1c) is an utterance adding 126 oh to (1b), but what is communicated by (1c) is quite different, because the implication ‘Mary failed to call Tom last night accidentally, not deliberately’, is communicated. Of course, in (1b) a similar implication could be communicated by exploiting non-linguistic devices, such as intonation or facial expression. Moreover, Mary can convey the same implication with oh alone, as in (1a). (1c) is, however, quite different from (1a) and (1b), because in (1c), by using both oh and the sentential utterance, Mary can communicate the implication much more strongly and clearly. Thus it can be said that oh is a DM affecting the hearer’s interpretation of the oh-marked utterance. In this paper I will show the function of the DM oh, and what pragmatic effects it yields in the framework of relevance theory (Sperber and Wilson 1995). 2. Previous work 2.1 Schiffrin’s coherence-based approach Schiffrin (1987), under the assumption that DMs are coherence options by which speakers and hearers can maintain the flow of text in discourse, closely analyzes the interjectional DM oh, by using data based on daily conversation in her neighborhood. Schiffrin claims, “[o]h is a marker of information management: it marks shifts in speaker orientation (objective and subjective) to information which occur as speakers and hearers manage the flow of information produced and received during discourse” (1987: 100-101). Consider her examples: (2) Jack: I think it was in seventeen: fifteen, or seventeen fifty five, I’m not sure when. Eh: oh I’m wrong. Seventeen seventeen. (1987: 75) (3) Jack : How bout uh. . .how bout the one uh. . .uh. . Death of a Salesman? Freda: Well that was a show, sure. Jack : Oh that was a movie too. (1987: 76) (4) Freda: Sometimes he got a notice for staying out past curfew. Recently. In August, that was. Val : Oh curfew? What’s curfew? (1987: 81) (5) Val : Is it safe? Freda: Uh: we found a safe way! But it’s the long way! Val : Oh it’s a special way? (1987: 85) (6) Debby: So what, you have three kids? 127 Irene : I have four. Three boys and a girl. Debby: Oh I didn’t know that. (1987: 89) According to Schiffrin, oh in (2) prefaces self-initiated repair, oh in (3) does other-initiated repair. Oh in (4) is used when clarification is requested, and oh in (5) occurs with a question where the speaker requests elaboration. In (6), oh occurs when the questioner acknowledges receipt of anticipated information. Thus, Schiffrin shows oh’s role in information state transitions, saying “oh marks a focus of speaker’s attention which then also becomes a candidate for hearer’s attention” (1987: 99). There are, however, several problems in Schiffrin’s view. First Schiffrin’s claim that the DM functions of oh and well have to be identified by examining characteristics of the discourse slots in which the markers occur, because neither of them lacks semantic (referential) meaning (1987: 73) cannot explain why oh and well in the same slot play different roles.1 Consider the following example: (7) A: Are you happy? B: Oh, yes./Well, yes. (cf. Schourup 1999: 243) In (7), both oh and well are acceptable in the same slot as a response to A’s question. Oh, however, may inform the hearer that A’s question or A’s intention to ask the question or B’s thought ‘B is happy’ has just come into B’s mind as something appreciable. In a situation where A is a friend who is always playful and suddenly asks such a serious question, the recognition may be accompanied with ‘unexpectedness’ or ‘surprise’. On the other hand, well may inform the hearer that answering ‘yes’ requires B some consideration, that is, it is not so easy for B to answer ‘B is happy’ quickly. If Schiffrin’s view that oh and well are ‘meaningless’ is right, what each utterance conveys should be almost the same, because they fill the same discourse slot at least in this case. This fact suggests that examining the slots where the markers occur is not enough to fully explain their functions and that the function of each DM should also be derived from what is encapsulated in the DM (oh or well) itself. 1 Schiffrin claims that understanding DMs requires one to consider two kinds of contributions: the contribution made by the marker itself and that made by characteristics of the discourse slot in which the marker occurs (1987: 73). 128 Second, Schiffrin’s view cannot explain DMs which appear discourse-initially. Schiffrin defines DMs as “sequentially dependent elements which bracket units of talk” (1987: 31). Most DMs including oh, however, often occur without any preceding discourse (Blakemore 1987: 85-86; Schourup 1999: 230-231). Consider the following examples: (8) Tom: Why didn’t you call me last night? Mary: Oh, I forgot to call you last night! (9) [Mary recognizes that Tom seems angry at her, and says] Oh, I forgot to call you last night! Mary’s utterances in (8) and (9) can be interpreted as ‘Mary failed to call because of her forgetfulness, not malevolent intention’. Oh in (8) is, however, used to connect two utterances, Tom’s and Mary’s, while oh in (9) is used to link the assumption Mary made from Tom’s appearance (his angry expression) and Mary’s utterance. In both examples, oh seems to play almost the same role, because without it, Mary will appear unfriendly, blunt or even impolite. This fact suggests that what DMs connect is not two text segments but two propositional contents, information or assumptions which are drawn from on-going contexts. Therefore, Schiffrin’s model for DMs does not fully account for the true behavior of DMs in conversation. Third, Schiffrin’s claim that “well marks responses at an interactional level, and oh marks responses at a cognitive level” (1987: 127) cannot explain the fact that oh also contributes on the interactional or interpersonal level. (10) Teacher: Good bye, class. See you next week. Mary: . . .Well/Oh, can I ask one more question? According to Schiffrin, well in (10) occurs when coherence is violated in that Mary knows that it is too late to ask any questions to her teacher because the class is over, but she is going to ask a question anyway. In this case well is used as an interactional device by making the conversation coherent. Oh in (10), however, may be also used as a social device, because Mary, who knows that it is not a good time to ask, intends to show ‘politeness’ to the teacher. Using oh means that Mary has just thought of an important question to ask, not that she has intended to 129 bother her teacher. This fact suggests that oh works not only at a cognitive level but also at an interactional level. It seems that Schiffrin’s view misses this point. In sum, Schiffrin examines closely where the DM oh typically occurs, but she does not seem to fully account for its function. 2.2 Bolinger’s gestural approach Bolinger closely examines several pragmatic effects in the usage of oh in discourse on the basis of his gestural analysis of interjections, saying, “the contribution of oh to discourse exploits our inferential mechanisms in far more complex ways” (1989: 269).2 Bolinger writes, “[b]eyond whatever concentration of pure surprise oh may carry, we have the things that surprise may imply, and the facets of usage that are organized around those implications” (ibid.) and claims that these implications are ‘magnitude’ and ‘spontaneity’. According to him, the first implication comes from our tendency to be surprised at something large rather than something small and the second implication comes because oh is originally a spontaneous vocal reaction to stimulus. In addition, Bolinger claims that oh suggests ‘enhancement’ from ‘magnitude’, and ‘frankness’ and ‘sincerity’ from ‘spontaneity’. I will discuss Bolinger’s views on a range of discourse effects which oh may imply in context. Consider the following: (11) June’s getting divorced. ? Oh how awful! When? (1987: 270) (12) Is she good-looking? ? Oh beautiful! (ibid.) (13) I’ve just heard a juicy story. ? Oh tell me! (1987: 271) In (11) and (12), according to Bolinger, oh is used to express ‘enhancement’, where the speakers may strengthen their emotions and feelings. In (13) the speaker is conveying his or her strong ‘enthusiasm’ toward someone’s information by using oh. Bolinger’s claim that “[t]he sociability component of oh is as important as the surprise-spontaneity component” (1989: 272) is illustrated in the following cases (1989: 273-274): 2 Bolinger does not use the term DM in his analysis, but his research shows some aspects of the DM usages of interjections. 130 (14) I plan to take Nancy to the party. ? Oh you do? That’s great. (15) I plan to take Nancy to the party. ? Oh you do? I wonder if that’s wise. In (14) and (15), the fact that the second utterances would sound very flat or unfriendly without oh suggests that the speakers may implicate ‘friendliness’ or ‘sociability’ to the hearers by using oh, whether the speakers agree with someone as in (14) or disagree as in (15). (14) seems similar to ‘enhancement’ cases like (11) and (12), because (11) and (12) would sound unfriendly without oh. The same thing can be said in ‘enthusiasm’ cases like (13). It seems to be difficult, therefore, to differentiate ‘sociability’ cases and ‘enhancement’ or ‘enthusiasm’ cases. In my view, ‘enhancement’ and ‘enthusiasm’ may be more or less associated with ‘sociability’ in these examples. Bolinger points out the effect of ‘mitigation’ in the usage of oh, saying “[a]n oh blunts what otherwise might be objectionable” (1989: 276), and illustrates that oh is frequently used when we command as in (16) (ibid.), decline other person’s offer as in (17) (ibid.) and express doubt without being confrontational as in (18) (1989: 278): (16) Oh just be quiet for a minute! (17) Oh that’s too much! Half is enough! (18) It was all Mary’s fault. ? Oh I don’t know. Don’t you think she did the best she could? According to him, in (16) oh used before a command may mitigate the confrontational implication yielded by commanding. In (17), oh used before declination of someone’s offer may mitigate impoliteness yielded by declining. In (18), oh used before doubt or lack of confidence may also mitigate rude implication yielded by such behaviors. Bolinger closely examines several aspects accompanied by oh and beautifully explains how a range of pragmatic effects of oh can be gained. Bolinger’s idea is that the semantic components included in oh are exploited to yield implications in discourse. I agree with him in that what is encapsulated in oh itself affects a range of implications derived in discourse. Bolinger’s gestural analysis, however, cannot fully explain why and how each implication of oh can be derived in a certain context. Specifically his study cannot account for why oh implies ‘enhancement’ 131 in (11) but does not in other cases. In my view, in all the cases above oh is simply used to indicate its core ‘meaning’ related to the speaker’s mental activity and the various pragmatic effects are achieved via the interaction of oh’s core ‘meaning’, the contexts, and what is conveyed by the host utterances. 3. The core ‘meaning’ of oh3 Oh is often said to indicate ‘surprise’ (Carlson 1984: 69-75). The meaning of oh, however, appears to be more complicated than has been said, because there are some examples where oh does not necessarily indicate the speaker’s ‘surprise’ (James 1973; Bolinger 1989). In addition, as long as most interjections come from natural vocal reactions toward external or internal stimuli, they represent ‘surprise’ to some extent. The idea that oh simply represents ‘surprise’, therefore, cannot account for oh’s role in discourse. Nishikawa, regarding the assumptions strongly indicated by interjections as their core ‘meanings’, investigates them on the notion of natural resemblance (see Nishikawa 2007a, 2007b and 2007c for details). The natural resemblance of interjections occurs at the phonetic level; the resemblance lies between a naturally occurring vocal sound and the conventionalized sound of an interjection and thus the resemblance between assumptions is automatically derived from that phonetic resemblance. In the case of oh, as Bolinger claims, it is not only related to ‘surprise’ but also its rounding shape represents “a blend with the kissing gesture” (Bolinger 1989: 266) produced by care-givers. That is, oh is inherently a sound which is spontaneously and biologically produced by specific muscle movements when the speaker has become aware of something unexpected but somehow appreciated. In the following examples, it seems that what the speakers intend to convey by oh is not only ‘unexpectedness’ toward the information they got but also the 3 I will use interjection ‘meaning’ to differentiate it from the meaning linguistically encoded in an ordinary word. 132 ‘importance’ or ‘appreciability’ of the information or the fact that they have gotten the information.4 Consider: (19) [Miranda doesn’t find anything good for that day’s “run-through” and is complaining to her co-workers. Suddenly she has found a lovely dress.] Miranda: Oh. This is . . . this might be . . . what do you think? (The Movie: The Devil Wears Prada) (20) Bill: Who’s that over there? Violet: Where? Oh, Katherine Watson. New teacher. Art History. Dying to meet her. (The Movie: Mona Lisa Smile) In (19), Miranda is using oh to represent that she found a good dress unexpectedly and also that she is pleased with that. In (20), by oh Violet is expressing not only her ‘surprise’ on seeing Katherine Watson but also the fact is appreciable to her. This vocal sound has played an important role in basic human life and has been imitated and stylized into a conventional interjection, which has made it more effective for people to communicate with each other. In other words, people recognize that a vocal reaction makes a certain assumption strongly manifest and, exploiting that, use it as the ostensive communication called interjection. I conclude, therefore, that the core ‘meaning’ of oh is “Something appreciable has just come into the speaker’s mind” and show this ‘meaning’ is involved in various functions played by oh in discourse. In section 5, I will illustrate, by using the core ‘meaning’, how the hearer interprets oh-marked utterances. 4. Toward the functions of DM oh DMs and similar expressions put some constraints on the interpretation of their host utterances by providing additional information. In this section, by comparing these items with oh and showing the similarities and the differences between them, I will closely investigate the exact characteristics of the DM oh, based on its nature and core ‘meaning’. 4 The Ohs in (19) and (20) are not vocal reactions any more but conventionalized interjections. The assumptions, however, are the same. 133 4.1 DM oh and discourse connectives Blakemore (1987) proposes a new approach to so-called DMs, which she calls discourse connective, in the framework of relevance theory. She defines discourse connectives (e.g. so, therefore, however, but) as “expressions that constrain the interpretation of the utterances that contain them by virtue of the inferential connections they express” (1987: 105), and claims that their function is “to guide the interpretation process by specifying certain properties of context and contextual effects” (1987: 77). Consider Blakemore’s examples: (21) Tom can open Bill’s safe. He knows the combination. (22) Tom can open Bill’s safe. So he knows the combination. (23) Tom can open Bill’s safe. After all he knows the combination. In (21), the semantic relationship between two utterances is ambiguous. In other words, the hearer cannot determine whether what is meant by the second segment is a result of what is meant by the first segment or a reason for what is meant by the first segment. In (22), however, since two segments are connected by so, the hearer can identify that what is meant by the second segment is a conclusion which the speaker infers from what is meant by the first segment. In (23), since two segments are connected by after all, the hearer can identify that what is meant by the second segment is a justification which reminds the hearer of what is meant by the first segment. Blakemore concludes, therefore, that so encodes the procedural meaning ‘Process the utterance following so as an implicated conclusion’ and after all encodes the procedural meaning ‘Process the utterance following after all as a justification’.5 The DM oh is very similar to these discourse connectives, because it relates the assumptions, information or propositional contents which are gained from the preceding utterances or the context to the interpretation of the following utterances, because they are not truth-conditional, and because they do not have compositionality. The DM oh is, however, different in some ways. Compare so and oh : (24) a. A: Tom can open Bill’s safe. 5 Relevance theory assumes two meanings linguistically encoded: conceptual meaning and procedural meaning. See Wilson and Sperber (1993: 10) for the details. 134 B: Oh, he knows the combination. b. A: Tom can open Bill’s safe. B: So, he knows the combination. (25) a. A: Tom can open Bill’s safe. B: He knows, oh, the combination. b. A: Tom can open Bill’s safe. B: *He knows, so, the combination. Oh in (24a) may indicate that the speaker B has just become aware of the important fact that Tom knows the combination of Bill’s safe on hearing A’s utterance. On the other hand, so in (24b) may indicate that B concludes, on the basis of what is conveyed by A’s utterance, that Tom knows the combination of Bill’s safe. Oh in (25a) may indicate that B has just become aware of the important fact that it is Bill’s safe combination that Tom knows. On the other hand, in (25b), so is not acceptable because grammar does not allow it to appear in the middle of the sentence. It can be concluded, therefore, that the discourse connective so only occurs sentence-initially and the interpretive scope which so affects is always the whole host utterance. On the other hand, the DM oh can occur at several slots within the host utterances (perhaps not sentence-finally) and the scopes which oh affects are always the elements which follow it. Also so affects the interpretation of host utterances by clarifying the semantic relationship between two related segments, but, on the other hand, oh affects the interpretation of host utterances by showing the speaker’s current mental state. In other words, with the DM oh, by showing what mental state or mental activity the speaker is in, the hearer is instructed to interpret the following utterance in a particular way. 4.2 DM oh and adverbial discourse connectives Rouchota (1998) points out the difference between adverbial discourse connectives and non-adverbial discourse connectives, but treats adverbial discourse connectives, such as moreover and after all, like so, as constraining implicatures: they “encode procedural meaning which constraints the implicit content of the utterance” (1998: 119).6 Adverbial discourse connectives are 6 Relevance theory assumes two kinds of communicated assumption: explicature and implicature. See Sperber and Wilson (1995: 182-202) for the detail. 135 superficially more similar to the DM oh in that these connectives are used parenthetically. Several differences can, however, be found as follows: (26) a. Oh, Bill believes Mary to have left yesterday. b. After all/ Moreover Bill believes Mary to have left yesterday. (27) a. Bill believes, oh, Mary to have left yesterday. b. Bill believes, ?after all/?moreover, Mary to have left yesterday. (28) a. Bill believes Mary, oh, to have left yesterday. b. Bill believes Mary, after all/ moreover, to have left yesterday. (29) a. Bill believes Mary to have left, oh, yesterday. b. Bill believes Mary to have left, ??after all/??moreover, yesterday. (30) a. Bill believes Mary to have left yesterday, *oh. b. Bill believes Mary to have left yesterday, after all/ moreover. The most striking difference can be seen in their distribution pattern: the adverbial discourse connectives occur sentence-finally, while oh cannot occur at the end of the host-utterance. Another distribution difference is, as seen in the examples (27) and (29), that oh can occur in the slots where after all or moreover cannot appear. The reason for this is that while the distribution of these adverbial discourse connectives are restrained by grammar, a rule of language (Rouchota 1998: 123), the interjection oh has no syntactical relationship with other elements of the utterance but is related to “the real time moment of occurrence of [. . .] thinking in order to establish the timeliness of a speaker’s reaction” (Schourup 1985: 21). In other words, since DM oh indicates that a thought, an assumption or a piece of information has just come into the speaker’s mind, it has more freedom to occur in the host utterances. The distinction between adverbial discourse connectives and the DM oh can also be found in the interpretive scope that these expressions affect. For example, adverbial discourse connectives always affect the interpretation of the whole host- utterance wherever they occur, and thus (26b) can be interpreted in the same way as (28b) or (30b). On the other hand, the scope that the DM oh affects is determined by the elements that follow it, so (26a), (27a), (28a) and (29a) are all interpreted differently by the hearer. Specifically, what has just come into the speaker’s mind as an important fact is ‘Bill believes Mary to have left yesterday’ in (26a), ‘Mary left yesterday’ in (27a), ‘left yesterday’ in (28a) and ‘yesterday’ in (29a). Thus, (30a) is not acceptable because there are no elements following oh. 136 The final difference between them can be found in the basic functional contributions. Like so, adverbial discourse connectives guide the hearer to interpret the host-utterances in the way the speaker intends by making explicit the semantic relationship between two segments. They are relevant in reducing the hearer’s processing effort by clarifying that relationship. On the other hand, the DM oh guides the hearer to interpret the host-utterances, by showing the speaker’s mental activity or state which oh ‘means’ and establishing contexts to interpret the following elements of the host-utterances. In other words, the DM oh reduces the hearer’s processing effort by establishing the context related to the speaker’s mental state or mental activity. 4.3 DM oh and sentential adverbials Sentence adverbials, such as unfortunately, sadly and allegedly, also “contribute to interpretation by affecting the process of context selection” (Rouchota 1998: 122). In this subsection, focusing on attitudinal sentential adverbials, such as surprisingly and sadly, I will discuss the similarities and differences between them and the interjectional DM oh. Ifantidou regards attitudinal adverbials as expressions “which do not name a speech act but indicate the speaker’s attitude to the statement she makes” (1993: 70). Consider the following examples (ibid.): (31) a. Unfortunately, Mary has missed the deadline. b. Sadly, Paul’s car was stolen. According to Ifantidou, unfortunately and sadly in (31) are used to add the speakers’ attitudes to the propositions expressed by the host utterances. She analyzes these attitudinal sentential adverbials in the framework of relevance theory, claiming that they encode conceptual meanings to contribute to the construction of higher-level explicatures, although they are non-truth-conditional.7 Now, compare these attitudinal sentential adverbials with the DM oh in (32): (32) Oh, Paul’s car was stolen. 7 See Wilson and Sperber (1993: 5) for higher-level explicature. 137 In (32), oh occurs sentence-initially, as the attitudinal adverbials in (31) do. It seems that there are several things shared by attitudinal sentential adverbials and the DM oh. First, neither these attitudinal sentential adverbials nor the DM oh affect the truth of the host utterances. Second, both of them (at least in this case) indicate the propositional attitude which the speaker intends to convey. Third, they are usually separated with comma intonation from main utterances. There seem, however, to be some differences between them. First, attitudinal sentential adverbials have compositionality, a characteristic of conceptuality in relevance theory, but the DM oh does not. ‘Very unfortunately, Mary has missed the deadline’, and ‘So sadly, Paul’s car was stolen’ are acceptable. On the other hand, we do not say *‘So oh, Paul’s car was stolen’. The reason is that while attitudinal sentential adverbials are conceptual, the DM oh does not seem to be conceptual. Another difference can be found in the following examples: (33) a. Happily, John came back with his girlfriend. b. Oh, John came back with his girlfriend. (34) a. John happily came back with his girlfriend. b. John, oh, came back with his girlfriend. (35) a. John came back happily with his girlfriend. b. John came back, oh, with his girlfriend. (36) a. John came back with his girlfriend happily. b. John came back with his girlfriend, *oh. In (33), both happily and oh are acceptable and affect the interpretation of the following sentential utterances. In (34) and (35), both are also acceptable in the slots, but while the interpretations of (34a) and (35a) are almost the same, the interpretations of (34b) and (35b) are clearly different each other. Happily in (34a) and (35a) (maybe (36a)) is not regarded as attitudinal sentential adverbial any more but as a manner adverbial which modifies the agent’s actions. It is not separated by comma intonation from the host utterances and affects the truth of the proposition expressed by the host utterances.8 By contrast, oh can appear in several spots in the host utterance except for at the end of sentence and always 8 In (34a), if happily is separated with comma intonation from the host-utterance, it becomes an attitudinal adverbial. 138 affects the elements following oh by showing the speaker’s current mental activity. Thus the overall interpretations of (34b), (34b) and (35b) depend on where oh appears in the host utterances. Third, a distinction can be seen in cognitive function. Ifantidou claims that although attitudinal adverbials are not truth-conditional, they encode conceptual meaning, guiding the hearer to the construction of higher-level explicatures (1993: 85-86) by embedding the proposition expressed by the host utterance under attitudinal descriptions. On the other hand, the DM oh does not seem to be the same, because it is not necessarily the proposition expressed by the host utterances that oh affects. Unlike attitudinal sentential adverbials, therefore, the DM oh does not always contribute to interpretation by constructing the higher-level explicatures. Rather it guides the hearers to interpret following parts of the utterances in the way intended by the speakers, by showing their current mental states or activities. 4.3 DM oh and Attitudinal Particles Wilson and Sperber (1993: 22) mention two interjection particles, the question particle eh and the dissociative particle huh, as attitudinal devices, and claim that they add to the echoic utterances (or attributed contents) the speaker’s attitudes, encoding constraints on the construction of higher-level explicatures (Wilson 2000: 433). Consider the following examples: (37) a. [Mary says to Peter] You’re leaving, eh? (c.f. Wilson 2000: 433) b. Mary is asking Peter if he is leaving or not. (38) a. [Mary says to Bill] Peter’s a genius, huh! (c.f. Wilson and Sperber 1993: 22) b. Mary doesn’t think that Peter’s a genius. (ibid.) In (37a), where Mary uses eh to indicate her questioning attitude to the proposition expressed by the host utterance, the hearer Peter can interpret Mary’s utterance by recovering the higher-level explicature like (37b). In (36a), where huh is used to represent Mary’s dissociative or skeptical attitude to the proposition expressed by the host-utterance, Bill can interpret the utterance by recovering the higher-level explicature like (38b). 139 These interjections, eh and huh, are very similar to oh, because they are all gestural interjections, they are not truth-conditional and convey the speaker’s mental states or activities, and can also increase the salience of the interpretation intended by the speaker. Eh and huh do, however, by constructing higher-level explicatures, while the DM oh does by showing the speaker’s current mental state or activity. Another difference can be seen in the nature of distribution. The slots where eh and huh occur are more or less determined (eh and huh almost always occur sentence-finally), while that is not the case with oh. As we have seen above, the distribution of oh is freer than eh and huh in that it is possible for it to appear in several spots, wherever the speaker’s mental activity is carried out and according to the spot where oh occurs, the hearer has a different interpretation. In the cases of eh and huh, even if they should occur sentence-initially, the over-all interpretation does not change. This fact suggests that oh is more closely connected than eh and huh (strictly speaking, at least, a kind of huh) with the moment that the speaker perceives the information or the assumption. 5. Interpreting oh-marked utterances The concerns of this section are the role of the DM oh and the pragmatic effects which can be achieved in different contexts by imposing a certain constraint based on the core ‘meaning’ of oh onto what is conveyed by the host utterances. So far, we have seen that the DM oh does not encode procedural meanings to put several constraints on the implicit contents of the following utterances like discourse connectives and adverbial discourse connectives do. Neither does it encode conceptual meanings to contribute to the construction of higher-level explicatures as do attitudinal sentence adverbials. Finally oh does not encode procedural meanings to contribute to the construction of the higher-level explicatures like attitudinal particles do. Rather it leads the hearer to interpret the host-utterance by showing the speaker’s current mental state or activity. I will also explore pragmatic effects that are yielded by using the DM oh and how they are yielded. In addition, closely examining several cases of the DM oh, I will show, in the framework of relevance theory, that these pragmatic effects are 140 not derived from oh itself but they are implicatures which can be recovered by putting a particular constraint based on oh’s core ‘meaning’ on the inferential phase of the hearer’s interpretation of host utterances. The DM oh is frequently used as an interpersonal device. Bolinger claims that oh has ‘sociability’ because oh “may represent a blend with the kissing gesture produced by mothers in their interaction with infants” (Bolinger 1989: 265). That seems to me true, because these vowel interjections are originally vocal gestures which may be accompanied by facial gestures and thus their meanings are “associated with the corresponding facial gestures” (Bolinger 1989: 284). Following Bolinger, I will provide a new analysis of pragmatic functions of the DM oh showing that oh is mainly used to implicate ‘politeness’ by showing the speaker’s mental activity described by its core ‘meaning’ ‘Something appreciable has just come into the speaker’s mind’ and affecting the interpretation of the host utterances. First, Bolinger claims that in (11) and (12) (1989: 270, repeated here for convenience) ohs indicate ‘enhancement’, but it seems to me that ‘enhancement’ is also derived from what is said by the host utterances. Consider: (11) June’s getting divorced. ? Oh how awful! When? (12) Is she good-looking? ? Oh beautiful! (12’) Is she good-looking? ? Oh I didn’t think so. Oh in (12’) does not clearly implicate ‘enhancement’. This fact suggests that the implication of ‘enhancement’ is not only derived from oh itself but from the overall interpretation of the oh-attached utterance. In my view, oh is simply used to mark that something appreciable has just come into the speaker’s mind. In (11), by showing that the speaker appreciates the news of June’s getting divorced and making a comment and asking a question to get further information, the hearer may interpret it as recovering an ‘enhancement’ implicature. In (12), by showing that the speaker appreciates the question and giving a stronger comment ‘beautiful’, he or she implicates ‘enhancement’. Bolinger claims that in (13) (repeated here for convenience) and (39) (1989: 271) oh implicates the speaker’s ‘enthusiasm’. Consider: 141 (13) I’ve just heard a juicy story. ? Oh tell me! (39) We’ve been thinking of taking a cruise this summer. ? Oh let’s! In (13), oh is used to represent the speaker’s strong curiosity about the information in the preceding utterance. In (39), oh shows the speaker’s interest in the idea expressed by the preceding utterance. In (13) and (39), the fact that each utterance would also be understood even if there were no oh, suggests that oh is mainly used by the speakers to implicate ‘friendliness’ or ‘politeness’. Now consider: (13’) I’ve just heard a juicy story. ? Oh don’t. You’ve been fooled again. Oh in (13’) does not clearly implicate ‘enthusiasm’ any more. This shows that it is not oh that indicates ‘enthusiasm’. It seems to me that oh is simply used to show that something appreciable has just come into the speaker’s mind. Using oh before the sentential utterances conveying the speaker’s positive attitude and showing that the speaker appreciates the information that has just come into the speaker’s mind is very useful in conveying the speaker’s ‘curiosity’ or ‘interest’ toward the information received. Consider the following example: (40) Erica: So, Harry…What do you do? Harry: I’m one of the owners of a record company among other things… Erica: Oh, really? Which record company? (The Movie: Something’s Gotta Give) In (40), by indicating that what Harry said is appreciable to her, Erica is showing her interest in his occupation and trying to ask a further question nicely. ‘Impetuousness’ or ‘impatience’ could result if the hearer felt that the speaker received the information but has not enjoyed or understood it yet. Showing attention to other people’s information or opinions or conveying strong ‘interest’ and ‘enthusiasm’ to others is very helpful in enhancing on-going conversation and maintaining good social relationships. Bolinger claims that in (41), (14) and (15) (repeated here for convenience), oh implicates ‘sociability’. I agree with him totally, because if there were no ohs in these examples, the utterances would sound unfriendly or blunt. (41) What did you think of the party? ? Oh I loved it! (14) I plan to take Nancy to the party. ? Oh you do? That’s great. 142 (15) I plan to take Nancy to the party. ? Oh you do? I wonder if that’s wise. In my view, however, the ‘sociability’ is derived not only from oh itself but also from what is communicated by the host utterances. (41) is very similar to ‘enhancement’ and ‘enthusiasm’ cases. Here, the speaker intends to express ‘sociability’ by showing that he or she appreciates the sudden question and giving her positive reply. In (14) the speaker implicates ‘sociability’ by showing that he or she appreciates the information and giving a positive reply. In (15), where the speaker makes a negative comment, the speaker still intends to convey ‘sociability’. That is probably because the speaker shows appreciation of the unexpected plan before he or she suggests disagreement. In my opinion, by putting oh before making a comment on someone’s opinion or question and by showing that what has just come into mind is appreciable, we guide the hearers to interpret our opinion in a positive way. Consider the following example: (42) Mother: I thought he was never nice to you. Mia: Oh, well…I don’t know. He is now. I just kinda hope that if he kisses me…my foot pops. (The Movie: Princes Diary) In (42), by indicating that her mother’s opinion is appreciable, Mia is leading her mother to interpret her thought in a positive way. Thus showing that we appreciate another’s thoughts, whether or not we agree with them, is very effective in our social lives. Bolinger claims that oh ‘mitigates’ the confrontational implication derived from a command as in (16), declination as in (17) and disagreement as in (18). Consider the following examples (Bolinger 1989: 278, repeated here for convenience): (16) Oh just be quiet for a minute! (17) Oh that’s too much! Half is enough! (18) It was all Mary’s fault. ? Oh I don’t know. Don’t you think she did the best she could? I agree with his view that the confrontational implications yielded by the following utterances are softened in these examples. Without ohs these utterances would be taken as very blunt and impolite. It does not seem to me, however, that ‘mitigation’ is one component of oh. In my view, in (16), (17) and (18) ohs are simply used to indicate that something appreciable has come into the speaker’s 143 mind and the core ‘meaning’ affects the interpretation of the host utterances. In (16) by showing that the speaker appreciates the noisy situation which he or she has just been brought into or the command which has just come into his or her mind and were neither aforethought nor designed deliberately, the possible negative implication derived from command is mitigated. In (17) by showing that the speaker appreciates the kind offer, he or she intends to mitigate the negative implication derived by declination. In (18), by showing that the speaker appreciates the opinion although it is an unexpected one to him or her, he or she intends to mitigate the possible negative implication derived from disagreement. Consider the following examples: (43) Harry: So I’m gonna venture out there and go for a short walk to climb stairs and stuff like that. You wouldn’t want to join me? Erica: That means you don’t want me to? Harry: No, I was asking if you wanted to. Erica: Oh, yeah, well, I would but… I don’t usually like to break once I’ve started writing so I probably shouldn’t, I mean, not that a little fresh air would hurt but. . . (The Movie: Something’s Gotta Give) In (43), by showing her appreciation to his unexpected offer she is mitigating the negative implication which might be yielded by the declination. Also, in (42), Mia is trying to mitigate the negative implication derived from her disagreement by showing her appreciation to her mother’s opinion with oh. Bolinger claims that ‘trivialization’ or ‘lightness’ is also derived from oh in the following context. (44) What did you see in it? - Oh just the ordinary things. (Bolinger 1989: 277) This implication is also derived from oh’s core ‘meaning’ and the contents of the following utterances which reflect trivial things. In (44), by using oh the speaker seems to decrease the importance of something in the interlocutor’s mind. That is, the speaker intends to convey, by showing that the question has just come into the mind and is the last thing to be expected to ask although she appreciates the question, that the speaker does not think the question so serious and thus ‘trivialization’ occurs. 144 Oh is very often used as a politeness device, because if there were no ohs in most examples above, the sentential utterances would sound blunt, impolite, and/or unfriendly. In my view oh may contribute ‘politeness’ in two ways. When oh is mainly used in a response to the hearer’s utterances (or to something about the hearer), oh contributes to ‘politeness’ mainly by showing that the information which has just come into the speaker’s mind is very appreciable. Consider: (45) A: I have a glass of sherry. B: Oh, that’s nice of you. (c.f. Aijmer 1987: 80) (46) Miranda: Well, I have two girls. Daniel: Oh, two precious gems. No doubt, the jewel of your eyes. Miranda: And one boy. Daniel: Oh, the little prince. How wonderful! Miranda: I must tell you, there would be a little light cooking required. Daniel: Oh, I don’t mind that, dear. I’d love some heavy cooking. [. . .] Miranda: No. Um, would you mind coming on an interview, say, Monday night at seven thirty? Daniel: Oh, I’d love to, dear. Miranda: Wonderful. I’m at two-six-four-oh Steiner Street. Daniel: Steiner. Oh, how lovely. (The Movie: Mrs. Doubtfire) In (45), oh represents the speaker’s ‘gratitude’ or ‘appreciation’ for the offer.9 (46) is a conversation between Miranda (a working mother looking for a nanny to take care of her children) and her ex-husband Daniel (applying for the job to meet his children), wherein oh is used by Daniel, who wants to get the job as nanny at Miranda’s house, to indicate that he appreciates her information and wants to make a good impression on her. Especially when the speaker’s opinion disagrees with the other person’s as in (42) or the speaker declines the other person’s kind offer as in (43), oh plays a significant role in maintaining a good social relationship, because the speaker can convey ‘politeness’ by showing the appreciation first and softening the possible negative implications derived from these behaviors. When oh is used before utterances like requests, which might yield negative implication, oh contributes to ‘politeness’ mainly by showing that these acts are 9 Aijmer (1987: 80) claims that in (55) oh is used to mark ‘politeness’ by indicating that “the offer was unexpected and that one is pleasantly surprised”. 145 what the speaker has just become aware of, but not what has stayed in the speaker’s mind. Consider the next examples: (47) Mia: Oh, well, I have to go. (The Movie: Princess Diary) (48) Marin: Oh (Har), by the way, no smoking in the house. My mom doesn’t allow it. (The Movie: Something’s Gotta Give) By uttering oh, the speakers show that the impolite behavior, such as leaving as in (47) and a request as in (48), which are important to the speakers, have just come into the speakers’ minds and these thoughts, which might be accompanied by negative implications, have neither always been in their minds nor are designed deliberately to make trouble and thus these negative implications do not deserve consideration. In other words, using oh before a request or troublesome utterance and leading the hearer’s attention into the speaker’s feeling that ‘Something appreciable has just come into the speaker’s mind’ is useful for the speaker to guide the hearer to interpret them in an easier way. My analysis of the DM oh’s function might also account for Schiffrin’s observation of ohs in situations when a speaker repairs, clarifies, asks, and answers. These situations might often cause the interlocutors trouble or might yield negative implications by demanding of them more effort. In such cases, oh might be a very convenient device for the speakers to deviate from involvement in such negative implications, because oh indicates that they appreciate what is conveyed by the preceding utterances or they have just become aware of something which conveyed by the following utterances, which could contribute to reducing the hearer’s mental burden. In my opinion, by using oh especially before giving negative comments to someone’s opinion, asking troublesome questions or repairing and by showing that what has just come into the speaker’s minds is appreciable to him or her, the speaker can guide the hearer to interpret his or her opinion in a positive way. Showing that someone appreciates another’s thought is very useful in our social lives. I conclude that in every case, the role of the DM oh is to guide the hearer in interpreting the utterance that follows as one in which the speaker has just become 146 aware of something appreciable, and its various implicatures (e.g. ‘sociability’, ‘intensification’, ‘enthusiasm’ and ‘trivialization’) are pragmatically derived from the context, what is conveyed by the following utterance and oh’s core ‘meaning’. 6. Conclusion I have examined the role of the DM oh and the pragmatic effects recovered according to contexts by using the DM oh and showing the speaker’s mental state or activity. The DM oh’s function is to lead the hearer to interpret the utterance or the elements which follow in the way intended by the speaker by showing the speaker’s mental state or activity represented by its core ‘meaning’. The relevance of the DM oh lies not only in setting the context for interpreting the utterance which follows (or elements following oh) and in reducing the hearer’s processing effort but also in yielding contextual effects like ‘politeness’. The interjectional DM oh is strikingly distinct from lexical DMs (e.g. so, after all). First, while the relevance of lexical DMs lies in their way of clarifying the semantic relationship between two segments, that of the DM oh lies in its way of setting the context related to the speaker’s mental state or activity for interpretation of the utterances which follow. Second, while the functions of lexical DMs can be explained as encoding procedural meanings to put certain constraints on implicatures, that of the DM oh cannot be explained as linguistically semantic encoding. Third, unlike lexical DMs, the DM oh enables speakers to show evidence of their mental state or activity very vividly, naturally and spontaneously, because it is originally a spontaneous vocal reaction toward stimuli. That allows the speakers to convey implicatures in more casual and natural ways. Oh is also used for interpersonal purposes like ‘politeness’: to establish or maintain good social relations with others. The DM oh enables speakers to convey such implicatures in a more vivid, more natural and more casual way, because oh has a natural origin. These natural pragmatic effects can also be yielded in other ways, such as facial expressions or tone of voice. In relevance theory, it is 147 assumed that the speakers aiming at optimal relevance should not only try to give the hearers adequate contextual effects, but also try to give these effects with the minimal necessary processing effort (see Sperber and Wilson 1986/1995; Wilson and Sperber 1993). Therefore it is not surprising that a rational speaker uses the DM oh to direct the hearer to interpret the utterances which follow in the way intended by the speaker by directly showing mental states or activities. References Aijmer, K. 1987. Oh and Ah in English Conversation. In: W. Meijs (ed.) Corpus Linguistics and Beyond, 61-86. Amsterdam: Rodopi. Ameka, F. 1992. Interjections: The universal yet neglected part of speech. Journal of Pragmatics 18, 101-118. Blakemore, D. 1987. Semantic constraints on relevance. Oxford: Blackwell. Blakemore, D. 2002. Relevance and Linguistic meaning: The Semantics and Pragmatics of Discourse Markers. Cambridge: Cambridge University Press. Bolinger, D. 1989. Intonation and its Uses: Melody and Grammar in Discourse. Stanford: Stanford University Press. Carlson, L. 1984. 'Well' in dialogue games: A discourse analysis of the interjection 'well' in idealized conversation. Amsterdam: Benjamins. Fischer, K. 1998. Validating semantic analyses of discourse particles. Journal of Pragmatics 29, 111-127. Ifantidou-Trouki, E. 1993. Sentential adverbs and relevance. Lingua 90, 69-90. Jucker, A. H. 1993. The discourse marker well: A relevance-theoretical account. Journal of Pragmatics 19, 435-452. Nishikawa, M. 2002a. Interpretive Use of Interjections. Annual Reports of Graduate School of Human Culture 17, 72-85. Nara Women’s University. Nishikawa, M. 2002b. Discourse marker Oh. Eigo goho bumpo kenkyu 9, 931- 952. Tokyo: Kaitaku-sha. Nishikawa, M. 2003. Phatic Aspects of English Interjections. Annual Reports of Graduate School of Human Culture 18, 79-92. Nara Women’s University. Nishikawa, M. 2007a A Cognitive Approach of English Interjections, Unpublished Ph.D. dissertation, Nara Women’s University. Nishikawa, M. 2007b. Politeness in the Usage of the Interjection oh. Setsunan Journal of English Education 1, 77-87. Nishikawa, M. 2007c. Interpreting Interjections: A Perspective of Relevance Theory. 22 Essays in English Studies: Language, Literature and Education, 65-88. Tokyo: Shohakusha. Rouchota, V. 1998. Procedural meaning and parenthetical discourse markers. In A. Jucker and Y. Ziv eds. Discourse markers: descriptions theory. Amsterdam; John Benjamins, 97-126. Schiffrin, D. 1987. Discourse markers. Cambridge: Cambridge University Press. 148 Schourup, L. 1985. Common discourse particles in English conversation. New York/London: Garland Publishing. Schourup, L. 1999a. Discourse Markers. Lingua 107, 227-265. Sperber, D. and Wilson, D. 1986/1995. Relevance: Communication and Cognition. Oxford: Blackwell. Wharton, T. 2003. Interjections, language, and the ‘showing’/’saying’ continuum. Pragmatics & Cognition 11(1): 39-91. Wierzbicka, A. 1992. The semantics of interjections. Journal of Pragmatics 18, 159-192. Wilkins, D. 1992. Interjections as deictics. Journal of Pragmatics 18, 119-158. Wilson, D. 2000. Metarepresentation in linguistic communication. In D. Sperber (ed.), Metarepresentations: A Multidisciplinary Perspective. Oxford: Oxford University Press, 411-448. Wilson, D. and Sperber, D. 1993. Linguistic form and relevance. Lingua 90(1): 1- 25.