CHAPTER 5 Language and gender in Romanian: The qualitative analysis

This chapter applies the microanalysis proposed in Chapter 4 to excerpts taken from naturallyoccurring conversations among Romanian women living in Constanta in an attempt to substantiate the intuitive conviction that when they engage in conversation Romanian women put more emphasis on serving the need for positive face i.e. honouring their interlocutors need for involvement. Within the unifying theoretical framework we proposed in the previous chapter the two levels of analysis, the micro level (i.e. linguistic variables) and the macro level (i.e. sociological variables), are kept apart as two independent systems in their own right but are nevertheless mediated by such variables as those supplied by Brown and Levinsons pragmatic theory of politeness: face, power, social distance, the rank of imposition and hierarchiacally organized politeness mechanisms. Our analysis of gender-related linguistic variation will proceed in a bidirectional fashion: from the conversational strategies down to their linguistic embodiments (the micro level of analysis) and also from those strategies up to the sociological variables (the macro level) that the choice of these strategies signals. 5.1. Investigating Romanian conversational discourse Despite claims to universality, Brown and Levison do allow for a certain degree of cross-cultural variation, which they call ethos and define as the affective quality of interaction characteristic of members of a society (1987: 243). Their distinction between positive and negative politeness strategies, whether mixed and multifunctional or not, and the subsequent differentiation between positive and negative politeness societies can shed considerable light on differences between cultures, on the one hand, and various social groups within a given culture, on the other, and provide a secure basis for a comparative investigation. Complex stratified societies exhibit both kinds of politeness. However, various social groups within such societies may show differing preferences for one ethos or another. Upper classes, for instance, may have a negative politeness ethos and lower classes a positive politeness ethos (Brown and Levinson (1987: 20). Similarly, it is widely reported that women tend to value positive politeness strategies and informality more than men. On this issue Brown and Levinson (1987: 246) argue that: this distinction between positive and negative-politeness emphases not only marks class from class in hierarchical societies, but also marks different kinds of social roles from one another. Thus, we suspect that, in most cultures, women among women have a tendency to use more elaborated positive-politeness strategies than do men among men (our emphasis). Although we can distinguish societies according to the ethos predominant in daily conversational discourse, we are not implying here either that societies as a whole or that various social groups within a complex stratified society can clearly and exclusively be categorized as being either positively or negatively polite. Rather they can be categorized as relatively more positive politeness oriented or relatively more negative politeness oriented, according to the type of ethos which is given more play. 107

In what follows this book aims to fill a gap in our understanding of spoken Romanian by investigating the speaking practices of Romanian men and women when engaging in all-male, allfemale and mixed conversations and by identifying their politeness orientations. Our informants belong in three age groups: teenagers (thirteen- to nineteen-year-old secondary- and high-school pupils), young adults (twenty to thirty-year-old university students and graduates) and adults (thirtyfive to sixty years old). To our knowledge, gender-related conversational behaviour has not been the subject of linguistic investigation in Romanian up to this point, and information concerning informal conversational discourse among Romanian speakers is scarce. In this context, it is worth mentioning one empirical study dealing with the structures and strategies of conversational discourse in spoken Romanian, namely Ionescu-Ruxndoius Conversaia: structuri i strategii: sugestii pentru o pragmatic a romnei vorbite (1999). The Romanian linguist based her analysis on a corpus of dialectal texts published in the last three decades in various dialectal atlases and illustrating the dialects spoken in the Southern rural part of Romania since at that time there was no audio recorded corpus of naturally occurring conversations for the Romanian language. Her analysis established a preference for elaborate negative politeness strategies in interaction among in-group and out-group members or among in-group members of different social status (Ionescu-Ruxandoiu 1999:110). As far as positive politeness is concerned, several strategies are particularly well illustrated. Among them, she mentioned the following: expressing agreement and avoiding disagreement, a consistent use in-group identity markers and intensifying interest to the addressee (Ionescu-Ruxandoiu 1999: 114-15). However, the interactional strategies identified and analysed by Ionescu-Ruxandoiu are not correlated with gender affiliation. Moreover, as she pointed out they may differ from those favoured by members of urban communities (Ionescu-Ruxandoiu 1999:116-7). 5.1.1. Research questions In the preceding chapters we have set the scene for an analysis of gender-related linguistic variation in Romanian. We established that linguistic stereotypes play an essential role as guides to where to seek confirming or disconfirming evidence for gender-related linguistic differences. We then reviewed research on gender-related linguistic variation in English and established that this type of variation does make a difference and that it generally revolves around the communal-agentic dichotomy. Our next stage in the analysis of gender differences in language use was to examine some of the sources of gender-related linguistic variation: gender ideologies and gender-related patterns of socialisation. We highlighted the role of gender ideologies in gender identity construction in English and Romanian societies and provided empirical evidence that confirm the hypothesis of similar practices of socialisation stemming from similar gender ideologies. According to these ideologies, men are assigned social roles associated with power and status, while women are assigned subordinate roles. Children are socialised first within the family and then within institutionalised settings so as to conform to these differential gender roles that society assigns to them. One way of conforming to these gender roles is to employ gender-appropriate conversational styles. In light of the evidence provided so far, we suggest the hypothesis that gender ideologies are manifest in Romanian at the level of conversational strategies, as they are in English. Moreover, we further hypothesize a female conversational style primarily based on expressing solidarity with interlocutors. Thus, many of the interactional strategies for which Romanian women show a preference are positive politeness oriented, being thus ways of expressing similarity between the self and others. 108

5.1.2. Database and methodology Our research is based primarily on the analysis of original data from recorded face-to-face interactions. The extracts are grouped in two sets of audio-recorded data gathered in Constanta and in Bucharest. Both sets are uncontrolled samples of face-to-face naturally occurring interaction. The material we collected amounts to ten hours of naturally occurring conversation. It should be emphasized that the analysis presented in this study is based primarily on a study of our own corpus, henceforth referred to as the Constanta corpus. The Constanta set is part of our own research project comprising ten hours of both mixed and same-sex interactions gathered over the last five years with a view to exploring the speaking practices of Romanian women and men in both formal and informal settings. The participants include twenty-four individuals 1 (twenty females and four males), whose ages ranged from thirteen to sixty-four (including ten adolescents, eight in their twenties, two in their thirties, three in their forties, and one in her sixties (Hudson 1980). The Constanta corpus consists of ten hours of both spontaneously occurring conversations with friends recorded in settings chosen by participants themselves and spontaneous interaction in offices and various service encounters. The primary database was collected with one Panasonic MiniCassette Recorder (RQ-L30). Our primary concern in gathering the data on informal conversation has been to avoid the constraints inherent in a one-to-one interview where the interviewer is present. Therefore we have chosen not to be present while the informants were engaged in conversation hoping that the constraints stemming from the informants knowledge that they are being observed can be alleviated (Labov 1972b). We asked some of the participants to pair up with their same-sex best friend and talk about stuff in a familiar setting; the topic for discussion, however, was up to our informants. The choice to group them in dyads rather than in triads or in even larger groups was made with the view to avoiding the technical problem of recording each speaker on a different track. On the other hand, we have chosen to interview best friends because we hold the view that the closest we can come to getting natural speech in an interview situation is by interviewing groups of peer. This type of interview is the context most conducive to obtaining casual speech since the normal patterns of group interaction can direct attention away from the tape recorder (Spolsky 1998). All those involved in this project provided information on their social background and granted permission for the data to be used for linguistic analysis (Spolsky 1998). Throughout the process, participants were free to edit and delete material as they wished (Holmes 2000). By handing over control of the recording process in this way, we managed to develop a relationship with our informants based on mutual trust which, over a period of time, made it easy for our participants to ignore the recording equipment. As a result, in return for guarantees of anonymity and confidentiality, our informants trusted us with a wide range of fascinating material. All names are fictionalised to protect participants identity. We are satisfied that the material represents natural conversation and that there was no undue awareness of the recorder. The tapes sound natural. Some of our informants reported that they soon began to ignore the tape recorder. Moreover, they were apologetic about the material calling it trivial and uninteresting, just the ordinary affaires of every day life. The Bucharest set comes from the corpus of spoken Romanian established at the Romanian Language Department, Faculty of Letters, University of Bucharest. The recordings were made in a variety of settings including private homes, shops, offices and other public places in Bucharest, Ploiesti and Braila (Ionescu-Ruxandoiu 2002). Throughout this work, in addition to examples from our
1

Gennerally, one should aim at a minimum of five informants for each social variable (Hudson 1980).

109

primary database, we will also analyse examples from the Bucharest corpus to illustrate relevant points, and to show that the phenomena under discussion are not restricted to our data set, on the one hand, and to supply evidence of linguistic behaviour not coverd by the Constanta corpus, on the other hand. We can profit from the advantages of one method while overcoming the limitations of another, and have, therefore, attempted to collect data from a variety of sources using various techniques for gathering relevant data. Thus apart from naturally occurring conversations originating in two different corpora, we have also made use of questionnaires to obtain statistically analyzable data on attitudes (Chapter 1), discussions with informants, reports from works on sociology (Hudson1980). The recordings were transcribed using the transcription conventions employed in conversation analysis (see Appendix) including thus sufficient detail for a fine-grained analysis (Ochs, Schegloff and Thompson 1996). As a result, this corpus can be used to make observations of a quantitative nature (Chapter 6). Finally, following the ethnographic approach, we carefully wrote down occurrences of situations and patterns we found interesting wherever possible (Wardhaugh 1992). 5.2. Intra-group similarities: interactional strategies associated with positive politeness in female conversational style Positive politeness redress has been shown to be equivalent to everyday intimate language behaviour plus an element of exaggeration. This element of insincerity in exaggerated expressions of approval and interest carries the implication that the speaker sincerely wants the addressees positive face to be satisfied. Being a kind of metaphorical extension of intimacy, as they imply common ground or sharing of wants even between strangers, positive politeness techniques are used as a kind of social accelerator. By using them the speakers indicate that they want to come closer to the addressee, that they and the addressee share similar views, opinions and values (Brown and Levinson 1987). The interactional strategies for which our female informants manifest an interest can be grouped under the following headings: claim common ground with the addressee and fulfil the addressees want (for some X), headings that label two broad mechanisms of positive politeness (Brown and Levinson 1987). Their conversational style can thus be characterised, using a label supplied by Tannen, as high-involvement2. This is not to say that they do not employ other positive or negative politeness strategies as well, but rather that they have a strong preference for using those positive politeness strategies that enable them to assert common opinion or shared knowledge and to show their support for their conversationalists partners. In what follows the strategies identified are illustrated by excerpts from the corpus we gathered in Constanta as well as by excerpts from the Bucharest corpus 3 (gathered by researchers from the Department of Romanian Language, School of Letters, University of Bucharest) which we use as a control sample to see if the linguistic variation is caused by social or cultural factors or if it happens by chance. All the excerpts are analysed within the theoretical framework proposed in Chapter 3. On this view, the relationaship between aspects of the micro level, i.e. conversational strategies, and aspects of the macro level of analysis, i.e. gender is mediated by the following variables provided by Brown and Levinson s pragmatic theory of politeness: face, power, social distance, the rank of imposition and hierarchically organized politeness mechanisms. 5.2.1. Claim common ground with Addressee
2

This comes in sharp contrast with the high-considerateness style described by Tannen (1989) as being characteristic of those speakers who put more emphasis on serving the need for negative face i.e. honouring others need not to be imposed on, in other words their need for independence. 3 A selection of this corpus has been published in Ionescu-Ruxandoiu (2002).

110

The first mechanism involves speakers claiming common ground with the addressee by indicating that both the speaker and the addressee belong to some set of persons who share specific wants, such as goals, values, opinions, in-group membership etc. The ways of making this claim are: (i) (ii) (iii) the speaker may convey that he finds some want (goal, or desired object) of the addressee admirable or interesting; the speaker may claim in-group membership with the addressee; the speaker may claim common opinions/attitudes/point of view with the addressee.

5.2.1.1. Speaker conveys that he finds some want of Addressee admirable or interesting 5.2.1.1.1. Speaker notices and attends to Addressees interests, wants, needs, goods The use of compliments In general this output suggests that S should take notice of those aspects of Hs condition which looks as though H would want S to notice and admire (Brown and Levinson 1987). The analysis of our data reveals that this is achieved by the use of such expressions of solidarity and appreciation as compliments4. Compliments are essentially formulaic speech acts drawing on a limited number of lexical items and a narrow range of syntactic patterns. The few syntactic patterns identified can be arranged along a continuum stretching from the minimal pattern ADJ (NP) (e.g. Great shoes!) to the rhetorical pattern illustrated by What (a) (ADJ) NP! (e.g. What great kids!) and Isnt NP ADJ! (e.g. Isnt this food wonderful!). The former reduces the syntactic pattern to its minimum elements; by contrast, the latter is a syntactically marked formula, involving exclamatory word order and intonation. The other few syntactic patterns that can be identified lie somewhere along the continuum between these positions according to the high density of various intensifying devices such as exaggerated intonation, emphatic stress and/or extreme case formulations5. The rhetorical pattern can be regarded as emphatic and as increasing the force of the speech act, while the minimal pattern tends to reduce the force of the compliment and therefore it could be regarded as attenuating the impact of the compliment. Thus choosing a rhetorical pattern or a pattern rich in intensifying devices, as in the examples below, stresses its addressee- or interaction-oriented nature.

TEENAGERS

Research on the distribution of compliments and compliment responses across gender groups carried out in various speech communities shows that the overall pattern appears to be consistent in a number of English-speaking communites: women give and receive more compliments than men do (Wolfson 1983; Herbert 1990; Holmes 1986; Holmes 1995). This pattern is also confirmed by research on compliments between Polish speakers (Lewandowska-Tomaszczyk 1989). 5 Evidence from various English-speaking communities suggests that women used the rhetorical pattern significantly more often than men, whereas men preferred the minimal pattern. Womens compliments tend to be expressed with linguistically stronger forms than mens. The stronger form I love X (as compared to I like X) and a consistent use intensifiers such as really, very, particularly, seem to be preffered by women (Holmes 1986; Herbert 1990; Jonhson 1992).

111

Excerpt 1 is taken from a conversation between two high-school students who are planning a night out on the town.
Excerpt 1 (Constana corpus) 1 Irina: da ce-s tia? pantofii ti noi? but what are these? your new shoes? 2 snt su:per adevrai6 they are so cool 3 Raluca: da:::: yeah 4 Irina: snt suPErbi they are awesome

ADULTS Excerpts 1 to 3 echo the pattern that emerged from empirical studies carried out in English-speaking communities and supports a view of womens style as more interpersonal, affective and interactionoriented compared to the impersonal, instrumental and content-oriented style more typical of male interaction (Preisler 1986; Tannen 1990).
Excerpt 2 (Constana corpus) 1 Maria: va:i da tia snt su:per mito wow, but these are so cool 2 Iulia: i plac? do you like them? 4 Maria: mi plac la nebunie Im crazy about them. 5 vreau i io s-mi iau I wanna buy some too. 6 Iulia: da yeah 7 Maria: vreau i io s-mi iau I wanna buy some too. 8 vai ce mito snt wow, they are so cool Excerpt 3 (Bucharest corpus) A: o::: ce drgu brichet ai oh, what a nice lighter youve got

5.2.1.1.2. Speaker exaggerates (interest, approval, sympathy with Addressee) Emphatic stress, extreme case formulations
6

Items under discussion are given in bold type.

112

Exaggerating interest/approval/sympathy with the addressee can often be achieved by using exaggerated intonation, emphatic stress as well as extreme case formulations, that is descriptions or assessments that deploy extreme expressions such as the superlative forms of adjectives ( best, most, biggest, least), semantically extreme adjectives (total, absolute, marvellous, extraordinary, whole, etc.), various adverbs (always, never, perfectly, completely, absolutely, totally, etc.) and phrases (as good as it gets, brand new, etc. )7 The following examples come from female conversational discourse and are used to intensify the speakers approval with or admiration/appreciation of the addressee or to make the story more interesting or to signal ones interest and involvement in the conversation. As a result these strategies are instrumental in satisfying the addressees positive face. TEENAGERS Excerpts 4 to 15 (Constana corpus) (4)
L: adic s cO:pii NIciodat you mean cheat? Never!

(5)
L: i-a zis c-mi mai zice pedepse dac e s mai jucm tie superbeton pedepse and he said that he would tell me more game tasks if we were to play again he knows some super cool game tasks

(6)
R: chestia aia tii cnd vii vin acas dac el este rupt de beat that thing, you know, when you I come home if he is dead drunk

(7)
R : e o fat extrem de bun shes an extremely nice girl

(8)
D : i dai seama c l-a deranjat enorm you realize that it upset him enormously

(9)
R: exclusiv cu cear
7

For an extensive discussion of extreme case formulations in English, see Pomerantz (1986) and Edwards (2000) among others.

113

wax only

(10)
I: snt superbi they are superb

(11)
R: am un zmbet oribil I have a horrible smile

(12)
R : bine e o relaie extraordinar de frumoas i foarte serioas well its a extremely nice relationship and very serious

(13)
I : n-o s mai vezi un mesaj de la mine nici ntr-o mie de ani youll never get an SMS from me, not in a thousand years.

(14)
I : NU-i spui pe tonu sta you dont speak to him like this

(15)
R :io N-am s m ro :g n VIa ::a mea I will not beg anyone not ever

ADULTS Excerpts 16 to 23 (Constana corpus) (16)

I: auzi da nu-mi pui i mie puin suc c-mi faci poft teribil hey, wont you give me some juice youve made my mouth water

(17)
M: bi da snt nite ciutani de super detepi acolo m mori the kids there are sm super smart; theyre killing you

(18) 114

M: i: nu da a fost a fost atmosfera absolut relaxat tii and, well, the atmosphere was totally relaxing, you know

(19)
I: io mi voi face planu de lecie la superlativ normal Im gonna draw up a superlative lesson plan, of course

(20)
M: io p-tia am dat trei sute da m-neap ::nfiorto::r I paid three hundred for these but theyre itching me terribly

(22)
I: haide o li:ngur mcar s gu:ti come on, just a bite, to have a taste

(23)
1 I: erau un material splendi:d it was a gorgeous fabric 2 cnd m-am mbrcat va:i nepa:u nepau groa:znic but when I put them on, ay, they were itching horribly

5.2.1.1.3. Speaker intensifies interest to Addressee Historic present and constructed dialogue (i.e. directly quoted speech) Another way for the speaker to communicate to the addressee that he shares some of the latters wants is to intensify the interest of his own contributions (i.e. the speakers contributions) to the conversation by making a good story. This goal can be achieved by using the historic present8, a tense shift from past to present tense, constructed dialogue9 (i.e. directly quoted speech) rather than indirect reported speech and tag questions or expressions such as you know, see what I mean, isnt it that draw the addressee as a participant into the conversation (Cf. Brown and Levinson 1987). Research on language and gender in English-speaking communities has shown that women frequently use the historic present in narratives of past events and constructed dialogue (Coates 1993). Framing ideas as constructed dialogue rather than statements is a discourse strategy that is effective in conveying

This use of present tense in reporting events that occurred in the past is referred to by Tannen (1989) as the vivid present, presumably because it makes the events seem more real. 9 Following Tannen (1989) we will argue that one cannot, in any meaningful way, report speech. When speech uttered in one context is repeated in another, it is changed fundamentally even if reported accurately. First, in many cases what is labeled as direct quotation was never spoken in a form resembling the one constructed. Second, if dialogue is used to represent utterances that were indeed spoken by someone else, when an utterance is repeated by a current speaker it exists by virtue of the reporting context although its meaning continues to resonate with its original reported context. As Tannen has put it words have ceased to be those of the speaker to whom they are attributed, having been appropriated by the speaker who is repeating them. Tannens view is supported by recent research on reported speech in interaction which shed light on the subtle ways in which speakers comment on the utterances they report while simoultaneaously appearing to simply reproduce them. Buttny (1997), Holt (2000), to mention just a few researchers, have shown how in an environment where reported speech is claimed, and perhaps expected, to be a faithful rendition without interpretation by the speaker, speakers make use of prosodic devices, the design of the reported speech and the sequence in which it occurs to implicitly convey an attitude towards the reported utterances which quite often undermines what is reported. It is in this sense that directly quoted speech should be seen as constructed dialogue rather than reported speech.

115

information and creating invlovement. To support this claim consider the following examples taken from narratives recorded by participants in casual conversation with their friends. TEENAGERS Excerpt 24 comes from an informal friendly conversation between teenagers. Excerpt 24 (Constana corpus)
1 I: m-am plimbat azi cu ( ) ce mito hai s-i spun I went for a ride with ( ) today how cool! let me tell you 2 cnd am mers spre dacia la meditaie m-am ntlnit cu miki when I was going towards dacia for tutoring, I ran into miki 3 eram cu miki mergeam cu miki spre meditaie i la dacia ( ) a parcat tii maina I was with miki, going to tutoring and at Dacia ( ) parked the car, you know 4 a parcat maina i m cheam la el la cinci poi s fii aici fac eu he parked his car and called me I go you can be here at five 5 face da uite la cinci pot s fiu aicea da nu snt sigur d-mi un telefon he goes look I can be here at five but Im not sure, give me a call 6 bine ies io de la meditaie l sun p-sta bi poi s vii s m iei? well, the tutoring is over I call him hey can you come pick me up? 7 cic te-atept n fa la dacia de vreo zece minute hai c te-am pupat vino he says Ive been waiting for you in front of Dacia for about ten minutes, come on, kisses

In line 1 Irina introduces a news-update10 (m-am plimbat azi cu I went for a ride today with) whose verb is in a past tense which then she evaluates as being interesting and newsworthy by using the slang term mito (cool) which amounts to an extreme case formulation. In lines 3 to 5 she starts telling about her date using only past tenses (am mers, m-am ntlnit, eram, mergeam, a parcat) and in the middle of her reporting of past events she switches to present tenses and constructed dialogue in lines 5 to 8. Switching to present tenses in the narration of past events enables the speaker to pull the addressee in the middle of the events being narrated and to increase the immediacy of those events. ADULTS Constructed dialogue (i.e. directly quoted speech) has been shwon to serve a lot of functions. People often report their own thoughts as dialogue (Tannen 1989). Excerpt 25 represents constructed dialogue as inner speech. In this excerpt Maria starts telling how she managed her final lesson during practicum. At this point she mentions one important aspect in her story, namely time management.

Excerpt 25 (Constana corpus)

10

News-updates have been analysed as recognizably small talk topics (Drew and Chilton 2000).

116

1 Maria: tii c data trecut am am trecut peste deci cum s spun eu am you know, that last time I skipped so how shall I put, it I didnt 2 nu mi-am calculat timpu ca lumea tii I didnt estimate my time properly you know 3 Iulia: mhm mhm 4 Maria: i am intrat pe timpu ( ) i acuma acuma zic and I took ( )s time and so, I say to myself 5 hai s-mi pun mai puine activiti tii I should choose fewer activities, you know 6 ca s m ncadrez n timp s nu m trezesc iar c n-am timp to finish on time and not end up running out of time again 7 Iulia: mhm mhm 8 Maria: i mi-am pus prea [puine and I chose too few 9 Iulia: [aoleu i ce? i s-au terminat prea repede? ay ! and what happened? they were over too soon? 10 Maria: mi s-au terminat bine cu vreo dou trei minute nainte tii they were over in time, two or three minutes before the break, you know 11 Iulia: mhm mhm

In lines 1-2 and 4 she mentions an instance when she had problems with time management since she had not allocated enough time for the activities she intended to do: tii c data trecut am am trecut peste deci cum s spun eu am nu mi-am calculat timpu ca lumea tii i am intrat pe timpu ( ) (you know that last time I I skipped so how shall I put it I I didnt estimate my time properly you know). In line 3 Iulia confirms receipt of information by using a continuer (mhm) and then Maria moves on to telling about her final lesson in lines 4-6 when she chose fewer activities lest she should run out of time and in so doing she shifts from past to present tense in line 5: i acuma acuma zic (and now now I say) thus increasing the immediacy and interest of her story. The same effect is achieved by framing her thoughts as a dialogue with herself in lines 5-6: hai s-mi pun mai puine activiti ca s m ncadrez n timp s nu m trezesc iar c n-am timp (I should choose fewer activities to finish on time and not end up running out of time again). Excerpt 26 illustrates dialogue representing what was not said11. The speaker is expressing her dissatisfaction with the idea that she should teach a class she has not thought before. Excerpt 26 (Constana corpus)
1 Iulia: nu e posibil aa ceva i mine o s-i zic this is just not possible, and tomorrow Im gonna tell her 2 mi pare ru doamn da nu exist aa ceva cum s-mi dai mie- Im sorry madam, but this is just not possible how can you give me- 3 nu? spune i tu right? Dont you think so? 4 i o s vorbesc i cu doamna kaiter i i spun i ei nu? and Im gonna speak to mrs. kaiter as well, and Im gonna tell her too, right? 5 i cu asta basta da nu e normal
11

See Buttney (1997) for a comprehensive list of possible types of reported speech.

117

and thats that! its just not right

She represents explicitly in the form of a dialogue what she did not say but rather would rather say in line 2. There is no suggestion that the speaker actually said imi pare ru doamn da nu exist aa ceva cum s-mi dai mie- (Im sorry madam, but this is just not possible how can you give me - ) but simply that she felt in a way reflected by that utterance. It is clear in all these extracts that the lines of dialogue in the stories are not reported but rather constructed by the speakers, in the same way lines of fiction or drama are constructed and to the same effect. The use of the historic present and constructed dialogue (i.e. directly quoted speech) enable speakers to make their stories more interesting. Historic present pulls the addressee into the middle of the events being discussed, at least metaphorically, whereas constructed dialogue fires the individual imagination. The creation of voices triggers the imagination of a scene in which the characters speak in those voices. Such scenes occasion the imagination of alternative more or less familiar worlds (Tannen 1989). This increases both the speakers and the addressees involvement with the story and with each other. 5.2.1.2. Speaker claims in-group membership 5.2.1.2.1. Speaker uses in-group identity markers The speaker can claim common ground with the addressee by virtue of claiming that both he and the addressee belong in the same group. This can be achieved by using in-group identity markers. Brown and Levinson (1987) list amog in-group identity markers in-group usages of address forms, of language or dialect, of jargon or slang, and of ellipsis. As our analysis will show, we can supplement Brown and Levinsons list with what has beencalled collaboratively built sentences (Sacks 1992). Address terms: endearment forms and diminutives In English, address forms used to convey in-group membership include generic names and terms of address like mate, buddy, pal, honey, dear, luv, brother, sister, sweetheart, guys, fellas . In Romanian, an equivalent frequently used among adolescent females is fat (girl). Excerpt 27 (Constana corpus)
Maria: las ca ( now ( ) o data zic de ce n-ai venit fat astzi la coal ) once I said girl, why havent you come to school today

Diminutives and endearments have a similar function when it comes to claiming in-group solidarity. But whereas in English such endearment forms as duck, love and dear are not exclusively restricted to conversation among friends and may equally be encountered in service encounters, displaying the same function of expressing solidarity with the addressee, in Romanian their use is restricted to conversations among friends, as excerpts 28 to 36 show. TEENAGERS Excerpts 28 to 32 118

(Constana corpus) (29)

1 Irina: mama ce m-am enervat la faza asta god, I got so angry at this 2 Raluca: de ce drag? e adevru why dear? its the truth

(30)
Irina: ma:m raluca m n-ai mobil? i nu-i iei? man, raluca, you dont have a cell phone? and arent you getting one?

(31) 1 Mona: bi tii care-i faza you know what? 2 io nu tiu io trebuia s m fii schimbat pn-acuma
the thing is, I dont know, I should have changed by now

(32)
1 Laura: ct e ceasu? whats the time ? 2 Anca: doipe i doucinci twelve twenty-five 3 Laura: doipe doUAcinci? twelve TWENTY-five? 4 Anca: da fat yeah girl

ADULTS
Excerpts 33 to 36 (Constana corpus)

(33)
1 Iulia: unde-i culc? where is he going to put them up? 2 Maria: la taic-su la hotel At his dads hotel 3 Iulia: GRA:tis? For FREE? 4 Maria: of course c gratis of course for free 5 Iulia: HAI MA::: i taic-su e de acord? COME ON! And his father agrees? 6 Maria: nu e de acord he doesnt 7 i culc acolo gratis (i cu) mncare

119

he puts them up there for free, food included 8 Iulia: i mncarea tot gratis the food for free as well 9 Maria: i mincarea pi da m tre s le asigure mncare i the food too, well hes gotta provide them with food and

(34)
1 Iulia: p calculatoru lu lulu aa on Lulus computer right 2 Maria: bi da ciu:dat e mac-ul la de tot c [pn-i dai seama cum sGod, that Mac is totally weird it takes ages to figure out how to3 Iulia: [aoleu (unclear) ay 4 Iulia: bi calculatoru la zici c-i calculator de jucrie p cuvntu meu = God, that computer, I swear, youd say its a toy-computer=

(35)
1 Maria: da m io am neles chestia asta da tii care-i che::stia? yeah, I got this but you know what? 2 Iulia: mhm? mhm 3 Maria: io cre c voi n-ati neles I think you didnt get it 4 c fiecare s-i ia cte unu din alea sapte i la AIAs fac zece ex- exemple each should take one of those seven and supply ten examples for THAT 5 Iulia: na na na nu pi da ce pi oricum nu e nimica de creaie no, no, no anyway, theres nothing to create 6 practic tu tre s le identifici practically, youve got to identify them 7 Maria: da drag da i aia-i munc yes dear, but thats work too (36) (Bucharest corpus) 1 A: aa fetelei CE-ai mai fcut astzi so, girls, whats new for today 2 B: am chiulit de la german we skipped German 3 A: nu v-ar fi ruine. shame on you 4 B: mi # ce-ai fcut la german hey, what did you do at German?

According to Garnica (1977: 161) the use of a diminutive form usually marks a dependency relationship between the user and the person being referred to. Additionally, it marks a strong emotional bond between interactants to the extent to which diminutives of names, unless established as nicknames, cannot be freely used by everybody. Moreover terms such as fata and puiu meu, can also 120

be used in their diminutive forms (feti, puiu, puior) 12, enhancing thus the sense of intimacy and endearment. These terms of address and diminutives may be accompanied by lexical items, usually devoid of meaning, such as bi or m, which cannot be translated into English. Combinations of first names in either their full form or in a variety of diminutive forms and intimate address forms are also possible. Consider the following excerpt:
(37) (Constana corpus) Iulia: bi ma:ri eti gifted auzi hey, mari, youre gifted, you heard that

Terms of address can also be used in successive turns in an encounter. Although they are obviously semantically unnecessary, in most cases, they serve as expressions of strong emotion and solidarity, as excerpt 38 illustrates.
(38) (Constana corpus) 1 Maria: fat io cre c voi ai neles aiurea I think youve got it wrong 2 bi nu poi s- eventual fie:CAre well you cant lets say so : WHICH 3 Iulia: mari n-am neles aiurea da probabil c n-o s fie TExt de jumate de pagin mari, I didnt get it wrong but it probably wont be a halfpage text 4 o s fie acolo cTEva propoziii there will be FEW sentences, there 5 Maria: iulia da e imposibil s s gseti aptezeci de texte iulia, but its just not possible to find seventy texts

The flexibility in usage which diminutives offer to first names, kinship terms, and other endearment forms is extensive. This, we believe, is indicative of a sense of closeness in relationships among members of the same in-group. Using any of the innumerable ways to convey in-group membership is one of the characteristics of positive politeness through which the speaker can implicitly claim common ground with the addressee (Brown and Levinson, 1978: 112).

The use of jargon or slang In-group terminology functions in the same way as in-group language or dialect in forming emotional bonds and claiming shared attitudes13. By referring to an object or a person with a slang term, the speaker may evoke all the shared associations and attitudes that he and the addressee have towards that
12

Attempts to translate such terms into other languages can only sound ridiculous because these references will be deprived of the emotional connotations and values they have acquired in the language of origin.

121

object or person; then this may be used to redress an FTA or to reinforce existing emotional bonds between participants in a conversation and in a relationship. Not surprisingly, teenagers conversational style is particularly rich in slang terms, which makes them a marker of age rather than gender. Excerpts 39 to 46 illustrate some of the most commonly used slang terms in teenagerss conversational style in Romanian14. TEENAGERS
Excerpts 39 to 46 (Constana corpus) (39) Anca: ai copiat vreodat la geogra? have you ever cheated at geogra(phy)? (40) Anca: da-i napa dac n-ai fust its tough if youre not wearing a skirt (41) L: tie super beton pedepse he knows some cool game tasks (42) D: s-a matolit c ne-am certat noi mai mult de suprare cum ar veni he got blind drunk cause we had a fight that is cause he was angry (43) Raluca: am stat prin cartier la o caterinc we hung around the neighbourhood chewing the rag (44) Raluca: mam i m-am ofticat ( ) dac eti bulangiu man it ate me ( ) if youre an arsehole (45) Raluca: ai vzut voi dou ciumece Ive told you you smart ones (46) I: i gagica-i de gac
13

Research on code-switching has shown that switching from mainstream language to in-group dialect marks personal involvement, while switching to a code associated with external relations may mark withdrawal of positive politeness and its related emotional support (Gumperz 1970, 1975). 14 For an extensive analysis of slang in Romanian and its impact on written language, see Zafiu (2002). Although the Romanian linguist argues that slang is mainly used by young people, she does not consider it to be a marker of any social group.

122

and the chick is really cool

ADULTS The following general features worthy of note emerge from a comparison between the teeagers and the adults samples: none of the slang terms encountered in adults conversational style is unknown to teenagers as our informants reach adulthood, the percentage of slang terms falls significantly by fifty percent, which lends support to the view that a consistent use of slang terms marks membership in a certain age group rather than gender group in adulthood our informants tend to replace slang terms with jargon

Excerpts 47 to 51 illustrate the first two points.

Excerpts 47 to 51 (Constana corpus) (47) 1 Maria: vreau i io s-mi iau vai ce mito snt I wanna get some too they are so cool 2 Iulia: snt mito da i mie imi plac theyre cool I like them too (48) Iulia: aoleu da-i nasol ay, thats shit (49) 1 Maria: am avut o lecie napa de tot I had a really crap lesson 2 bi da snt nite ciutani de super detepi acolo m mori the kids there are sm super smart; theyre killing you (50) 1 Iulia: uite-i uite-i stai c ia snt ai alinei there they are! wait! those are Alinas 2 sub oalele mele acolo= under my clothes, there 3 Maria: =aha marf aha, cool (51) 1 Anca: secretara de la coal de la la care a fost omort la tic tac pe acolo the secretary at the school of the guy who was killed near tic tac

123

cic era prieten cu ea n gac they say they were friends in the same gang

Excerpts 52 to 54 are illustrative of the third point: the tendency among adults to use jargon instead of slang terms. The use of jargon enables our adult informants to express their affiliation not so much in terms of the age-group they belong into, but rather in terms of professional groups.
Excerpt 52 (Constana corpus) 1 Iulia: chestia e c e aiurea m pentru c i ia mult timp the thing is, its like shit, because it takes too much time 2 este parataxa este hipotaxa its parataxis, its hypotaxis 3 Maria:faci referat de cinpe pagini naibi you cant write a damn fifteen-page paper 4 Iulia: ce naiba mai e? what else do we have? 5 auzi i mai e o chestie la aia cu:: a:::: coordonare aa listen, and theres one more thing for that ah coordination, thats it 6 Maria : mmm mhm 7 Iulia : snt o mulime de conjuncii there are a lot of conjunctions 8 Iulia: ce le vrea pe toate? she wants all of them? 9 Maria: pi da well yeah 10 Iulia: copulative adversative= copulative adversative Excerpt 53 (Bucharest corpus) 1 B: bine c zice c dac eti hipotiroidian i afecteaz : memoria i aa [este well they say that if youre a hypothyroid, it affects your memory and its true 2 A: [da.da yes yes

Excerpt 54 (Bucharest corpus) 1 B: m gndesc s nu iau enunul asa ca: palier de mijlo:c pentru ce este sub enun i Im thinking of not taking the utterance as a midway for whats below it 2 pentru ce este peste enun pentru ca o parte s fie de pragmatic a textului for whats above the utterance, so that part of it can be about the pragmatics of the text

124

dialogului nu tiu ce of the dialogue, whatever, 4 A: care nu i-e inclus ca subwhich is not included as under5 B: pentru c titlul meu e aa enunul stru- enunul structuri semantice i m because my title goes like that the utterance stru-the utterance, semantic structures and 6 ca unitate as a unit 7 A: da yes 8 B: da m rog unitate minimal de mesaj s zicem well, any way, as minimum discourse unit, so to say

As the above excerpts have shown, in using in-group expressions known only to and used by a limited number of people, the speaker expresses a sense of solidarity with the addressee and emphasizes an assumption of shared experiences, goals, and values. Collaboratively built sentences Recall that Brown and Levinson (1987) mention the use of slang, jargon, dialect, address forms and ellipsis as devices that mark in-group identity. We can also add a further member to this list of ingroup identity markers, namely the phenomenon referred to by Sacks (1992) as collaboratively built sentences. In producing a collaboratively built sentence, a speaker A produces a complete turn made up of a single sentence which is then turned by another party B into a seemingly incomplete turn which then B completes by producing a dependent clause or prepositional phrase which are syntactically and semantically consistent with the first part. The first utterance is a possibly complete sentence, but the second utterance is not a sentence. Only with the first is it a sentence. As Sacks (1992) has pointed out it: If you want to find a way of showing somebody that what you want is to be with them, then the best way to do that is to find some way of dividing a task that is not easily dividable and which clearly can be done by either person alone. Collaboratively built sentences serve precisely this purpose: they show that the people who produce them are close to each other and they know what is on each others minds. TEENAGERS Oddly enough, the analysis of our data reveals that collaboratively built sentences are practically nonexistent among Romanian teenagers living in Constanta. Producing a collaboratively built sentence is no easy task since it requires not only linguistic competence (i.e. being able to produce a part of a sentence that is semantically and syntactically consistent with what went before presupposes mastering the syntactic processes of the language in question), but also the ability to read ones interlocutors mind. This may create the false impression that the level of linguistic and communicative competence among teenagers is not as high as the one among adults. Sacks (1992), however, provides examples of collaboratively built sentences produced by teenagers. Presumably, collaboratively built sentences convey the highest degree of solidarity and ingroup membership and in my data, at least, making such strong claims to solidarity do not appear to be an issue among my teenager informants. When it comes to claiming in-group membership Romanian adolescent girls seem to show a preference for the consistent use of address forms and slang terms. 125

ADULTS Excerpt 55 is taken from a conversation between two female students in their early twenties. The excerpt is a continuation to excerpt 26. In lines 1 and 2 Iulia is expressing her growing dissatisfaction with the idea that the teacher should expect her to teach a class she has not thought before.
Excerpt 55 (Constana corpus) 1 Iulia: deci ea care vorbete de metodic cum poa s zic so she whos talking about methodology how can she say? 2 s-mi dea o clas pe care eu nu o cunosc nu? to give me a class to teach that I dont know, right? 3 Maria: ca lecie final as a final lesson 4 Iulia: ca lecie final as a final lesson

In this exchange Iulia produces in lines 1-2 an utterance that is a possibly complete sentence, whereas Maria produces an utterance which is not a whole sentence but a part of a sentence that is syntactically and semantically coherent with the prior turn. Moreover, Marias utterance becomes a sentence only together with Iulias utterance. By fitting her talk together with Iulias talk, Maria shows that she and Iulia are close to each other that they belong in the same group. This is reinforced by Iulias repetition in line 4 that ratifies Marias contribution. 5.2.1.3. Speaker claims common point of view/attitudes with Addressee 5.2.1.3.1. Speaker seeks agreement with Addressee It has been shown that expressing agreement with what the preceding speaker has said is common in female conversational discourse in English (Maltz and Borker 1982; Coates 1993). Agreement may be expressed by using exact or slightly varied allo-repetitions and minimal responses (Schegloff 1982, Brown and Levinson 1987; Tannen 1989). Repetition Agreement may be stressed by repeating part or all of what the preceding speaker has said. In addition to demonstrating that one has heard correctly what was said (satisfying thus another positive politeness output, namely notice, attend to H), exact or slightly varied allo-repetition may be used to stress emotional agreement with the utterance and the addressee or to stress interest or surprise. In what follows I exemplify the following functions served by repetition of words, phrases and clauses in conversation: participatory listenership, ratifying listenership, providing back-channel response, expanding, and 126

persuading the addressee15.

In all these examples, repeating part of what the preceding speaker has said is done with the view to showing acceptance of the others utterances, of their participation and personality. TEENAGERS The analysis of our teenager data16 shows that the instances of repetition functioning as indicators of agreement are scarce when compared to adults style. Unlike adults style, where repetition is highly versatile, displaying a multiplicity of functions, all repetitions identified in my teenager data serve one function only: showing participatory listenership17. The following example is a case in point: (i) Repetition as a way of showing participatory listenership Extract 56 illustrates what Tannen (1989) has called displaying participatory listenership through repetition, i.e. the exact or slightly varied repetition of a previous speakers utterance.
Excerpt 56 (Constana corpus) 1 Raluca: ai nceput anu bine a::: youve started the year well, ah 2 Mona: da am nceput anu numai bine l-am nceput well, weve started it well, all right

At line 2 Mona does not simply confine herself to agreeing with Raluca monosyllabically, rather she repeats with slight variation Ralucas comment in line 1 showing thus listenership and acceptance of Ralucas utterance. ADULTS Research on repetition has shown that not all conversations exhibit a high percentage of repeated words and phrases, but many do and all show some. This raises the question of the extent to which frequency of repetition is culturally variable. Since repetition of sentences and ideas is a means of keeping talk going in interaction, the relative frequency of repetition can be correlated with the cultural value placed on the presence of talk. In addition to tying parts of discourse to other parts, repetitions bonds participants to the discourse and to each other, linking individuals in a conversation and in a relationship. Thus we would expect cultures and speech communities that place a relative positive value on talk in interaction to exhibit a relative high frequency of repetition. Research on high-involvement styles provides support for these predictions (see Tannen 1989).
15 16

These functions are identified, discussed and illustrated in Tannen (1989). This section on repetition is based on our paper Repetition in female conversational style presented at the Annual Conference of the English Department Occidentalism and Orientalism , Bucharest, University of Bucharest in 2002 as well as on Hornoiu (2003a). 17 For a detailed discussion of repetition as participatory listenership see Tannen (1989)

127

Repetition as a way of ratifying listenership In line 1 of excerpt 57 Iulia uses a couple of phatic questions to introduce another topic for discussion18. She asks Maria whether Bogdan, Marias husband, has finished organizing a certain concert. Maria answers that Bogdan is still busy with the concert and, in order to provide evidence as to how busy he is, she lists several activities that Bogdan has to do, by using three self-repetitions with slight variation in lines 5 and 6 (du-i la hotel take them to the hotel, du-i s fac probe take them to rehearsal, du-i s fac aia take them to do this, that and the other).
Excerpt 57 (Constana corpus) 1 Iulia: bogdan? = ce face? a terminat cu concertu?= bogdan? what is he doing? has he finished with the concert? 2 Maria: =eh bogdan = nu smbt eh, bogdan no, on Saturday 3 Iulia: =aha aha 4 Maria: tre s i dai seama c alearg ca nebunu pn vin ia youve got to know hes running around like crazy until they come 5 ia-i p-ia du-i la hotel pune-i la mncare pn nu tiu ce pick them up, take them to the hotel, to eat til I dont know what 6 du-i s [fac probe du-i s fac aia take them to rehearsal, take to do that and that 7 Iulia: [pe cine pe cine la hotel? whom whom is he taking to the hotel? 8 Maria: pi p-ia care cnt well, the singers 9 Iulia: da ce vin din ar? trupe? why, are they coming from other parts of the country? bands? 10 Maria: pi vin din ar well, theyre coming from other parts of the country 11 unii vin din timioara some are coming from Timioara 12 unii vin din bucureti some are coming from Bucharest 13 unii vin din (craiova) some are coming from (Craiova)

After a couple of more questions whose main function is to keep the flow of conversation going rather than to ask for information or clarification, proving thus Iulias interest in the topic, Iulia enquires in line 9 about the whereabouts of the bands performing in the concert ( vin din ar? are they coming from other parts of the country?). Maria ratifies Iulias contribution in line 9 by repeating it in line 10 (vin din ar there are coming from other parts of the country) and incorporating it into her narrative. She then lists the cities from which the bands are coming by making use of three self-repetitions in lines 11-13; she sets up a paradigm in line 11 and slot in new information:
unii vin din timioara some are coming from Timioara
18

As we shall see later in this chapter, this strategy is commonly used in female conversational style.

128

unii vin din bucureti some are coming from Bucharest unii vin din craiova some are coming from Craiova

The establishment of the pattern allows Maria to utter whole new sentences while adding only the names of the cities as new information. This is a convenient way of producing more language, more fluently and therefore of carrying on conversation with relatively less effort. Another commonly used type of repetition is the exact or slightly varied repetition of a previous speakers utterance, as we have already seen above shows participatory listenership. The following example coming from the same conversation illustrates participatory listenership:
Excerpt 58 (Constana corpus) 1 Maria: i le dau stora le dau drumu i cazare tii and theyre paying for their travel and accommodation, you know 2 [drumu cazare i mas travel, accommodation and meals 3 Iulia: [drumu le pltesc le pltesc drumu la tia travel expenses they cover travel expenses for them 4 Maria: mhm mhm 5 Iulia: da yes 6 Maria: snt vreo trei care vin about three are coming 7 unu vine din bucuresti one comes from Bucharest 8 deci o trup vine din bucureti so a band comes from Bucharest 9 o trup vine din timioara a band comes from Timioara 10 i o trup (nu- de unde vine) and a band (dont know where from)

Iulias repetition with slight variation in line 3 (drumu le pltesc le pltesc drumu la tia), echoing Marias utterance in lines 1-2 (i le dau stora drumu) seems to be a way for Iulia to participate in the interchange by showing listenership and acceptance of Marias utterance. In lines 11-13 of extract 57 and 8-10 of extract 58, (separated by a page and a half of transcript) Maria repeats the clauses with slight variation. By restating her contribution, she continues to take part in conversation even though she has nothing new to add. As these examples show, repetition provides a means of keeping talk going, where talk itself is a sign of involvement, of willingness to interact, serving thus a positive face. The analysis of our data reveals repetition to be a pervasive feature in female conversational style in Romanian. This is in keeping with research on female conversational style in English which has been characterised as collaboration oriented or affiliative, i.e. as focusing on the relationship. Repetition is a resource by which conversationalists together create a discourse and a relationship. Together with asking questions, it is an important strategy for keeping the flow of conversation going and creating interpersonal involvement when taking part in conversation is equivalent to being part of a relationship. 129

Minimal responses as an indication of agreement In addition to functioning as continuers, minimal responses can also be used to signal the passing up of an opportunity to initiate repair (Schegloff 1982). If any talk can be viewed as a trouble-source, then after any talk can be a place for repair to be initiated on it. Vocalizations like uh huh, mm hmm, yeah, right as well as head-gestures such as nods, in passing an opportunity to initiate a full turn at talk, can be seen to be passing an opportunity to initiate repair on the immediately preceding talk. Passing the opportunity to raise problems of understanding may be taken as indicating the absence of such problems and hence the absence of disagreement. Thus minimal responses that occur in places where other-initiated repairs are potentially relevant, i.e. after completion of any unit of talk (Schegloff 1982), are taken as indications of agreement. Consider the following two excerpts that exemplify the use of minimal responses as indicators of agreement in all female discourse. TEENAGERS
Excerpt 59 (Constana corpus) 1 Raluca: dac ar fi totul prea frumos aa tii cnd e prea bine tii c a greit undeva if everything was too good, then, you know, when its too good, you know that hes done something wrong 2 Mona: mhm mhm 3 Raluca: tii i te i plictiseti la un moment dat you know, and you get bored at some point 4 Mona: mhm mhm 5 Raluca: de atta bine omu se plictisete if everything is too good, one gets bored

ADULTS
Excerpt 60 (Constana corpus) 1 Iulia: uite eu chiar snt curioas c ieri mi-a luat msura look, Im really curious, cause yesterday she took my measure 2 Maria: mhm mhm 3 Iulia: snt curioas s vedem cnd m cheam la prob Im curious to see when she calls me to try it on 4 c dac m cheam la prob i iese bine p orm imediat i face tii cause if she calls me to try it on and it turns out well, then she immediately makes you one, you know 5 Maria: mhm mhm 6 Iulia: pi eu i-am zis lu ele- elenei i-a fcut costum I told you she made Elena a suit 7 sacou i fust cu cptueal n trei zile

130

blazer and a skirt with lining in three days n trei zile era gata i l-a dat deja (2) in three days it was ready and she gave it to her 9 e foarte bine da its very good, yeah 10 Maria: mhm mhm 8

In both extracts these minimal responses occur at transition relevant places, i.e. after the completion of a unit of talk when speaker change may occur and other-initiated repairs are potentially relevant. However, in all cases those who utter them refrain from initiating repair, thus implying that they agree with their interlocutors opinions (Schegloff 1982). 5.2.1.3.2. Speaker avoids disagreement 5.2.1.3.2.1. Token agreement The desire to agree or at least appear to agree with the addressee leads to mechanisms for pretending to agree, instances of token agreement19. More often than not, speakers may alter their utterances so as to appear to agree or to hide disagreement (Brown and Levinson 1987). To take an example, one may respond to a preceding utterance with Yes, but rather than a blatant No. The following examples illustrate this positive politeness strategy: TEENAGERS In line 1 of excerpt 61 Mona refers to a mutual acquaintance describing him as someone who occasionally is a heavy drinker.
Excerpt 61 (Constana corpus) 1 Mona: sava vreau s spun cteodat exagereaz cu butura I mean Sava sometimes overdoes it when it comes to drinking 2 Raluca: da bea foarte mult= nu:: cteodat n gener[A::l exagereaz yes he drinks a lot not sometimes, he usually overdoes it 3 Mona: =da [n general yes usually

Her interlocutor, Raluca, agrees with her and reinforces Monas description of Sava as someone who drinks heavily da bea foarte mult (yes he drinks a lot) and then she moves on to an opinion which contradicts, to a certain extent, Monas assertion and argues that this feature does not apply to Sava occasionally, as Mona said, but generally: nu cteodat n general exagereaz (not occasionally he exaggerates in general). ADULTS

19

The strategies for avoiding disagreement in Romanian conversational discourse have been initially analyzed in Hornoiu (2003b).

131

In excerpt 62 Iulia expresses her dissatisfaction with the idea that, for her final lesson during practicum, she should teach a class she has never thought before, saying in lines 1-2 dar ideea este c nu este corect cum s m duc io s s fr ni- nici un (but the idea is that its not fair how shall I go to to without any any).
Excerpt 62 (Constana corpus) 1 Iulia: dar ideea este c nu este corect the thing is, its not fair 2 cum s m duc io s s fr ni- nici un how am I going to go, to to with no no 3 Maria: i literatur and moreover its literature 4 Iulia: da (3) m rog asta nu-i o problem c pn la urm dac ia nu tiu vorbeti tu yeah, well, thats no problem, cause, after all, if they dont know anything you do the talking

Maria intervenes in line 3 adding i literatur (and moreover its literature) implying that the subject of study, German literature as opposed to German language, makes Iulias task even more difficult. After agreeing with Maria in line 4 ( da yes), Iulia uses (in the same line) the item m rog, well, which, in this case, is an appositional term signalling disagreement and then she moves on to saying something that may be seen as standing in complete disagreement with Marias turn: asta nu-i o problem c pn la urm dac ia nu tiu vorbeti tu (thats no problem cause if they dont know anything you do the talking). In addition to the item m rog well, the dispreferred nature of her turn is also signalled by the three-second pause. Thus Iulia uses a token agreement to avoid disagreement and to convey that she and Maria share common opinions. 5.2.1.3.2.2. Hedging opinion Hedges vary greatly not only in terms of form, but also with regard to the functions they serve. The use of hedges is one of the negative politeness strategies identified by Brown and Levinson. Hedges are used to redress various kinds of face threatening acts (such as criticism, complaints, requests, suggestions, etc) or to strengthen the force of other acts that may be seen as beneficial to the addressee (e.g. promises). They may also be used to stress speakers commitment to the truth of their utterance or to suggest that they are not taking full responsibility for the truth of their utterance, in which case they become simple yet efficient devices for avoiding disagreement with the addressee. Hedges are important devices used in marking topic changes. Such changes are face threatening and therefore are often done off record, the use of hedging serving precisely this purpose rather than signalling lack of confidence. In such cases hedges redress the imposition on the addressees face perhaps partially apologise for it (Brown and Levinson 1987). The use of softeners We have already seen that one positive politeness output (exaggerate interest, approval, sympathy with H) leads S to exaggerate and this is often manifested by employing emphatic stress and extreme case formulations. Thus apart from its self-disclosing nature that accounts for womens consistent use of hedges, all-female discourse is particularly rich in exaggerated intonation, emphatic stress as well as 132

extreme case formulations like marvellous, extraordinary, wonderful divine, delightful absolutely, incredible, completely, etc. Nevertheless, in light of the desire to agree, this element of exaggeration is risky unless the speaker is certain of the addressees opinion on the topic. Therefore one characteristic device in positive politeness is to hedge20 these extremes by using softeners21, making thus ones own opinion safely vague (Edwards 2000) and thereby avoiding disagreement. Again the use of softeners becomes a powerful device for saving both speakers face. However, like many other pragmatic particles softeners are multifunctional. It is not always clear whether the basic function of softerners is to modify the propositional content or the illocutionary force in order to avoid or minimize interactional face threats (Sifianou 1999:164). In addition to mitigating the imposition of face-threatening acts, a function mainly characteristic of societies with negative politeness orientation, we can safely argue that in societies with positive politeness orientation softeners tend to serve another equally important interactional function: that of expressing shared knowledge, albeit to a limited extent, thereby offering the addressee the opportunity to provide support, understanding, participation, in other words, to show that both speaker and addressee are on the same wavelength. Softeners include diminutives, tag questions, and a variety of other devices that enable the speaker to weaken or qualify the force of an extreme case formulation so as to invite shared knowledge and thus to avoid disagreement. The use of softeners: diminutives Romanian is a morphologically rich language, both inflectionally and derivationally. Among the derivational processes we can mention the production of diminutives by means of special types of suffixes. Morphologically, diminutives are produced in Romanian from a variety of word classes, although the most productive class of all is that of nouns. Sometimes the same stem can be given either one (e.g. ut; -el) or another (e.g. sor; -as) diminutive suffix (for instance, puiut, puisor; baietel, baietas); occasionally the diminutive suffix may indicate sex; for instance, pisi small cat can be either male or female, but pisoi or pisoias small cat is male and pisicuta small cat, female only. It is worth noting that some diminutive forms are produced through the addition of more than one diminutive suffix, as in pisicpisoi pisoia (small cat). This is a feature characterictic of many other European languages as well; however, it does not apply to English on such a large scale. In English, on the other hand, diminutive suffixes are fewer and diminutives themselves are not so frequently used. Examples, such as -ette in kitchenette, and -let in piglet and starlet are used with decreasing frequency (Quirk et al., 1972: 994). Another group of suffixed diminutives common in English consists of words accepting the -ie suffix (sometimes -y) such as doggie and mummy. Although the flexibility of this suffix is greater than those mentioned above, it is still limited because although doggie and birdie are common, there is nothing like headie or cowie (Wierzbicka, 1984: 126). This ending can also be observed in shortened forms, such as hankie for handkerchief and hubby for husband. Another group of morphological derivatives in English akin to diminutives are contracted forms of longer words. Examples such as daffs for daffodils and veg for vegetables vegans for vegetarians are common and new ones are continually being introduced.
20

According to Brown and Levinson (1978: 150), a hedge is a particle, word, or phrase that modifies the degree of membership of a predicate or noun phrase in a set; it says of that membership that it is partial, or true only in certain respects, or that it is more true and complete than perhaps might be expected. In general their function is to to soften the force of the face threatening acts. Brown and Levinson argue that hedging can be achieved in indefinite numbers of surface forms (ibid.: 151) and they group these surface forms into two classes weakeners (mainly acting as tentativizers) and strengtheners (mainly acting as emphatic items). 21 Hedges have been grouped into softeners and intensifiers, with softeners mitigating the force of the imposition in at least the two ways stated above and intensifiers aggravating the impact of the face threatening act.

133

Abbreviations such as info for information and memo for memorandum are common in the jargon of certain groups with shared interests, groups that tend to use the same stock of words very often signalling thus in-group membership. As the term suggests, the primary function of diminutives is to express the idea of smallness. However, they also carry a number of affective connotations ranging from endearment to tenderness through mild belittlement or deprecation to outright derogation and insult (Haas, 1978: 82). We have already seen that diminutives frequently accompany various classes of address forms. In Romanian, however, the use of diminutives may be extended beyond their function as address forms. Perhaps the most interesting feature of diminutives in Romanian is that, although they mainly concern morphological alternations at word-level, they can also be used in ways that affect the force of the whole utterance or interaction. A consistent use of diminutives is characteristic of positively polite conversations, especially among women, where they function as an overall endearment for the topic of the interaction. Thus diminutives can be freely added to inanimate nouns as well as adjectives and adverbs when no indication of smallness is involved or implied. Under such circumstances they function as markers of small talk, rather than purely descriptive items that indicate the smallness of the referent. Consider, for instance, extracts 63 and 64 which are part of an interaction between friends. Both extracts illustrate adults use of diminutives to indicate social relationships. In excerpt 63 Maria is admiringly describing a pair of trousers Iulia has recently bought:
Excerpt 63 (Constana corpus) 1 Maria : i snt d-ia and theyre that kind 2 Iulia: snt cu brdu= theyve got a fir tree (dim.) pattern 3 Maria: =lrgui nu? loose (dim), arent they? 4 Iulia : da yes 5 Maria : snt un pic evazai [.] theyre a bit flared 6 Maria : i ia i vin- i tia tia tii cu ce merg ? those fit you and these, do you know what they go with? 7 Iulia : mhm ? mhm ? 8 Maria : cu ia ai ti de iarn ia nchii chiar la culoarea asta with the winter ones youve got, the dark ones, exactly this colour 9 Iulia : care de [iarn ? which winter ones? 10 Maria : [ghetuele the boots (dim.)

In this excerpt Marias use of diminutives in lines 3 and 10 conveys endearment and sympathy towards a particular item and enables her to extend these feelings towards the addressee-owner22.
22

When speakers use diminutives to refer to their own possessions, achievements, or characteristics, the connotations may be those of affection but may also be attempts to reduce the possibility of the utterance being interpreted as self-praise (Sifianou 1999:167).

134

In excerpt 64 Maria, who is paying a visit to her friend, is offered in line 1 a helping of spaghetti which she declines in line 2. In line 3 Iulia describes the spaghetti as having been made with crni de pui poultry-dim.
Excerpt 64 (Constana corpus) 1 Iulia: nu vrei spaghete? wont you have some spaghetti? 2 Maria: vai las-m c snt oh, dont tell me! Ive grown 3 Iulia: a facut alina nite spaghete cu crni de pui alina made some sphagetti with poultry-dim 4 Maria: [e:::::: e :::: 5 Iulia: [n-ai mncat n viaa ta aa ceva youve never eaten something like this

Since there is no imposition to be minimized in extract 63, whereas in extract 64 the imposition due to the sequence offer-refusal is kept to a minimum given the fact that the extract is part of a spate of small talk between two long-time friends and university colleagues, the diminutives in these two examples indicate in-group solidarity. Diminutives are especially frequent in everyday informal speech where they mainly involve routine actions dealing with the exchange of free goods. In such environments they serve various positive politeness needs, as the extracts above have shown. Diminutives are not usually used when there are status differences between the interactants (Sifianou 1999:167) presumably because the conflict between intimacy and status makes diminutives expressing intimacy and familiarity inappropriate in interactions where participants are of different social status. However, extract 65 which comes from an encounter with a dressmaker illustrates the use of diminutives in an interaction characterised by differentials between interactants in socioeconomic status. Here various diminutive suffixes added to adjectives and nouns occur in an extended spate of task-oriented talk and move from the thinness of the fabric, to the skirt, to the undershirt, to the thickness of the fabric.
Excerpt 65 (Bucharest corpus) 1 C: da sta CE are but WHATS wrong with this one? 2 E: e mai subirel poate i se ia p picior maybe its thinner-dim. and it fits on the leg 3 B: e mai subiric [] its thinner-dim 4 A: nu mi-am dat seama c-i materialu subire I havent realized the fabric is thin 5 B: da ce eu mi-am dat seama? e materialu subire you think I realized? the fabric is thin 6 A: l fceai tot cu fust you would have made it with a skirt as well 7 B: da

135

yes 8 A: tot cu fusti [] with a skirt-dim. as well 9 A: i-o-ndoi doar un pic tnica you fold it just a little tanta-dim. 10 C: h yep 11 B: da da yes yes 12 C: pi da da mi-am luat i maieuu ala aa s port pantalonii cu maieuu la aa [.] well, yes! but Ive also got that top-dim. to wear the trousers with that top-dim. 13 A: l croieti mai larg mult mai larg spatele [tnica you tailor it looser, much looser at the back, Tanta-dim 14 B: [mult mai larg SPA[TELE the BACK much looser 15 C: [LARG ct cuprinde as LOOSE as possible 16 B: da da yes yes 17 A: h [...] yep 18 E: nu da materialu subire ntotdeauna se ia p but the thin fabric always fits on 19 B: da da yes yes 20 E: i cnd e mai grosu el st mai lrgu [...] and when its thicker-dim. it falls looser-dim. 21 B: chiar nu mi-am dat seama. cu de toate am lucrat I really didnt realize it Ive worked with all the kinds 22 da uite nu mi-am dat seama c se poate ntmpla aa but, look! I havent realized that this might happen 23 A: BINE c mai are material i facem spatele din la GOOD thing weve got some fabric left and we make the back from that 24 B: da. pi da bine c n-am scos mnecile c n-am scos mnecile i:: yes well, its a good thing I havent made the sleeves, I havent made the sleeves and 25 E: astea ies [din these will come out from 26 B: [da yes 27 C: din astelalte from these other ones 28 B: aa i cu gulerau-la like this, and with that collar-dim. 29 A: i uite c vine frumos [...] and see? it looks nice 30 A: nu tnica dac-i faci spatele un pic mai mare no, Tanta-dim. if you make the back a bit larger 31 ai s vezi c-i d drumu youll see, it will loosen 32 B: da da pi d-aia zic well, yeah! thats what I was saying

136

Since interactants perform specific roles, requests are not perceived as impositions, and diminutives show readiness for co-operation in a friendly atmosphere. On the other hand, there is some imposition involved in this extract which stems from the social distance between the service provider and the clients due to social status and age differences; these differentials in socioecomic status are acknowledged and mitigated through other means: the assymetrical use of polite and diminutivized forms of address. The redressive force of diminutives stems from their association with in-group language; this feature enables the interactants to express their wish to establish a successful encounter where they are co-operating members. Clearly the function served by diminutives in these excerpts is to stress the emotional bond between the participants in the interaction, rather than being intended as literal descriptions of their referents as being small. These affective connotations thus shift from applying to one lexical item to applying to the whole conversational encounter. This points to a solidarity-oriented view of the whole encounter and of the relationship in general. It has been argued that rich systems of diminutives seem to play a crucial role in cultures in which emotions in general and affection in particular is expected to be shown overtly (Wierzbicka, 1985a: 168). As the excerpts analysed so far have shown Romanian womens preference for a consistent use of diminutives is indicative of their tendency to value spontaneity and to express both their negative and positive emotions overtly. Thus, diminutives appear to be instrumental in facilitating the expression of feelings. By contrast, the Anglo-Saxon culture does not ecourage such an unrestrained display of emotions and feelings. Consequently, the English system of diminutives has not been developed to the same extent, nor are diminutives so extensively used. The use of softeners: miscelaneous Another way of expressing diminution in Romanian is by means of syntactic modification, namely, by using the words puin and un pic little and a noun. Moreover, the collocation of such items with already diminutivized nouns is also common. Similar constructions can be found in English as in little doggie, but they are rather restricted to such environments as baby talk. Excerpts 66 and 67 illustrate the use of such combinations in contexts where they further emphasize the smallness and thus the endearing aspect of what is mentioned.
Excerpt 66 (Constana corpus) Iulia: auzi da nu-mi pui i mie puin suc c-mi faci poft teribil hey, wont you give me some juice, cause youve made my mouth water

Excerpt 67 (Constana corpus) 1 Iulia: da mai pune-i mai pune-i please, have some more, have some more 2 Maria: hai s mai pun i io puin Ill have a little bit too

Although theoretically these combinations with nouns and puin can be ambiguous between a quantifier interpretation and a politeness marker, the context makes the intentions and meanings clear. 137

Evidence for this is the fact that interactants do not usually act on the literal meaning of the utterance: the quantity offered is not determined by the presence or absence of puin/un pic. A variety of other expressions can perform a similar politeness function. For example, un moment a moment, o secunda a second, and o gura a sip. Moreover they themselves can also be modified by mcar, numai or doar only. The modifiers puin/un pic a little or a bit and mcar/numai/doar only can also modify verbs, encoding extra politeness as the following example illustrates.
Excerpt 68 (Constana corpus) 1 Iulia: alina nclzete o porie de spaghete pentru mari= alina heat up a helping of spaghetti for mari 2 Maria: =nu iulia [nu: no, Iulia, dont 3 Iulia: [ha:ide o lingur come on, just a bite 4 mcar s guti taste it at least 5 Maria: nu nu mnnc [.] no, I wont eat 6 Iulia: i ciorba de perioare e foarte bun the meatball soup is very good too 7 Maria: [oh s n-aud oh, I wont hear of it 8 Iulia: [bi numai o lingur come on, just a sip 9 mcar s guti uite well, taste it at least 10 Maria: s nu mai aud de mncare no more talking about food

These modifiers are more flexible than their English counterparts. They can collocate with a variety of verbs and are not restricted in terms of the grammatical construction with which they are employed. These collocations are far from being ambiguous. They cannot be interpreted as meaning that the requester or the one who makes the offer wants the actions carried out partially or for a short time. Clearly then putin/un pic (little/a little) in the above example does not minimize the imposition by minimizing the action requested. It could be suggested that in such contexts, putin/un pic (little/a little) has lost its literal force and is an informal variant of te rog/va rog please, which can be used instead of or along with it. This use of putin is not peculiar to Romanian since it can be regarded as similar to what Brown and Levinson (1978: 144, 182) report for Tamil and Malagasy, where the words for please literally mean a little. Sifianou (1999:171) argues that the Greek counterpart behaves in a similar way. It is not a coincidence that all these cultures have been shown to favour a positivepoliteness orientation. There are a number of extremely common modifiers in Romanian whose function is to tentativize what speakers say, thus allowing them not to fully commit themselves to what they are saying. They include such items as cteodat, n general, cam, un fel de, oarece, sometimes, in general, rather, a sort of and many others. The following excerpts show speakers as being oriented 138

towaeds the requirement of extreme case formulations to be occasionally weakened, or qualified so as to avoid disagreement. TEENAGERS Excerpt 69 is taken from a conversation between two Romanian adolescents and reproduces the first line of extract 61. We have already seen that in this excerpt Mona refers to a mutual acquaintance arguing that he is a heavy drinker.
Excerpt 69 (Constana corpus) Mona: sava vreau s spun cteodat exagereaz cu butura I mean Sava sometimes overdoes it when it comes to drinking

Since her description amounts to an extreme case formulation and since, on the one hand she wants to avoid potential disagreement, and on the other hand, she seems to resent the conversation drifting towards malicious gossip about their friend, she uses the adverb cteodat (sometimes/occasionally) to hedge her description: sava vreau s spun cteodat exgereaz cu butura (I mean sava sometimes overdoes it when it comes to drinking). Romanian adolescent girls do not seem to be particularly concerned with hedging their opinion in order to avoid disagreement. Rather they seem more preoccupied with asserting their point of view. Indeed, softeners are an isolated occurrence in Romanian adolescentsconversational style as Table 3 shows but are not totally unknown. As they reach adulthood, they apparently become more attentive to their interlocutors positive face and more preoccupied with avoiding divergent views, at least in friendly talk ADULTS Excerpt 70 is taken from a conversation on complementary medicine and herbal remedies and exemplifies the intricately woven fabric of extreme case formulations and softeners.
Excerpt 70 (Constana corpus) 1 Iulia: asta pentru ce o iei? what do you take this for? 2 Maria: uite s-i spun alina [pentru de toate well, alina can tell you (its good) for all kinds of things 3 Iulia: [s nu mai ai poft de mncare= to assuage your appetite 4 Maria: =pentru de toate for all kinds of things 5 Alina: n general cele care snt bune de pentru toate nu prea snt bune in general those that are good for- for all kinds of things arent that good 6 Maria: deloc ? at all? 7 Alina: da m rog au oarece

139

well theyve got some 8 Maria: dar am neles c but Ive heard that9 Alina: da spirulina asta cre c e- a ajuns aa: un fel de ap sfiinit yeah, the spiruline has come to be some sort of holy water

Both Maria and Alina are at pains not to disagree with each other while stating their rather conflicting opinions. This leads them to hedge their extreme case formulations so as to hide disagreement conveying thus that they share common opinions. Previously in the conversation Maria mentioned spirulina, a drug based on algae extracts that she takes as part of her slimming diet, arguing in line (2) that it is good for everything. Notice that she uses an extreme case formulation. Since her opinion is delivered in overlap with Iulias turn which supplies in line (3) a possible reason for taking this drug (s nu mai ai poft de mncare to assuage your appetite, Maria repeats her extreme case formulation in line (4) when she emerges into the clear. Moreover, she expects Alina, who is a doctor, to agree with her opinion: uite s-ti spun Alina (e) pentru de toate (well, Alina can tell you its good for all kinds of things). Marias use of an extreme case formulation proves to be a risky strategy, in light of her desire to agree, since she is unaware of her addressees opinion. Alina does not seem to agree with Maria since the point she is about to make in line (5) is that drugs that are claimed to be good for all kinds of things are not very efficient, and this would stand in total disagreement with Marias opinion. However, in line (5) Alina seems to be attending to the capacity for simple unmodified generalizations to be taken as equivalents to extreme case formulations. Consequently, she uses two softeners in general (in general) and (nu) prea (hardly) which enable her to mitigate the force of an utterance that might stand in total disagreement with her interlocutors assertion. Thus she makes a more plausible claim that is in keeping with her wish to avoid disagreement with Maria. Maria, on the other hand, shifts to another point of view and aligns with Alinas position by supplying in line 6 the adverb deloc (not at all), an anticipatory completion that reinforces Alinas opinion. Thus Maria abandons her initial point of view in order to claim common opinion with Alina. The upshot that emerges from this collaboratively produced sentence amounts to another extreme case formulation that stands in complete disagreement with Marias initial opinion: drugs that are good for all kinds of things arent good at all. Apparently they have reached agreement, but reaching this agreement means rejecting Marias initial opinion, something Alina would rather not do. Hence Alinas unexpected turn in line (7): da, ma rog, au oare ce (well, yes they have sort of) consisting of a token agreement (da, ma rog), meant to hide her partial disagreement with the assertion jointly produced by her and Maria, and a softener (au oare ce), meant to mitigate the force of the jointly produced utterance that amounts to an ECF (which as we have already seen stands in disagreement with Marias initial opinion). Alinas turn in line 7, far from showing lack of assertiveness, actually signals her concern for Marias positive face. The same goes for Maria who, after stating an opinion which later seems not to be favoured by Alina, embraces Alinas totally conflicting opinion by producing a collaboratively built sentence meant to signal shared point of view. The whole exchange is the interplay between extreme case formulations meant to create involvement and instances of token agreement and hedging meant to avoid the disagreement that may be triggered by the former. Occasionally such tentativizing modifiers can also be followed by diminutivized nouns or adjectives as the following excerpt shows. Excerpt 71 reproduces part of conversational exchange in which Iulia and Maria are discussing their personal experiences in teaching.
Excerpt 71 (Constana corpus)

140

1 Iulia: ia de a nou snt [cam slbui those in the ninth grade are a bit weak dim. on grammar 2 Maria: [cam slbui nu se prind ce greeli faci a bit weak -dim. on grammar they dont realise when you make mistakes

The combination between the tentativizing modifier cam rather/ a bit and the diminutivized adjective slbui weak-dim/mediocre-dim.allow the interactants to further tentativize what they say and to minimize the force of the face-threatening act. The use of softeners: tag questions The English system of tag questions is highly developed both in terms of form and function 23. The equally developed system of modal auxiliaries24 has perhaps contributed to such a wide formal variety of tags in English25. It has been argued that English tags enable speakers to mitigate the force of the speech act by sounding more tentative and less committed to the truth of the proposition. Lakoff (1977: 83) contends that tags have the effect of hedging protecting a speaker from the consequences of his speech acts. Thus, in using tags, speakers not only express their concern not to impose on the addressee but they also protect themselves against a possible refusal. Lakoff (1975: 16-17) claims that in general women use more tag questions than men, exactly because this enables them to avoid commitment and thus conflict. She adds that tags also signify that speakers are not really sure of themselves and might not even have views of their own; thus, according to Lakoff, womens consistent use of tag questions would be indicative of their insecurity and of their need to seek confirmation from the addressee. Lakoffs claims have been attacked on the grounds that women do not always use tags more often in all situations and that the functions served by tags cannot restricted to just signalling lack of confidence26. Tannen (1990: 228) suggests that it seems more a matter of expectation and stereotyping rather than actual difference between the two gender groups in the frequency of employing tags. By contrast, the Romanian system of tag questions includes two items, nu-i aa isnt it and nu no, which are invariable in form and can collocate with any verb. In addition to these two items, Romanian uses two token tags tii you know, nelegi you understand and vezi you see whose function is to invite the addressees involvement and thus to establish agreement and common ground27. TEENAGERS The following example illustrates the use of the token tag stii you know among our adolescent informants. It is interesting to point out that troughtout my adolescent corpus this was the only (token) tag questions used. Apparently, the use of tag questions, like other interactional strategies analysed here, diversifies significantly as our informants reach adulthood.
23 24

The literature on this issue is extensive. See Lakoff (1969 and 1975), Cattell (1973) among others. If we consider the tag questions in such sentences as Leave it here, will you? or Have a glance at these, won't you?, we could supply a number of alternative possibilities for these tags, such as can you? or can't you?. In Romanian, however, the only possibility, if used at all in such contexts, would be the token tag bine , which means OK. 25 This is not true of other languages, such as Greek (Sifianou 1999), Hebrew (Blum-Kulka, 1982 and 1983) and Polish (Wierzbicka, 1985). 26 See, for instance, Dubois and Crouch (1975); and Holmes (1986), among others. 27 Brown and Levinson (1978:235) argue that When token tag questions are tacked on to a presumptuous positively polite request ... the results are basically still positive-politeness strategies, even though they make use of essentially negativepoliteness techniques to soften the presumption.

141

Excerpt 72 reproduces part of a conversational exchange in which the interactants are making small talk. At this point Mona is making confessions about the misfortunes of her love affair.
Excerpt 72 (Constana corpus) 1 Mona: pi vreau s spun c cre c-a fost cel mai nasol weekend de cnd snt cu george well, I mean I think it was the worst weekend since george and me started going out together 2 a fost smbta cnd s-a-ntmplat smbt sear it was on saturday when it happened on saturday evening 3 a fost duminic care pur i simplu n-aveam ce vorbi cum it was on sunday when we simply had nothing to talk about as4 dac deschidea unu gura ne certam tii if one of us opened the mouth we started fighting, you know 5 i duminic iari ( ) a fost urt luni iar ne-am certat tii and again on sunday ( ) on monday it was ugly we had arguement again, you know 6 Raluca: acum v-ai revenit la starea iniial? now youre back to the way it used to be? 7 Mona: da cred mai avem unele scpri d-alea tii yes, but I think weve still got some outbursts of anger, you know 8 mereu apropouri unu la altu [i :: we always drop innuendoes to each other and 9 Raluca : [da e normal s se-ntmple aa well, its normal to be this way 10 deci nu o s ai niciodat ceva de genu tot timpu s fie frumos so, youll never have something like good all the time

You know is basically a marker of information state transition. It has two discourse functions: first it is a marker of meta-knowledge about what speaker and hearer share; second you know is a marker of meta-knowledge about what is generally known. The fact that you know verbalizes speakers handling of cognitive tasks has interactional consequences. You know may open an interactional negotiation over the informational status of a generalization, i.e. the degree to which a certain piece of information is really shared knowledge. Thus you know may also be used to seek interactional alignments by establishing shared opinion (Schiffrin 1987). In both of the above examples stii you know marks transition from a state in which the addressee was not aware of a certain piece of information to a state when that piece of information becomes shared knowledge. Thus speakers may use you know to establish common ground/opinion and to enlist hearer agreement when such agreement is not otherwise forthcoming. Token tags far from being markers of insecurity and lack of assertiveness, as Lakoff claimed, are used to draw the addressee as a participant, into the conversation, reassuring the latter that the speaker cares about his opinions. When serving this purpose tags become an efficient way of avoiding potential disagreement. ADULTS Excerpts 73 and 74 below illustrate the use of stii you know in adults conversational style to mark the transition from a situation when speaker knows that addressee does not have particular information to a situation when speaker knows that hearer shares knowledge. 142

Excerpt 73 (Bucharest corpus) 1 B: eram TErifiat de ideea c vor veni i vor sta la noi I was TErrified at the thought that they will come stay with us 2 mai ales netiind nimic ct timp au especially not knowing anything about how much time theyve got 3 c dac au timp puin i nu tiu ce mcar ameninarea era mai limitat cause if theyve got little time and, whatever, at least the threat was reduced 4 da aa m gndeam c poate vin i stau toat vara toat toamna tii yeah I was thinking that maybe they come and stay the whole summer, autumn, you know

In extract 74 Maria has been telling how she managed her final lesson during practicum. At this point she mentions one important aspect in her story, namely time management.
Excerpt 74 (Constana corpus) 1 Maria: tii c data trecut am am trecut peste deci cum s spun eu am you know that last time I skipped over so how shall I put it, I have 2 nu mi-am calculat timpu ca lumea tii I didnt estimat my time properly, you know 3 Iulia: mhm mhm 4 Maria: i am intrat pe timpu ( ) i acuma acuma zic and I took ( )s time and so, I say to myself 5 hai s-mi pun mai puine activiti tii I should choose fewer activities, you know 6 ca s m ncadrez n timp s nu m trezesc iar c n-am timp to finish on time and not end up running out of time again 7 Iulia: mhm mhm 8 Maria: i mi-am pus prea [puine and I chose too few 9 Iulia: [aoleu i ce? i s-au terminat prea repede? ay and what happened? they were over too soon? 10 Maria: mi s-au terminat bine cu vreo dou trei minute nainte tii they were over in time two or three minutes before the break, you know 11 Iulia: mhm mhm

In lines (1) and (2) she mentions an instance when she had problems with time management as she had not allocated enough time for the activities she intended to do but because she cannot be sure that Iulia knows of that experience she checks Iulias knowledge with stii c data trecut am am trecut peste deci cum s spun eu am nu mi-am calculat timpu ca lumea stii (you know that last time I skipped so how shall I put it I didnt estimate my time properly you know). In line Iulia (3) confirms receipt of information and then Maria moves on to telling about her final lesson when she chose fewer activities lest she should run out of time again. You know/Stii in line 5 occur in direct quote which provides one of the main aspects of her story namely finding solutions to problems related to time management in teaching, and through which Marias story may become understood as an instance of a more general 143

situation. The third instance of stii you know follows another piece of new information in line 10 where she concludes that this time she has been successful in managing teaching time. Again Iulia acknowledges receipt of information by producing a minimal response in line 11. We have seen that stii you know brings about information states in which the speaker knows of speaker/hearer shared knowledge. Moreover it turns out with great regularity that stii you know does not work alone in these transitions. An important part of these transitions is for the hearer to acknowledge his receipt of information either by affirming that piece of information using yeah, mhm, aha or other minimal responses showing agreement or by marking its reception with oh. Consider the following example:
Excerpt 75 (Bucharest corpus) 1 B: da da trebuie s m hotresc de fapt ce fac pentru c am: aa cam o turm de iepuri yes, but Ive got to make up my mind what Im going to do actually, cause Ive got about a herd of rabbits 2 fiecare dat drumu-n alt direcie every time running in different directions 3 A : aa eram i eu I used to be like that too 4 B : i trebuie s m hotresc pe unde s-o iau tii and Ive got to make up my mind which way to go, you know 5 A : aa eram i eu I used to be like that too

When such markers of receipt of information are not provided, the speaker is likely to repeat that piece of information or to try again with a different bit of information (as in the initial part of extract 74: lines 1 and 2). This suggests that you know is complementary in function to the marker oh or to minimal responses which display the addressee as an information recipient, i.e. one that is actively involved in the production of talk through the process of receiving information. Because you know induces the addressee to act as an information recipient, it has the complementary function of ratifying the speaker as an information provider whose provision of information is contingent upon addressees reception. This allows us to understand why you know prefaces background information. Speakers may need to introduce background material before an upcoming narrative event will make sense to their addressees. Bracketing such material with you know marks its special status as to-be-shared information, as well as the speakers dependency on addressees reception of that information prior to his continued role as information provider. Returning to excerpt 74, Marias point regarding the difficulties involved in managing time while teaching, may not be understood unless Iulia is supplied with information about a situation when too little time is allocated for too many activities. Thus Marias status as information provider depends on her ability to establish that information as shared. She does this by prefacing it with stii (you know) in line 1, and when Iulia delays her receipt of information, she repeats that bit of information in line 2, trying again to get it acknowledged. Maria then continues to tell her story in her role as information provider. The excerpts analysed above illustrated the one of the discourse functions displayed by stii you know and similar token tags: marking meta-knowledge with regard to what the speaker and the addressee share. Another discourse function is to signal meta-knowledge about what is generally known, as excerpt 76 illustrates. 144

Excerpt 76 reproduces the last two lines of excerpt 70 on complementary medicine and herbal remedies
Excerpt 76 (Constana corpus) 1 Alina: da spirulina asta cre c e a ajuns aa un fel de ap sfiinit tii? yeah, the spiruline has come to be some sort of holy water, you know? 2 Iulia: e vedet e medicamentul vedet its a star its the star medicine

The same function is performed by other tag questions such as nu-i aa (isnt it) or its shorter form nu (no), and nelegi (you understand). Consider the following examples:
Excerpt 77 (Constana corpus) 1 Iulia: deci ea care vorbete de metodic cum poa s zic she, whos talking about methodology, how can she say? 2 s-mi dea o clas pe care eu nu o cunosc nu? to give me a class that I dont know, right? 3 Maria: [ca lecie final as a final lesson 4 Iulia: [ca lecie final as a final lesson 5 nu e posibil aa ceva thats just not possible 6 i mine o s-i zic mi pare ru doamn da nu exist aa ceva and tomorrow Im gonna tell her Im sorry madam but thats impossible 7 cum s-mi dai mie- nu? spune i tu how can you give me- right? dont you think so? 8 i o s vorbesc i cu doamna kaiter i i spun i ei nu? and Im gonna speak to Mrs. Kaiter and tell her too, right? 9 i cu asta basta da nu e normal and thats that its just not right Excerpt 78 (Constana corpus) 1 Alina: dac-ai tii cum e do:amne bietu copil if you knew how he looks God poor child 2 Iulia: ca un pianjen like a spider 3 Alina: da yes 4 Iulia: i culmea c e superngrijit i stau o grmad [de capu lui and, on top of all, hes more than taken care of, and there are a lot of people looking after him 5 Maria: [da snt m unii copii aa snt nu? yeah, some kids are like that, right ?

145

The excerpts analysed above show tag questions such as nu-i asa isnt it and nu no and discourse markers like stii you know and intelegi you understand as performing important interactional functions in both arguments and narratives. Arguments involve an asymmetric distribution of knowledge and/or opinion. These (token) tags allow the speaker to check on how the discourse is progressing away from this initial distribution. It also appeals to shared knowledge as a way of converting an opponent to ones own side in a dispute, achieving thus consensus. In narratives, they help the addressee filter through the story and select what is important for understanding the narrative point and thus they enable the addressee to respond to the story as an audience. Responding to a story as an audience implies showing interest in the speakers story and establishing bonding to the discourse and to each other (Schiffrin 1987). We have already seen that achieving consensus, negotiating and building a relationship that should be in the form of support and closeness are major traits of womens conversational style. Thus we expect women to show a preference for employing such (token) question tags in both arguments and stories. This expectation is confirmed by the interactional behaviour of our informants. Our adolescent informants show a preference for only one tag question, namely stii you know and employ it in stories only as a marker that invites common ground. As they approach adulthood their use of such tag questions becomes significantly more diversified and elaborate to the extent to which they make consistent use of other members of the same category which they employ not only in stories and in making generalizations but in arguments as well. The use of these tag questions in this last environment proves that Romanian women attach growing importance to achieving consensus by converting the addressee to their own side in an argument. This enables them to avoid potential disagreement. 5.2.1.3.2.3. Achieving alignment with Addressee Brown and Levinson (1987) have shown that claiming common opinion with the addressee can be achieved in two ways: by expressing agreement and hiding disagreement. It has become apparent that token agreement and hedging opinion, two strategies proposed by Brown and Levinson, are instrumental in avoiding disagreement. The same effect may be achieved, although indirectly, by employing various techniques of persuasion. Providing a convincing argument is a means of converting the addressee to ones point of view, avoiding thus potential disagreement. In what follows we will focus on two kinds of strategies that were particularly common in our data: constructed dialogue (i.e. directly quoted speech) and repetition. We have already considered these strategies and their role in making the story more interesting and expressing agreement. However, the analysis of our corpus reveals their multifunctional nature: in certain environments repetition and constructed dialogue are simple yet effective ways of achieving alignment with the addressee. When performing this function they become ways of avoiding disagreement28. Constructed dialogue (i.e. directly quoted speech) It has become apparent from recent research that constructed dialogue (or directly quoted speech) can perform several actions. We have considered above its potential in increasing both the speakers and the addressees interest with the story. In what follows we will consider its potential to provide evidence while reporting a previous utterance (Holt 1996). According to Clark and Gerrig (1990: 764), quotations (particularly direct ones) are a type of demostration because they differ from descriptions
28

This section on achieving alignment with the addressee is a revised and considerably extended version of Hornoiu (2005a).

146

to the extent to which they depict their referents. Li (1986: 41) argued that a direct quote communicates a more authentic piece of information that an indirect quote in the sense that a direct quote implies a greater fidelity to the source of information than an indirect quote. Philips (1986), in an analysis of reported speech during a courtroom trial, concluded that the lawyers assume directly quoted speech is more accurate and consequently more reliable than other forms. As the extracts below will show, the same can be argued of talk-in-interaction where directly quoted speech is frequently used to provide evidence and convince the addressee of the validity of ones claim. Confirming previously established patterns regarding the distribution of constructed dialogue or directly quoted speech (Holt 2000), we will focus on two types of environments that are particularly common in our data. The analysis of the data reveals that directly quoted speech is recurrently associated with two speech acts: making complaints and recounting amusing incidents. These environments generally revolve around an interaction. (i) Constructed dialogue in making complaints It has been shown that once placed in a new context, directly quoted speech takes on a dual role 29: it reproduces a former utterance for the current recipient and at the same time it supplies evidence of it. Thus reported speech is simultaneously a report of a previous locution and part of a new sequence used to perform a new speech act. This, and the fact that complaints often focus on what someone previously said, make directly quoted speech particularly appropriate for use in making complaints. In excerpt 79 the complaint focuses on something that a third person said, and constructed dialogue or directly quoted speech is used to convey the reprehensible comment.
Excerpt 79 (Constana corpus) 1 Iulia: ai fost la croitoreas? have you gone to the dressmakers? 2 Maria: e:h am fost pe naiba to hell Ive gone 3 Iulia: de ce? [p:i n-ai zis ca ] te duci astazi why? didnt you say youd go today? 4 Maria: [p:i ( ) ]nu ti-am zis c-am sunat well, havent I told you that I called her 5 Iulia: eh i:? ce-a zis? and? what did she say? 6 Maria: am fost <bolNA::v> Ive been <ILL> 7 Iulia: da ce-a avut? what was wrong? 8 Maria: e::h a avut pe dracu ( ) a avut de-o luna de zile eh, she didnt have shit ( ) shes been ill for a month 9 Iulia: n-a avut clieni she had no clients 10 Maria: A:re prea muli clieni shes got too many clients 11 Iulia: are clieni?
29

According to Voloshinov (1971: 149), reported speech is speech within speech, message within message, and at the same time also speech about speech, message about message.

147

shes got clients? 12 Maria: <s ia clieni= prea: muli i p-orm nu le face fa> she takes on too many clients and then she cant make it 13 Iulia: =i and 14 e:h pi da e neserioas eh, well, then shes not reliable

In line 6 Maria uses constructed dialogue to convey the dressmakers comment ( am fost bolnava: I was ill) which is the focus of the complaint. The device shows the addressee what happened as well as it gives evidence of it. Thus the addressee is given direct access to the source of complaint and can judge its reprehensibility for herself. At a superficial examination of the locution, Maria seems to describe the incident in an objective way by simply recalling what happened. However, as recent research has shown, verbatim reporting is unlikely to be possible (Buttny 1997; Gunthner 1998; Holt 2000). Even if reported accurately, speech uttered in one context and repeated in another is fundamentally altered. In most cases, speakers do comment on the utterances they report while simultaneously appearing to reproduce them accurately. What is expected and even claimed to be a faithful reproduction without alteration or interpretation on the part of the reporting speaker is often accompanied by prosodic devices meant to communicate affect and even to undermine the reported utterance. Closer examination of excerpt 79 reveals that Maria adopts this strategy. There are substantial clues that inform the recipient of the nature of Marias reaction to the reported comment. First, the story begins with a gloss in line 2 (e::h am fost pe naiba the hell Ive gone ) which answers Iulias question in line 1 (ai fost la croitoreasa have you gone to the dressmakers?) and which indicates that the telling will be a complaint. Second, the exaggerated prosody of the reported utterance (line 6) which belongs to the current speaker rather than the reported one conveys the formers attitude towards the reported utterance. Thus the dramatic step up in volume and pitch on bolNA::va, the stress on the last syllable, the lengthened vowel as a well as the smile voice used to recall an utterance presumably delivered in a serious manner was all these features contribute towards implicitly communicating her disbelief. This, coupled with a slow tempo in line 12 and the stress on dracu, A:re, prea, iain lines 8,10 and 12 convey her growing irritation with the dressmakers excuse. By making it sound like a blatant lie, she invites the addressee to also find it so. Notice that Maria avoids any explicit assessment of the reported speech. It is the recipient who prompted by the speakers implicit attitude towards the comment in question offers such an assessment in line 14 ( e:h pi da e neserioas eh, well, then shes not reliable). Excerpt 80 evinces a slight departure from the expected pattern. As in the previous extract, the complaint focuses on something that a third person said. This time however, directly quoted speech is not used to convey the reprehensible comment but to give voice to mothers thoughts and worries revolving around the reprehensible behaviour. The excerpt belongs in a conversation between two workmates in their forties and recounts how the daughter of one of the interactants pretended to be ill so that she could go to her mothers workplace.
Excerpt 80 (Bucharest corpus) 1 A: cnd m-a sunat brbati-miu intrase n cas i se vita d: burt nu tiu ce when my husband called, she had just got it and was complaining about her stomach, and things 2 m s nu fie apendicita s dau de alte belele

148

Im afraid it could be appendicitis and Id get into some more trouble i m-am dus la ea c nu m-am gndit c se plnge i ( ) so I went to her cause I didnt think she was just whining 4 mi f-i un ceai de ment make some mint leaves tea for her 5 zic trebuie s fie c a mncat pepene rece de la frigider p stomacu gol I thought she must have eaten cold melon straight from the fridge on an empty stomach 6 i s-ar putea de la asta s fie. and this might be why. 7 i (.) de unde ! (.) se certase afar c i-a fost puintel ru not in the least! she had been fighting, besides that she had been feeling a bit sick 8 B: ( ) 9 A: se certase p-afar cu copiii pesemne i se plictisea she must have been fighting with the kids outside, and she was getting bored 10 i avea chef s vin la m-sa la servici and she felt like dropping by her mothers work place 11 i a gsit o metod c-i ru copilului and she came up with the thing that shes sick 12 i s-a fcut c o doare burta s vin la m-sa la servici ( ) and she pretended she had stomachaches, so that she could come to my work place 13 B: a DRAC de copchil ce i-a trecut prin cap what a DEVIL of a child, to cook this up 14 A: uit-te i tu uit-te i tu s n-o strngi de git ? fancy that! Fancy that! Dont you feel like killing her? 15 B: c te bag i-n speriei she is giving you the willies 16 A: da m-a bgat n speriei D-ABIA asear cnd am ajuns acas but she really gave me the willies last night when I got home 17 B: i-a spus ce: what did she tell you? 3

The utterance that constitutes the source of the complaint (line 1) is an indirect third-person report. The current speaker moves on to directly quoted speech in lines 2 and 4-6 where she voices her and her husbands concern about their daughters health. These verbatim quotes appear to describe the incident in an objective way and give the recipient direct access to, and evidence of what triggered the complaint. Again, the details regarding the reprehensible incident are recalled with no explicit evaluation and assessment (lines 7-12). However, there are clues that convey her attitude and make clear what kind of response will be appropriate. Constructed dialogue in lines 2 and 4-6 coupled with the step up in pitch in line 2 supplies evidence for the fact that she had every reason to be worried. In contrast to what she feared, the details in lines 7-12 and the prosody accompanying them make the incident appear a trivial matter. This, then, adds to the reprehensible nature of the incident and implicitly communicates the speakers attitude to it and invites a particular response from the recipient. As in the previous extract, explicit assessment of the reported utterance and the incident in question is made by the recipient in the next turn (line 13). The reporter agrees to the recipients assessment in line 14. The assessment of the incident as reprehensible is carried on by the recipient in line 15 which takes on a duality: it can either be a continuation of the assessment in line 13 or it can be designed as a sequel to the utterance in line 14 with which it forms a collaboratively built sentence. This second interpretation seems to be more appropriate since the recipient makes use of a similar 149

strategy in line 17 where she delivers her turn in anticipatory completion of the current speakers turn in line 16: what the recipient says in line 17 is semantically and syntactically consistent with what went before in line 16. Thus the end of this extract shows both the reporting speaker and the recipient as reacting in the same way independently and producing a highly collaborative and affiliative sequence. (i) Constructed dialogue in amusing stories In instances when constructed dialogue is used to report amusing stories, the prosody of the reported utterance(s) can, at least in part, belong to the current speaker. An utterance that is likely to have been delivered in a serious manner, may later be reported with a smile voice or interspersed with or followed by laughter, which signals the current speakers attitude towards the reported comment. Excerpt 81 is a case in point. In Extract 81 Maria describes her final lesson during practicum recalling how she intended to make use of a photograph of Thomas Edisons as part of a lead-in exercise.
Excerpt 81 (Constana corpus) 1 Maria: trebuia s le fac ceva cu invenii i ce- trebuia s le introduc cuvintele astea tii= I had to prepare something with inventions and what- I had to introduce these words you know 2 Iulia: =aa yeah 3 Maria: i cum s-i fac ( ) i le-am pregtit o poz d-asta cu thomas edison and how could I make them ( ) and I chose a picture with thomas edison for them 4 Iulia: mhm mhm 5 Maria: i [cu and with 6 Iulia: [cu becul inventatorul becului with the bulb, the inventor of the bulb 7 poftim here you are 8 Maria: i m rog era o poz d-aia proast nu tiu ce and, any way, it was a poor picture, whatever 9 mersi thanks 10 Iulia: aa so 11 Maria: eu m-am gndit m::: contam pe pe faptul c ei n-o s [recunoasc poza i cI thought I was counting on the fact that they wont recognize the picture and that12 Iulia: [i ei au recunoscut [i and they recognized it and 13 Maria: [nu no 14 Iulia: i-au [stricat scenariul they ruined your scenario 15 Maria: [NU:: stai s vezi i porm urma s le arat becul NO, wait and see. And then I was supposed to show them the bulb 16 i s zic care-i cine-i sta and say Whos- whos this

150

17 Iulia: mhm mhm 18 Maria: p:i sta-i inventatoru [becului well, this is the inventor of the bulb 19 Iulia: [la care a apasat pe buton ((laughing)) the one who pushed the button 20 Maria: aa aa right right 21 Iulia: i s-a corentat and got electrocuted 22 Maria: i nu spun c a a art poza art poza aa: prima dat and believe me I I show the picture like that, the first time 23 cine-i sta hai s facem un guessing game nu tiu ce whos this lets play a guessing game whatever 24 Iulia: mhm mhm 25 Maria: i se trezete unand somebody comes up with 26 una din spate mi spune sta seamn seamn <cu cu taic-meu> some girl at the back tells me this looks like like <my father> 27 Iulia: ((laughing)) n englez in English 28 Maria: aa i porm una se trezete pi cic right and then another girl comes up with 29 asta-i thomas edison care a inventat becu thats thomas edison who invented the bulb 30 zic mersi mi-ai stricat tot jocu ((laughing)) I say thanks youve ruined my game 31 [ce sa mai fac? what was I supposed to do next ? 32 Iulia: ((laughing)) [haide m:: come on

In lines 26 and 29 she directly quotes two students thus giving Iulia access to their inappropriate reaction. It seems highly unlikely that the laughter interspersing the utterance in line 30 is part of the original prosody especially since the dramatic step up in pitch at the beginning of the utterance and the stress on mersi and stricat give it a disapproving tone matching the incident which spoilt her lead-in. However, the prosody and the accompaning laughter convey the speakers attitude and invite a particular kind of response from the recipient. As Jefferson (1979) pointed out, current speaker can invite laughter from the addressee by including it in, or immediately after, the current turn, thus initiating a sequence of laughing with. Marias laughter in lines 30 and 31 points towards recalling the event as amusing and indicates to Iulia that laughter is an appropriate response. In lines 31 and 32, the interlocutors response to this amusing incident is delivered in overlap creating thus a highly collaborative and affiliative sequence. It has been shown that directly quoted speech often forms the punchline in amusing stories and anecdotes (Bauman 1986). Sacks (1978) argued that punchlines in amusing stories or jokes must be implicit since the recipient must be allowed to interpret the humorous meaning out of the punchline. Thus the reason why jokes and humorous anecdotes usually climax in a directly quoted utterance is that such utterance gives the recipient direct access to the amusing locution apparently without any 151

shades of interpretation, assessment or rephrasing on the part of the speaker. The recipient is not told why an utterance is funny but is enabled to find it so. Repetition In addition to the functions of repetition discussed above (see Section 5.2.1.3), functions that allow both speaker and addressee to express common opinion, common point of view, another type of repetition in all-female discourse is the exact or slightly varied repetition of ones own words within the same turn or across several turns used with a view to persuading the addressee of the correctness of the speakers argument. Thus, achieving alignment becomes a way of avoiding disagreement. In excerpt 82 Rodica seeks her interlocutors endorsement of a generalisation. Previously in the conversation Rodica has expressed her dissatisfaction towards the medical system in Romania. As she does not know whether this is a position Anca is aligned with, she offers one of her acquaintances experience as evidence that would support her point, and in so doing she makes use of directly quoted speech.
Extract 82 (Constana corpus) 1 Rodica: i spunea asta ce nenorocire e n spital cum trebuie and she said that what a tragedy there is in hospitals how you have 2 de la poart trebuie s stai cu banii-n mn from the gate on you have to give them the money 3 portar asistent infirmier tot tot doorman, nurse, everybody everybody 4 a doua zi la fel portar asistent infirmier tot tot the next day, the same: doorman, nurse, everybody everybody 5 altfel nici nu se uit otherwise, they dont even look at you 6 Anca: de ce? ca [s intri? why? to get in? 7 Rodica: [ca s intri c orele de vizit nu tiu au ei un interval to get in, cause the visiting hours I dont know they have a time period 8 i cred c snt anumite zile mai ales la ginecologie tii si cred c sint anumite zile mai ales la ginecologie stii 9 c probabil acolo e internat cause shes probably an in-patient there 10 i trebuie s le umpli buzunarul i mai mult mai puin de zece mii nu suport and you have to stuff their pockets with money and more they wont accept less than ten thousand 11 spunea c au cheltuit o grmad de bani she said they spent a lot of money 12 n primul rnd banii chirurgului apoi banii anestezistului first, the money for the surgeon, then the money for the anaesthetist 13 Anca: unde mi? aicea-n constana? where? here in constanta? 14 Rodica: da:: deci stia snt pltii separat cel care opereaz cel care o s-o aib n grij i cel yeah, so these are paid separately: the one who operates, the one whos gonna take care of her and the

152

15 Anca: anestezistul the anaesthetist 16 Rodica: i anestezistul da plus asistentele the anaesthetist too, plus the nurses 17 io tiu de la mtu-mea de ast var I know it from my aunt last summer 18 ca s-i schimbe perfuziile s-i schimbe sonda i aa mai departe in order to give her intravenous nutrition to change her probe and so on 19 trebuia s le dai bani you had to give them money

While quoting in lines 1-5 her acquaintance who argued that she had to tip hospital porters and nurses, Rodica makes use of the first self-repetition to draw her addressees attention to a piece of information that is important for understanding why the story is being told. Consequently, Anca asks a question in line 6 (ca s intri, in order (for them) to (let you) enter) which then Rodica ratifies by repeating the subordinate clause of purpose and by incorporating it in her own story. After restating in line 10 (through paraphrase) the idea that one has to tip medical staff ( i trebuie s le umpli buzunaru, and you have to stuff their pockets with money), she carries on with her story introducing a related idea in lines 11 and 12, tipping doctors (banii chirurgului, banii anestezistului the money for the surgeon, then the money for the anaesthetist ), which is then expanded in line 14 ( tia snt pltii separat cel care opereaz cel care o s-o aib-n grij so these are paid separately: the one who operates, the one whos gonna take care of her). From this point on, Rodicas argument is structured by a series of selfrepetitions and paraphrases as each turn-constructional unit picks up a word or a phrase from a previous unit making thus the fabric of conversation and contributing to the main point of her argument. In line 15, Anca signals her active listenership by completing Rodicas utterance. Her contribution is ratified by Rodica in line 16 where she repeats the noun phrase ( anestezistul, the anaesthetist). In order to support her point, that one has no choice but to spend a lot of money on tipping medical staff, she gives further evidence in lines 17-19 where she mentions her aunts experience (ca s-i schimbe perfuziile ca s-i schimbe sonda trebuia s le dai bani in order to give her intravenous nutrition to change her probe we had to give them money). Making use of directly quoted speech, Rodica describes the grim reality to the recipient and gives evidence of it: Anca is given direct access to it and can judge its reprehensible nature. This suggests that Rodica describes the incident in an objective way, simply reporting what happened. The reprehensible utterance at the beginning of the extract and the details following it are recalled with no explicit evaluation or assessment. However, there are substantial clues that convey her attitude. First, the story begins with a gloss in line 1: ce nenorocire este in spital (what a tragedy there is in hospitals). From the outset, this indicates her attitude to the harsh reality the story depicts. Second, the design of the telling and the prosodic contour add up to the morally reprehensible behaviour of doctors and nurses. Her story is structured by a series of self-repetitions meant to obsessively convey the idea that one has to spend a lot of money on tipping medical staff. This obsessively repeated idea is voiced in a low monotonous tone that renders this fact inescapable. In this excerpt thus repetition and paraphrase bound various episodes or points within a larger conversation, operating thus as a kind of theme-setting at the beginning and forming a kind of coda at the end (see lines 2-4 and 19). The excerpt also demonstrates that repetition is instrumental in attempting to convert the addressee to ones own side in an argument which leads to avoiding disagreement. 153

The excerpts above (79-82) have shown that repetition and constructed dialogue in complaints and amusing stories are simple yet effective strategies in converting the addressee to ones own side in an argument, avoiding thus potential disagreement. Both devices are delicate ones. When telling stories, people want to convince the recipient of the correctness of their argument, they want the recipient to agree with their interpretation or assessment of the incident (for instance, that it was complaint worthy or funny). However, rather than making their assessment of the incident explicit, they resort to directly quoted speech embedded within a sequence containing implicit assessement. By making use of this strategy, the reporting speaker gives the recipient direct access to the original utterance and enables him to make an independent assessment of it. This explicit assessment of the reported utterance is commonly supplied by the recipient in the next turn. It turns out with great regularity that the reporting speaker often collaborates with the recipients response by producing a similar one in overlap. The result is that both speakers appear to react independently but also harmoniously to the reported utterance, and both are in alignment as to their assessment of it. Sometimes repetition is added to reinforce the point that has just been made. These strategies can thus be viewed as subtle ways in which intersubjectivity is established and maintained. 5.2.1.3.3. Speaker constructs shared meaning/knowledge Imagery and details Some conversational samples are characterized by a tendency towards concreteness and imageability (Chafe 1984: 1099) which lends them a sense of particularity. The examples below (excerpts 83 to 86) are illustrative in this sense. The events described are represented as scenes. The numerous details supplied by the speaker inspire the addressee to create sounds and scenes in their minds, scenes in which the described characters, objects and actions figure. Thus, it is in the individual imagination that meaning is constructed. And it is the creation of such shared meaning that turns a collection of individuals into a community and unites individuals in relationships giving thus cohesion to the group.

TEENAGERS The participants in excerpt 83 are two secondary-school pupils who are not only friends but also live in the same neighbourhood. Laura recounts how she, her mother, her brother and a friend of her brothers spent the evening playing cards and whoever lost the game received a punishment.
Excerpt 83 (Constana corpus) 1 Laura: am mai jucat un joc cu ei trei I played another game with the three of them 2 i-au zis aa cine pie:rde and they said this: the one who loses 3 a zis c dac io pierd io le iau lor un pachet de tigri tii i:: he said that if I lost Id buy them a pack of cigarettes, you know, and

154

4 Ana: ai pierdut? did you lose? 5 Laura: am pierdut da oricum nu le-am luat I lost, but I didnt buy it anyway 6 Ana: ((laughter)) 7 Laura: i am jucat juma de joc and we played half of the game 8 i p-orm am nceput s ne batem c daniel era:and then we started to fight, cause Daniel was:9 el sttea n capu mesei he was sitting at the head of the table 10 Ana: mhm mhm 11 Laura: mihai sttea fa-n fa cu mine tii mihai was sitting in front of me, you know 12 i daniel i fcea cu picioarele tii and daniel was kicking him, you know 13 i fcea cu picioarele lui mihai tii he was kicking mihai, you know 14 i f- i-a-nceput s-l njure i aa i-am nceput s ne ba:tem and he was- he started to him names, and so we started to fight 15 p-orm avea daniel un lan n buzunar tii de aicea pn aicea then daniel had a chain in his pocket, you know, from here to here 16 [tii you know 17 Ana: [mhm mhm 18 Laura: i zic io ce-i cu lanu meu i l-am rupt [i:: and I said what are you doing with my chain? and Ive broken it, and 19 Ana: [i era al tu? and it was yours? 20 Laura: nu era al lui cred no, it was his, I think 21 i ne-am fugrit i am intrat n cealalt camer i era i mihai and he chased me, and we got into the other room, and mihai was there too 22 i: a-nchis repede ua i-am zis hai s-l batem and he closed the door quickly, and we said lets kick him 23 i l-am btut chestii de-astea and we kicked him, and stuff

ADULTS What is worth pointing out and this seems to be characteristic of adults conversations as opposed to teenagers is that the interlocutors also supplement the description with their own comment providing or eliciting further details that help conjure up the images that create the scene. Excerpt 84 is part of an encounter with a beautician 30. S, the beautician, is recounting to her client how her daugher-in-law has celebrated her birthday. Notice how both the details supplied by the beautician throughout the excerpt and those supplied by the client in lines 5 and 15 contribute
30

For an analysis of all-female Romanian conversational discourse in service encounters, see Hornoiu (2004).

155

significantly to the construction of a scene that is familiar not only to the speaker but to the interlocutor as well.
Excerpt 84 (Constana corpus) 1 S: era mndr c de ziua ei i-a luat cosmin o pereche de cercei de argint mari aa she was proud because cosmin bought her a pair of large silver earrings for her birthday 2 C: a::::h a ::: 3 S: cnd i-am vzut la (.) ureche zic io nu port aa i era numai cu argint when I saw them on her ears I said I dont wear this and she was covered in silver 4 da verigheta lor este combinaie de aur cu aur alb but their wedding ring is a combination of gold with white gold 5 C: mhm ce frumoas mhm, how beautiful 6 S: i merge i la argint i la aur da i avea o groaz de brri pe mn and it goes with both silver and gold yes, and she had a lot of bracelets on her wrist 7 cerceii ia mari mi-a luat cosmin de ziua mea cosmin got me those large earrings for my birthday 8 c pe nou februarie a fost ziua lu ctlina pe cinpe a lu cosmin cause on the ninth of February it was Catalinas birthday, and on the fifteenth it was cosmins 9 i acum pe doutrei a lu felicia da i::: and now on the twenty-third its Felicias, yeah and 10 (4) 11 i luase a venit c-o sear nainte naa de cununie a lor i i-a adus o bluz she had brought her their matron of honour came a night before and brought her a blouse 12 foarte frumoas cu fir prin ea aa very beautiful, purled like that 13 (4) au fcut la clubu unde lucreaz cosmin au serbat amindoi pe:::: smbt the party was at the club where cosmin works they celebrated both their birthdays on Saturday 14 pe paipe on the fourteenth 15 C: de valentine on valentines day 16 S: da au fcut platouri reci le-am fcut io un tort de fructe le-am dat yes, they served a cold buffet, I baked them a fruit cake, I gave them

The protagonists in excerpt 85 belong in different age groups and have different educational background. A is a teacher in her sixties while B is a nurse in her forties and she regularly helps A with the housework. Despite this assimetry in social status, their conversational exchange resembles a conversation between close friends displaying most of the positive politeness strategies analysed so far that act as social accelerators in friendly talk.
Excerpt 85 (Bucharest corpus) 1 A: i CUM a fost la brila la: VIe? and how was it at braila at the vineyard ?

156

2 B: no::::c dac v povestesc facei un roman well if I tell you you can write a novel 3 A: ha ha ha poate c m pregtesc s scriu romanu ha ha ha maybe I am going to write that novel 4 B: am plecat joi deci joi a fost o sptmn de cnd am plecat la ar we left on Thursday, so on Thursday it was one week since wed gone to the countryside 5 nu m-am dus [la servici joi I didnt go to work on Thursday 6 A: [cu copii cu tot did you take the children as well 7 B: pi da i zic de diminea las i eu casa curat terg prafu well yes, and I say to myself in the morning Ill clean the house, dust it 8 n-are rost s m mai duc la servici fac puin mncric pentru doru theres no point in going to work. Ill cook some food-dim. for doru 9 el rminea acas he stayed at home 10 i am fcut piaa pentru la ar c tiam c: and I did the shopping for those in the countryside, cause I knew that 11 A: da CE le ducei la ar a : zahar ulei chestii d-aste nu but what do you take them at the countryside? sugar, oil, stuff like that, right? 12 B: am cumprat salam I bought some salami 13 A: a : a :::: 14 B: crnai crenvurti [.] sausages 15 i brnz CE-am mai luat pate de ficat and cheese what else did we get? liver pt 16 deci aproape cinsute de mii am dat numai pe mncare so I spent about five hundred thousand on food alone

B recounts a journey in the countryside as she is cleaning the windows. The story is elicited by A in line 1 where she uses a news-up-date (i CUM a fost la brila la: VIe? and how was it at brila, at the vineyard?), a speech act characteristic of small talk and particularly frequent in all-female conversational discourse. Bs answer in line 2 (no::::ca dac v povestesc facei un roman) prefaces and shows her as being oriented towards a narrative rich in details. Similarly, A is equally involved in the construction of a scene that evokes familiar experiences by supplying her own comment that either provides or elicits further details in lines 6 and 11. Excerpt 86 is a continuation to excerpt 85. B carries on with her description of the journey in the countryside. The recipients own comment providing further details in lines 2, 5, 8 and 11 or eliciting further details line 4 shows her as being equally oriented towards the construction of shared meaning.
Excerpt 86 (Bucharest corpus) 1 B: i-am ajuns la nou i jumate n cireu and we got to ciresu at nine fifty 2 A: seara

157

in the evening 3 B: i de la nou i jumate ia-o p JOS and from nine thirty we started walking 4 A: i::: e drum aa: sau mergei peste cmp aa and is there a road, or you go across the field? 5 B: e: asfalt its paved 6 A: e asfalt its paved 7 B: da dar ntuNEric. Nu e: deci nu e SAT cmp cmp cmp yes, but it was dark theres no no village, just fields and fields all around 8 A: o sosea: n [cmp a road in the field 9 B: [snt cteva stne de oi. vreo patru stne there are a few sheep folds about four sheep folds 10 se-auzeau cinii [ltrnd you could hear the dogs barking 11 A: [cinii alt belea the dogs, another trouble 12 B: aa i-am ajuns la soacr-mea la doisprezece i un sfert and so we got to my mother-in-laws at a quarter past twelve 13 A: DOAmne god 14 B: bici p TLpi aveam my feet were in blisters 15 A: m::: Mmm

As the above excerpts have shown both the speaker and addressee are involved in the shared meaning through the use of details that conjures up images of familiar experiences. This type of coconstructed meaning reinforces the group cohesion, on the one hand, and the protagonists involvement in the conversation and the relationship, on the other. The fact that the addressee takes part in the creation of scenes evoking familiar worlds seems to be a characteristic of adults rather than adolescents conversational style. General extenders The forms to be analyzed in what follows constitute a class of linguistic expressions that typically occur in clause-final position and have the structure conjunction + NP. They are generally referred to as general extenders: general in the sense of being non-specific or vague with regard to their referent, and extenders because they are added to grammatically complete utterances, thus extending them31. They have been divided into two sets: adjunctive general extenders (i.e. phrases beginning with and such as and stuff, and everything, and things like that, etc) and disjunctive general extenders, this latter set including such phrases beginning with or as or something, or anything (Overstreet 1999:3)32. This is a rough classification, however, since there are occasions when the actual
31

For an extensive study on general extenders, see M. Overstreet, Whales, Candlelight, and Stuff like That: General Extenders in English Discourse. New York: Oxford US, 1999. 32 The following list, long yet not exhaustive, based on Overstreet (1999:3) provides an idea of the range of possible types of expressions that could be classified as general extenders: and stuff (like that), or something (like that); and all (that), or anything (like that); and everything (like that), or what; and blah blah blah, or whatever; and that, or what have you;

158

conjunctions are missing. Given their widespread use in both spoken and written language which suggests that they are indeed important for language users, general extenders have become the focus for extensive study and research. The focus of early research has been on the referential function served by general extenders. Thus some researchers have analyzed these expressions as items which are used to complete three-part lists (Jefferson 1990, Lerner 1994). Working in the tradition of conversation analysis, Jefferson argues that, in constructing lists, speakers and hearers in naturally occurring conversation orient to what she calls the programmatic relevance of three partednesss (1990:66) which means that participants are aware of the fact that lists not only can and do occur in three parts, but should so occur (1990: 66). As a result, according to Jefferson, general extenders (which she calls generalized list completers) are employed by conversational participants to complete three-part lists, providing thus a methodic solution to the problem of three-partedness (1990: 67). Still others view general extenders as setmarking tags (Dines 1980, Ward and Birner 1993), or vague category identifiers (Channell 1994)33. In most cases, general extenders are analyzed as indicating additional members of a list, set, or category. Those who adopt a cognitive categorial approach to general extenders argue that these linguistic expressions co-occur with a named exemplar (or exemplars) whose characteristics enable the hearer to infer additional or alternate members of the category the speaker has in mind. Following this approach, a noun phrase such as apples and stuff might be analyzed as having the same referent as the category fruit. There are several reasons why a speaker may choose to use a general extender rather than to refer to the category by name. First, by naming an item (or items), and then using a general extender to implicate a lexicalized category, speakers can refer to a category whose name they either do not know, or cannot recall. Second, in using a general extender a speaker can identify members of a category for a hearer who may be unfamiliar with the category, or with its name. Finally, naming a specific exemplar and indicating more with a general extender allows a speaker to emphasize or highlight certain members of the category (Overstreet 1999:45). However, it has become apparent that in actual usage general extenders do not appear to be used with category implication as their most obvious function. Rather, when viewed in terms of their role within the interactive exchange of talk, general extenders appear to take on a new dimension that is interpersonal and tied to the nature of the social relationships holding among participants. In natural conversation general extenders are not so much list completers or category identifiers as markers of shared knowledge and experience, or markers of the speakers attitude towards the message expressed, or towards the addressee. In what follows the focus of our analysis will shift from a concern with the cognitive and the categorial to the role general extenders play in creating inter-subjectivity34

and the like, or anyone (like that); and such, or anybody (like that); and what have you, or someone (like that); and so on, or somebody (like that); and so forth, or someplace (like that); and whatnot, or somewhere (like that); and the rest; and this and that; and whatever; and you name it; and the whole kit and caboodle; and the whole nine yards; and the whole bit/thing; and (all) {this/that}; and (all) (this/that) {sort/kind/type} of; {crap/thing/jazz/junk/mess/nonsense/shit/stuff}; and {crap/things/junk/shit/stuff} (like this/that); and {business /crap/things/junk/shit) of (this/that}; {kind/sort/ilk/nature}; et cetera. 33 If treated as examples of vague language they take on negative connotations. In one reported study, Dines (1980) found that middle-class judges in Australia associated the use of general extenders with working-class speech. Such forms are stigmatized because they are assumed to reflect vague and inexplicit speech (1980: 19). This negative value attached to the use of general extenders seems to come from a feeling that vagueness in reference stems from vagueness in thinking, and hence stupidity. 34 Inter-subjectivity is the process whereby participants can reach similar interpretations of the world despite the fact that individual mental worlds are necessarily distinct and no two individuals will share identical concepts; this achievement is ascribed to an assumption of shared knowledge and a shared world (Husserl 1929/ 1977: 92; Schegloff 1992; Schiffrin 1990, 1994a); social actors behave as if the external world is sufficiently the same for them as it is for others.

159

or reciprocity of perspectives (see Cicourel 1974, Schegloff 1992, Schutz 1962, Schutz and Luckmann 1977) by invoking shared knowledge and shared experience. According to Cicourel (1974), the reciprocity of perspectives is an interpretive procedure basic to all interaction consisting of two parts: the first part instructs the speaker and addressee to assume their mutual experiences of the interaction are the same even if they were to change places; the second part informs each participant to disregard personal differences in how each assigns meaning to everyday activities, thus each can attend the present scene in an identical manner for the practical matter at hand. (1974: 34) Given these operational procedures for social interaction, it is not surprising that talk-in-interaction contains expressions that are conventionally used to invoke an assumption of shared experience or inter-subjectivity. Some of the most easily recognizable of such expressions are discourse markers 35 like I mean, you know what I mean, or simply you know. Recently researchers have drawn attention to the similarity between discourse markers and general extenders with regard to their potential to create shared knowledge36. As part of her analysis of you know, Schiffrin argues that once we acknowledge that the information state is verbally displayed (as it is in the extracts analysed in this section) we may also assume that the marker by which such verbal displays are solicited functions not merely as a cognitive marker, but as an interactional marker (1987: 273). Following this description, the same can be argued about general extenders which are more appropriately viewed as interactional markers. Recent studies (Overstreet 1999) conducted in the framework of conversation analysis emphasize this interactive or interpersonal function served by general extenders to dispel the pervasive notion that the only function served by general extenders is the referential one proposed by studies of categorization. Although they can be used on occasion as category implicative devices in shared referential worlds, this function is not incompatible with their status as interactional markers of intersubjectivity in shared social worlds. On the contrary, it has become apparent that they are used with greater frequency when performing this latter function. Extracts 87 to 96 are illustrative of the wide variety of general extenders used in both teenagers and adults conversational styles in Romanian to mark an assumption of shared knowledge and experience37. Although the focus of this chapter is on the qualitative rather than the quantitative analysis, it is worth noting that the two age groups under scrutiny differ both with regard to the range of the general extenders used and the frequency with which speakers employ them: adult speakers use general extenders not only with increasing frequency but also display a wider repertoire of such linguistic expressions. It has been argued that the type of shared experiential knowledge that the addressee needs in order to infer what the speaker may implicate in using a general extender ranges along a continuum
35

For an extensive study on discourse markers, see Schiffrin (1987). It is worth noting here that although related functionally, there are important differences between discourse markers and general extenders. Unlike discourse markers, which represent a disparate list of items, belonging to different word classes (Schiffrin 1987: 40), general extenders are a relatively homogeneous set of forms consisting of a conjunction ( and or or) plus a noun phrase. Discourse markers function parenthetically and are therefore independent of sentence structure, while general extenders are attached to utterances so as to become syntactically consistent with what precedes them, thus general extenders are part of sentence structure. Finally, whereas several discourse markers (you know, I mean, oh, like ) can occur quite freely within a sentence at locations which are very difficult to define syntactically (Schiffrin 1987: 32), general extenders typically occur in clause-final position (Overstreet 1999:12). However, research has shown that certain forms, such as and stuff, may undergo a process of becoming more flexible with regard to their position in the sentence (Overstreet 1999). 36 In fact, the close co-occurrence of you know and I mean with general extenders supports the idea that there is some connection or shared function among these forms. 37 This section on general extenders in all-female conversational discourse in Romanian is a revised and extended version of Hornoiu (2006).

160

from broad, cultural, or general knowledge potentially shared by many, to knowledge that is shared only by a small number of interactants (Overstreet 1999:69). Excerpts 87 and 88 illustrate the use of general exteders that implicate broad cultural knowledge shared by many. In extract 87 Rodica is expressing her dissatisfaction towards the medical system in Romania quoting an acquaintance who argued that in order to receive proper medical care one has no choice but to tip medical staff. In order to infer what is implicated by the general extender tot tot (everybody everybody) in lines 3 and 4, the addressee is assumed to recognize the named items, portar asistent infirmier, as members of the medical staff.
Excerpt 87 (Constana corpus) 1 Rodica: i spunea asta ce nenorocire e n spital cum trebuie and she said that what a tragedy there is in hospitals how you have to 2 de la poart trebuie s stai cu banii-n mn from the gate on you have to give them the money 3 portar asistent infirmier tot tot doorman, nurse, everybody everybody 4 a doua zi la fel portar asistent infirmier tot tot the next day, the same: doorman, nurse, everybody everybody the next day, the same: doorman, nurse, everybody everybody 5 altfel nici nu se uit otherwise, they dont even look at you

Excerpt 88 taken from the same conversational exchange offers a similar example. Here Rodica mentions another example (her aunts experience) that lends weight to her view that one has no choice but to spend a lot of money on tipping medical staff.
Excerpt 88 (Constana corpus) 1 Rodica: io tiu de la mtu-mea de ast var I know it from my aunt last summer 2 ca s-i schimbe perfu::ziile s-i schimbe so::nda i aa mai departe in order to give her intravenous nutrition to change her probe and so on 3 trebuia s le dai bani you had to give them money

In using the general extender i aa mai departe (and so on) after the subordinate clauses of purpose ca s-i schimbe perfuziile ca s-i schimbe sonda (in order to give her intravenous nutrition to change her probe), the speaker assumes that the addressee will infer that the general extender implicates various other ways in which members of the medical staff take care of their patients. The conventional set of ways of taking care of patients in hospitals may qualify as a kind of broad or general knowledge that the analyst, as a non-participant in the interaction, can attribute easily to the participants. Sometimes, the speaker may assume a less common form of knowledge that is shared by a smaller subgroup of the culture. In excerpt 89 the speaker is discussing the structure of her doctoral dissertation mentioning that one of the chapters will concern itself with a pragmatic approach to 161

discourse. In order to understand the function of the general extender nu tiu ce in line 3 one needs to be familiar with various concepts and topics that are central to pragmatics.
Excerpt 89 (Bucharest corpus) 1 B : ca un capitol gndesc s nu iau enunul aa ca : palier de mijloc pentru ce este sub as a chapter, Im thinking of taking the utterance as a midway for whats below 2 enun i pentru ce este peste enun the utterance and for whats above the utterance 3 pentru c o parte s fie de pragmatic a textului dialogului nu tiu ce so that a part can be about the pragmatics of the text, of the dialogue, whatever

Although it may be difficult for an individual unfamiliar with this branch of linguistic investigation to understand this excerpt, there are still a relatively large number of individuals who would be able to make sense of it, based on shared knowledge, namely that the chapter in question will probably address, for instance, such issues as implicature, speech acts or deixis. As we move towards the far end of the continuum, there may be occasions when the knowledge shared between participants seems inaccessible to non-participants. In a discussion of the architecture of intersubjectivity, Rommetveit (1974: 29) argues that similar states of nearly perfect complementarity and synchronization of intentions and thoughts may result in conversations that are cryptic to outsiders but are perfectly understood by the interactants. For the purposes of illustration, consider extract 90 which, although it is not based on a private way of seeing the world shared exclusively by the participants only, demonstrates that the interpretation of the general extender i cu draci i cu toate astea (and with shit and all the stuff) does depend on having access to some specialized knowledge shared by the participants . But for this knowledge, a non-participant would not know that Marias use of the general extender implicates a reference to Hermann Hesses work.
Excerpt 90 (Constana corpus) 1 Maria: a:::: mi am gsit s tii ah, Ive found some more, you know 2 am gsit i cu:: lupul de ste::p i cu jocu cu mrgelele i cu siddhartha Ive found some with the prairie wolf and playing with beads and with siddhartha as well 3 i cu draci i cu toate astea tii and with shit and all the stuff, you know

The activation of shared experiential knowledge as a result of the speakers use of a general extender leads to intersubjective understanding which is co-constructed by participants in the course of the interaction. As illustration, consider extract 91 which includes a continuation to excerpt 90 and also repeats the latter. The point of interest in this extract is in line 7, where Maria mentions that she found text analyses focusing on characters (presumably on ways characterization is achieved) and then uses another general extender cu nu tiu ce (literally: with I dont know what, i.e. and the rest/and the like), thus demonstrating an expectation that Iulia will recognize that she implicates other issues that are relevant in a literary analysis, i.e. setting, atmosphere, narrative technique, etc. Indeed, Iulia demonstrates such an understanding by offering an evaluation of the piece of information supplied by Maria as being mito cool. The fact that Iulia does not make use of requests for clarification, but on 162

the contrary uses two questions in lines 4 and 6 ( n german totu all in German; i comentariu de text and text analysis as well) that elaborate on the topic under discussion and an additional characterization at the end of the extract ( ce mito how cool) demonstrates that she inferred correctly the meaning of the general extenders used by Maria and this correct inference is based on intersubjective understanding.
Excerpt 91 (Constana corpus) 1 Maria: a:::: mi am gsit s tii ah, Ive found some more, you know 2 am gsit i cu:: lupul de ste::p i cu jocu cu mrgelele i cu siddhartha Ive found some with the prairie wolf and playing with beads and with siddhartha as well 3 i cu draci i cu toate astea tii and with shit and all the stuff, you know 4 Iulia: n german totu? all in German? 5 Maria: da m n german [i am gsit yeah, in German, and Ive found 6 Iulia: [i comentariu de text and text analysis as well? 7 Maria: i comentariu de text am gsit cu personaje cu nu tiu ce Ive found some text analysis as well, with characters, with whatever 8 Iulia: ce mito how cool

As markers of intersubjectivity, general extenders frequently co-occur with another expression that is often used to indicate assumed similarity of participants experience: the discourse marker tii, nelegi you know, you understand. According to Schiffrin, one function of you know is to mark general consensual truths which speakers assume their hearers share through their co-membership in the same culture, society, or group (1987: 274). Excerpt 92 illustrates the close co-occurrence of tii you know with a general extender. In excerpt 92 Andreea, a teenager, is telling her friend that she is going to her grandmother to get grsua the little fat one, her dogs puppy, because she misses her.

Excerpt 92 (Constana corpus) 1 Andreea : io m duc s-mi iau grsua astzi Im going to take my plump one today 2 m duc la bunic-mea s mnnc ceva Im going to my grandma to eat something 3 da i sracii- vezi nu tiu cre c la btrinee cnd te ataezi yeah, and poor them see, I dont know, I think that when youre old you grow fond 4 face- bine bunic-mea a-nceput cu teatru bine mam dac- my grandma goes well she started to put on an act well dear if- 5 la unchi-meu dac nu-i place cum e ngrijit de noi to my uncle if you dont like the way we take care of her

163

6 7 8

cu chestii de genu sta tii and stuff like this c s-a ataat de ea i acuma i e aa s-o dea cause she has grown fond of her and now she doesnt feel like sending her back da tii i mie mi-e dor de ea but I miss her too, you know

In lines 4 to 6 she is telling her friend that her grandmother has grown very fond of the puppy and she would rather Andreea didnt take it away. She carries on arguing that her grandmother is acting as if she felt offended by Andreeas intention: bunic-mea a-nceput cu teatru bine mam dac nu-i place cum e ngrijit de noi (grandmother started acting all right my dear if you dont like how we take care of her) and then she uses a general extender followed by the discourse marker tii you know cu chestii de genu sta tii (and stuff like this you know). The general extender evokes other ways of complaining. In using a general extender, Andreea is relying on an assumption of shared knowledge or experience (see also Aijmer 1985: 378-379; Overstreet and Yule 1997b) which is then reinforced by tii you know, a discourse marker used to seek interactional alignments by establishing shared opinion (Schiffrin 1987). As you have already seen, intersubjectivity is interactionally co-constructed by both the speaker and the addressee. If, in using a general extender, the speaker may invite an assumption of shared knowledge, which may be additionally reinforced by the use of the discourse marker you know, eliciting thus displays of understanding, the addressee, in his turn, may affirm his shared knowledge or perspective by using backchannels such as yeah, umhm, uh huh, okay, or right or phatic questions that elaborate on the topic under discussion. Excerpt 93, in which Maria is telling her friend how she planned to teach vocabulary to highschool students, illustrates this interactional strategy of co-constructing shared knowledge by both the speaker and the addressee. The general extenders, the backchannels and other instances of supportive feedback are given in bold type.
Excerpt 93 (Constanta corpus) 1 Maria: mhm c trebuia s le fac ceva cu invenii i ce- trebuia s le introduc cuvintele astea tii mhm, cause I had to prepare something with inventions and what- I had to teach these words, you know= 2 Iulia: =aa yeah 3 Maria: i cum s-i fac ( ) i le-am pregtit o poz d-asta cu thomas edison and how could I make them ( ) and I chose a picture with thomas edison for them 4 Iulia: mhm mhm 5 Maria: i [cu and with 6 Iulia: [cu becul inventatorul becului with the bulb, the inventor of the bulb 7 Maria: i m rog era o poz d-aia proast nu tiu ce and, any way, it was a poor picture, whatever 8 Iulia: aa [.] right 9 Maria: i nu spun c art poza art poza aa prima dat and believe me I show the picture like that, the first time

164

10

cine-i sta hai s facem un guessing game nu tiu ce whos this? lets play a guessing game, whatever 11 Iulia: mhm mhm 12 Maria: i se trezete una and some girl comes up with 13 una din spate mi spune sta seamn seamn cu cu taic-meu some girl at the back tells me this looks like like my father 14 Iulia: ((laughing)) n englez in English 15 Maria: aa i porm una se trezete pi cic sta-i thomas edison care a inventat becu and then another girl comes up with thats thomas edison who invented the bulb 16 zic mersi mi-ai stricat tot jocu ((laughing)) I say thanks, youve ruined my game 17 [ce s mai fac? what was I supposed to do next ? 18 Iulia: ((laughing)) [haide m:: come on

It is worth pointing out at this stage that whenever speakers use a general extender, our data do not contain any instances of requests for clarification by hearers. Instead, supportive feedback indicating that they are following the communication and questions or utterances (sometimes delivered in supportive overlap) that elaborate on the topic that is under discussion are the order of the day 38. In the above excerpt the addressee acknowledges each piece of new information either by using minimal responses as continuers (lines 2, 4, 8 and 11) or by elaborating on the topic under discussion (lines 6, 14 and 18). In using a general extender, the speaker appears to communicate the following message to the addressee: Because we share the same knowledge, experience, and conceptual schemes, I need not be more explicit; you will be able to supply whatever unstated understandings are required to make sense of the utterance. Functioning in this way, general extenders are recognizable indications that all talk is, in some sense, incomplete (see Garfinkel 1967), and that each of us expects our conversationalist partners to co-operate with us in the process of creating whatever sense of completeness is sufficient for a particular occasion. This is particularly true when speakers use adjunctive general extenders, since the more that is implicated is typically treated as known. In Brown and Levisons pragmatic approach to politeness phenomena (1987), one strategy of positive politeness is to presuppose, raise, or assert common ground with the addressee by demonstrating an assumption of shared knowledge. Thus the use of adjunctive general extenders can generally be viewed as a strategy of positive politeness whereby shared knowledge is created and solidarity is invited, much like discourse markers such as you know (see Brown and Levinson 1987, Ostman 1981, Schiffrin 1987). The examples so far illustrated the wide variety of adjunctive general extenders that can be used to mark an assumption of shared knowledge. In what follows we will focus on one particular pattern: when general extenders do not actually intimate additional instances; instead, they are simply used for the purpose of marking invited solidarity with the interlocutor, much like the form you know (see Brown and Levinson 1987, Ostman 1981, Schiffrin 1987). In excerpt 94, the general extender i aa (approximately and so on /an stuff like that) is used in a context where one speaker (Alina) is arguing that, contrary to popular belief, herbal medicines are expensive and to support her point she mentions an acquaintance who called her the
38

This phenomenon is not limited to our data set. See, among others, Dines (1980) and Overstreet (1999).

165

previous night telling that she bought her son a herbal medicine that contained a lactobacillus that regulates the small bowel flora and that she paid one hundred thousand lei for two Martian-shaped tablets.
Excerpt 94 (Constana corpus) 1 Alina: da pi e asear m-a sunat melania s-mi spun yeah, well, its melania called me last night to tell me 2 c a luat pentru copilu ei tii that she has bought for her child, you know, 3 c a luat un medicament m rog care avea that she has bought a medicine, anyway, which had 4 Iulia: aa yes 5 Alina: un lactobacilus i a dat a lactobacilus and paid for it 6 tot chipurile naturist i aa supposedly a herbal medicine and so on 7 o sut de mii pentru la mic a hundred thousand for the kid 8 c-i sub form de mini [marieni cause its shaped like mini martians 9 Iulia: [da pentru ce-i? but what is it for? 10 Alina: pentru reglarea florei intestinale to regulate the small bowel flora 11 Iulia: aha c el are probleme aha, cause hes got problems

Alina describes this medicine as being naturist i aa (approximately: herbal and so on/and stuff like that). It is obvious that Alinas general extender i aa an stuff like that/and so on is not actually intimating other herbal medicines that her conversationalist partner will be able to think of, but rather that the latter will provide supportive feedback and agree with her. Excerpt 95 provides another example when the speaker uses a general extender, this time nu tiu ce (approximately: I dont know what), as an appeal to solidarity. Here, Maria is telling Iulia how she managed her final lesson during practicum.
Excerpt 95 (Constana corpus) 1 Maria: trebuia s le fac ceva cu invenii i ce- trebuia s le introduc cuvintele astea tii I had to prepare something with inventions and what- I had to teach these words, you know= 2 Iulia: =aa yeah 3 Maria: i cum s-i fac ( ) i le-am pregtit o poz d-asta cu thomas edison and how could I make them ( ) and I chose a picture with thomas edison for them 4 Iulia: mhm mhm 5 Maria: i [cu and with

166

6 Iulia:

[cu becul inventatorul becului with the bulb, the inventor of the bulb 7 Maria: i m rog era o poz d-aia proast nu tiu ce and, any way, it was a poor picture, whatever 8 Iulia: aa [.] right 9 Maria: i nu spun c art poza art poza aa prima dat and believe me I show the picture like that, the first time 10 cine-i sta hai s facem un guessing game nu tiu ce whos this? lets play a guessing game, whatever 11 Iulia: mhm mhm 12 Maria: i se trezete una and some girl comes up with 13 una din spate mi spune sta seamn seamn cu cu taic-meu some girl at the back tells me this looks like like my father 14 Iulia: ((laughing)) n englez in English 15 Maria: aa i porm una se trezete pi cic sta-i thomas edison care a inventat becu and then another girl comes up with thats thomas edison who invented the bulb 16 zic mersi mi-ai stricat tot jocu ((laughing)) I say thanks, youve ruined my game 17 [ce s mai fac? what was I supposed to do next ? 18 Iulia: ((laughing)) [haide m:: come on

As the focus of her final lesson was teaching vocabulary related to inventors and their inventions, she used a picture of Thomas Edison as part of a lead-in exercise. In line 7 she refers to a problem she encountered because the picture was of poor quality and uses nu stiu ce as a marker of solidarity. The same item is used again in line 10 when Maria starts telling how she began her lead-in exercise art poza aa prima dat cine-i sta hai s facem un guessing game (first I show the picture to them whos this lets play a guessing game) and she uses the general extender nu tiu ce as an appeal to solidarity. It is obvious that on both occasions, in using nu tiu ce, Maria has no additional instances in mind. Instead, she appeals for understanding regarding the difficulties she encountered: a poor-quality picture and a couple of students who ruined her lead-in by making inappropriate comments. Throughout the conversation, Iulia demonstrates her co-operation and understanding with mhm in lines 4 and 11 and asa in lines 2 and 8 and provides supportive feedback by elaborating on the topic and initiating a sequence of laughing-with in lines 14 and 18. Regardless of whether the implicated concept is culturally established (e.g., ways in which medical staff take care of patients in hospitals) or is restricted to a smaller subgroup of the culture (e.g., issues that pragmatics concerns itself with), general extenders serve the same basic function: in using them, the speaker invites the addressee to act as if he or she is familiar with what is being described and expects the latter to supply whatever unstated understandings are required to make sense of the utterance. By demonstrating such an assumption of shared knowledge, the speaker emphasizes a similarity between himself and the addressee39.

39

This can be done either to reinforce existing familiarity between interlocutors, or to decrease social distance (Overstreet 1999:72).

167

Whether or not the addressee actually shares the shared experiential knowledge assumed by the speaker is of little consequence. What remains of the greatest importance is the assumption of shared knowledge that is marked by the general extender, not the fact, and that assumption is rarely challenged. Thus by treating the speakers utterance as unproblematic, the addressee reciprocally underscores the participants similarity. Indeed, raising questions of understanding regarding an utterance containing a general extender might be perceived as a breach of the reciprocity of perspectives (the assumption that the hearer will supply whatever unstated understandings are required to make sense of the speaker's utterance). Rather than affirming the participants similarity, this would draw attention to their differences and potentially increase the social distance between them. Thus general extenders are not so much category implicating devices as markers of intersubjectivity. As such they are associated with positive politeness being instrumental in establishing solidarity and common ground. 5.2.2. Fulfil Addressees want (for some X) 5.2.2.1. Give symbolic gifts to Addressee: sympathy, understanding, cooperation This strategy involves speaker satisfying the addressees positive face by fulfilling some of his wants. This goal is achieved by performing the classic positive-politeness action of gift-giving. However gifts need not tangible; symbolic gifts such as satisfying human-relation wants (i.e. the addressees wants to be liked, admired, cared about, understood, listened to, etc.) are just as appropriate. In their seminal study of politeness Brown and Levison do not offer details as to the output of this politeness mechanism. The analysis of our data suggests that giving the symbolic gifts of understanding, cooperation and sympathy may be achieved through the use of such conversational strategies as minimal responses as continuers, phatic questions, supportive overlapping talk, anticipatory completion and consistent use of imagery and details.

The use of minimal responses as continuers Minimal responses, also referred to as back-channel communication (Yngve 1970), assent terms (Woods 1988) or accompaniment signals40, are vocalization such as uh huh, mm hmm, yeah and many others as well as head-gestures such as nods. These verbal and non-verbal signals are produced by others than the main speaker and are meant to indicate a persons co-participation in a conversation. Schegloff (1982) identifies two usages of minimal responses. The most common usage is to exhibit on the part of its producer an understanding that an extended unit of talk is underway by another and that this unit is not yet complete. When so used utterances such as uh huh are termed
40

Throughout the literature two main characterizations have been offered to deal with minimal responses. According to one, they are evidence of attention and understanding on the listeners part. Thus according to Kendon (1967:44), in producing a minimal response the addressee appears to do no more than signal that he is attending and following what is being said. A second characterization of such behaviour is that it keeps the conversation going smoothly (Dittman and Llewellyn 1967:342) or appears to provide the auditor with a means for participating actively in the conversation thus facilitating the general coordination of action by both participants (Duncan and Fiske 1977: 202-203).

168

continuers41. In producing a minimal response functioning as a continuer, speakers display their understanding that an extended turn is underway and show their intention to pass the opportunity to take a turn at talk and in so doing they facilitate the continuation of the turn that is underway. TEENAGERS Extracts 96 and 97 are reportings of a love letter and a date respectively. Since these extracts are rather long and continuers scarce, they are not given in their entirety rather the passages relevant for the use of minimal responses as continures have been selected. In both extracts Raluca is delivering an extended unit of talk while Mona inserts continuers in lines 4 of extract 96 and lines 3 and 11 of extract 97, at points where the in-progress character of the talk is clearly visible and in doing so she shows that she is listening and encouraging the speaker to continue talking.
Excerpt 96 (Constana corpus) 1 Raluca: cea mai frumoas surpriz care io consider c i-am fcut a fost scrisoarea I think the nicest surprise Ive had for him was the letter 2 nu tiu dac v-am zis de ea I dont know if Ive told you about it 3 am luat un pergament i l-am ars zi:: pe margini I took some parchment and burned it along ahhh the edge 4 Mona: mhm mhm 5 Raluca: i: i-am scris o declaraie de dragoste pe careand I wrote him a love letter which6 io n-am crezut n viaa mea ca pot s scot asemenea cuvinte [] I never ever thought I could come up with such words Excerpt 97 (Constana corpus) 1 Raluca: uite irina nu nu irina a fost cu sorin are e cu dou luni naintea mea look, irina, no, no, irina hung out with sorin. shes got, shes two months before me 2 sau cu o lun jumate or a month and a half 3 Mona: mhm mhm 4 Raluca: i:: cnd a fost cu sorin (.) chestia cum a fost (.) io la plajand when she hung out with sorin (.), the thing was (.), at the beach, I 5 io-l tiam pe sorin din liceu i-l tiam din vedere Ive known sorin since high school and Ive known him by sight 6 de cnd am fost cu tibi de fapt il tiam actually, Ive known him since I started hanging out with tibi 7 i:: bun bun ce faci dup aia la plaj and hello hello how are you after that, at the beach 8 am fost odat la plaj cu ia din cartier i era sorin i mai era un tip [.] Ive gone to the beach once with the guys from the neighbourhood and sorin was there too, and there was another guy
41

Other researchers have used the term facilitative for a continuer (Reid 1995).

169

a fost (.) mito tii it was cool, you know 10 exact o-ntilnire d-aia cnd te ia cineva i are grij de tine per total just like those dates, when somebody takes you and takes care of you totally 11 Mona: mhm mhm 12 Raluca: ne-am plimbat puin io m-am dus cu sta n fa s vorbesc we walked around a little I went ahead with this one to talk

ADULTS Excerpt 98 shows Maria to be attending to the ongoing talk at lines 2, 8, 13, and 18. This involves, on the one hand, Marias refraining from initiating a turn in order to show that she does not object to her partners having the floor and producing an extended unit of talk.
Excerpt 98 (Constana corpus) 1 Alina: i acum io tiam c-aveam nite oase n frigider de vit and now I knew I had some beef bones in the fridge 2 Maria: mhm mhm 3 Maria: i zic las din oasele alea fac ciorb i din carne fac spaghete and I thought Id make some soup with those bones and spagetti with the meat 4 i cnd dimineaa am constatat c de fapt aveam aveam numai carne but in the morning I noticed that in fact I had only meat 5 i zic a:: i ciorb din carne de vit i spaghete tot and I thought, well, beef soup and spagetti as well 6 toate snt prea cu carne de vit too much beef in everything 7 las o s fac p[ui well, Ill cook some [poultry 8 Maria: [mhm= mhm 9 Alina: =cu spaghete= with spaghetti 10 Alina: =i le-am fcut so I cooked them 11 le-am fiert pe ele Ive boiled them 12 apoi am fcut un sos cu ceap, usturoi, bulion i un cub de Knorr de pui then I made a sauce with onion, garlic, tomato paste and a Knorr poultry cube 13 Maria: mhm mhm 14 Alina: ca s aib ct de ct gust de pui to taste a little bit like poultry 15 Maria: [((laughing)) 16 Alina: [((laughing)) 17 Alina: aa i apoi am pus carne de pui tiat aa fii lun [gi so, and next I put some beef carved in long strips 18 Maria: [mhm

170

mhm

Not objecting to her partners contribution and encouraging her to carry on with her stories enables the conversation to flow smoothly, when willingness to take part in a conversation is equivalent to willingness to be part of a relationship. On the other hand, fitting to the details of the locally preceding talk is achieved through the use of minimal responses functioning as continuers. In this example, Maria satisfies her partners desire to be liked, admired and listened to. Throughout the tapes, when their conversationalist partner is talking, our adult informants, unlike the teenagers, are particularly skilled at inserting mm, aha, oh, aa (right) and other such comments throughout streams of talk rather than placing them at the end. The inserter signals that she is constantly attending to what is being said, demonstrating her participation, her interest in the interaction and in the speaker since none of these minimal responses is delayed. These minimal responses are skilfully inserted between the breaths of a speaker, causing sometimes slight overlaps and there is nothing in tone or structure to suggest that the person who produces them is attempting to take over the talk. Thus they are signals of solidarity and support. Phatic questions as a means of keeping the conversation going Questions are interactionally powerful utterances, as they open a two-part sequence. In conversation questions primarily serve two functions. They often function as information-seeking devices. A speaker can assume the role of expert, while the others ask the expert questions. On the other hand, asking question may be a way of inviting others to participate; of checking that what has been said is acceptable to those present, of ensuring that conversation continues (Coates 1993; 1995) as the following excerpts show. TEENAGERS The analysis of our corpus shows that out of a total number of 95 questions asked by teenagers, 35 are phatic questions (i.e. 36.84%), questions meant to keep the conversation flowing. In the fifteen-minute conversation from which excerpt 99 is taken, the eight-grade girls touched on 8 topics, most of them being initiated through the use of phatic questions. They began by mentioning Lauras brother and a mutual acquaintance in the neibourhood and went on to such topics as Lauras birthday present, the test in geography, cheating in written tests, a bartender training course that Lauras brother is taking, Lauras supposedly broken bicycle, etc.
Excerpt 99 (Constana corpus) 1 Anca: cnd vine maic-ta? when is your mother coming? 2 Laura: pe la:: nu tiu pe la doipe unu cam aa around, I dont know, twelve - one oclock, or something 3 ct e ceasu? whats the time? 4 Anca: doipe i doucinci twelve twenty-five 5 Laura: doipe doUcinci? twelve twenty-five?

171

6 Anca: da fat yeah girl 7 n-ai ochelarii la [tine? do you need glasses? 8 Laura: [ua::u ce mU:lt wow, that late 9 Anca: ce:: unde-i frate-tu? what? wheres your brother ? 10 Laura: la cursurile de barman at the waiter training classes 11 tiai [c face? did you know he was taking them? 12 Anca: [cte: ore? da mi-ai zis how many hours? yes, youve told me 13 Laura: trei ore parc sau dou ore= three hours, I think, or two 14 Anca: =de la ct la ct? from what time to what time? 15 Laura: de la unpe la:: from eleven to 16 Anca: de la unpe? from eleven ? 17 Laura: de la nou la unpe de la no- nu tiu nu tiu nu m prea from nine to eleven, from I dont know. I dont know, I dont really 18 Anca: (bine tu) i- i-a rupt bicicleta? (well youve) hes, hes broken your bike? 19 s-a ru:pt s-a strica:t? its broken, its broken down? 20 Laura: cine? who 21 Anca: bicicleta the bike 22 Laura: fra- fra- frate-miu? my my my brother? 23 Anca: nu tiu I dont know 24 andu (mi-)a zis c i-a stricat bicicleta andu told me that he had broken your bike 25 Laura: a::h vroia andu bicicleta tii a:::h, andu wanted the bike, you know 26 i io cre c i-a zis frate-miu c s-a stricat sau c nu o am and I think my brother told him that its broken or that I dont have it

In the extract above the phatic questions function either as a means of establishing the topic for discussion (lines 1, 9, 18 and 19) or as devices that expand on a topic that is currently under discussion (lines 11, 12 and 14). The other questions that are not phatic in intent function as elicitations of information (line 3, 7), clarifications (lines 20, 22) or they initiate repair (line 16). As this excerpt shows the topics are introduced rather suddenly and topic switches are usually occasioned by phatic questions. No topic is extensively elaborated, no topic extends over more that a few turns. Our adolescent informants show a preference for using phatic questions as means of introducing a new topic rather than as ways of expanding the topic which is currently under discussion. This comes in 172

contradistinction to adults who sustain conversation topics considerably longer and tend to use phatic questions as means of expanding on the currently discussed topic. ADULTS Throughout the tapes, 61% of the questions asked by our female adult informants initiate the conversation or keep the flow of conversation going either by expanding on a certain topic or by establishing some other topic for discussion. Excerpt 100 reproduces a short segment of all-female conversation that illustrates the consistent use of phatic questions. We have chosen this segment because it appears to be representative of the way the women in our sample use questions as a means of keeping the flow of conversation going.
Excerpt 100 (Constana corpus) 1 Iulia: ai fost la croitoreas? have you gone to the dressmakers? 2 Maria: eh, am fost pe naiba to hell Ive gone 3 Iulia: de ce? [pi n-ai zis ca te duci astzi? why? didnt you say youd go today? 4 Maria: [pi ( ) well ( ) 5 Maria: nu i-am zis c-am sunat-o m I havent told you that I called her 6 Iulia: eh, i? ce a zis? and? what did she say? 7 Maria: am fost bolNA:v Ive been ILL 8 Iulia: da ce a avut? what was wrong ? 9 Maria: eh, a avut pe dracu ( ) a avut de o lun de zile eh, she didnt have shit ( ) shes been ill for a month 10 Iulia: n-a avut clieni she had no clients 11 Maria: a::re prea multi clieni shes got too many clients 12 Iulia: are clieni? shes got clients? 13 Maria: ia clieni= prea muli i p orm nu le face fa= she takes on too many clients and then she cant manage them 14 Iulia: =i and 15 Iulia: =e:h, pi da e neserioas las c te duc e::h, well, then shes not reliable Ill take you 16 s vedem cum mi ies mie pantaloni i te duc la:: la tipa asta tii lets see how my trousers turn out, and Ill take you to this lady, you know 17 Maria: pi chiar m-am gndit la ea tii well, actually. I thought about her you know 18 Iulia: da i i le-a dat pn la urm sau nu?

173

well, and did she give them to you or not after all? 19 Maria: nu mine tre s m duc la ea s vd dac no, Ive got to go to her tomorrow to see if 20 deci s-o sun nainte so, Ill call her first 21 cre c m-au costat telefoanele I think Ive run up huge phone bills 22 am sunat-o de a:pte mii de ori Ive been calling her seven thousand times

In line 1 Iulia uses her first question (ai fost la croitoreas ) to establish a topic (i.e. the dressmaker) on which to talk. Since Maria seems to be unwilling to discuss it, Iulia makes a second attempt to launch the conversation in line 3 where she makes use of another question to invite Maria to participate. Marias answer in 5 seems to be more encouraging, offering a piece of information that can be turned into a good take-off point. From this point on, Iulia asks a series of questions in lines 6, 8 and 18. Our assumption is that Iulia is not as much interested in the information these questions elicit as she is in ensuring that conversation goes on. In other words, these questions are rather phatic in intent: what exactly is talked about seems to be less important than the fact that talk itself occurs. Apparently Iulia has achieved her aim of keeping the conversation flowing: Maria seems to be willing to contribute to the topic as her turns at speaking become longer. Both conversational exchanges demonstrate how questions can be used to keep the flow of conversation going when willingness to engage in interaction is equivalent to being part of a relationship. Interruptions, overlapping talk and anticipatory completion The occurrence of overlapping talk is one of the two major departures42 from one-speaker-talking-ata-time rule, the basic design feature of conversation (Sacks, Schegloff and Jefferson 1974). Consequently various elements of the organization of talk in interaction press for its resolution. Although the organization of interaction is indifferent to how the overlap is resolved, i.e. to who ends up with the turn, participants may display an obvious interest in the outcome by actively and persistently seeking to be the survivor, i.e. to emerge into the clear while bringing their turn to a completion43. The onset of a particular episode of simultaneous talk may subsequently be considered by either party (the prior speaker or the new starter) as an interruption and this decision may figure in how the parties conduct themselves in dealing with that instance of simultaneous talk; an essential ingredient of interruption appear to be complainability about the overlapping utterances (Schegloff 2002). Although voiced complaints are rare in ordinary conversation, in each instance we may note that parties make use of various resources that signal their orientation to intersecting talk as being problematic or not. The manner of their withdrawal from the overlap and their conduct in overlap aftermath can display the stance they take up towards the treatment of an overlap as a possible interruption. The phenomenon of overlapping talk is a paradigm case of the ambiguity of power and solidarity. Some speakers assume that only one voice should be heard at a time, so for them any instance of overlapping talk is disruptive and may be taken as an interruption, a fight for the floor and therefore a move motivated by power. However, on other occasions, if the initial and subsequent speakers end up talking at the same time, this may treated as non-competitive, non-problematic. This collaboratively-oriented simultaneous talk can serve as a way of indicating that one is interested in,
42 43

The other departure is silence, i.e. fewer than one speaker talking at a time. For a detailed analysis, see Schegloff (2000a).

174

enthusiastic about and highly involved in the interaction. We have already mentioned that this is particularly true of a certain type of conversational style which serves to carry a metamessage of interpersonal rapport and which is therefore referred to as a high involvement style44. Supportive overlapping talk Despite Romanian womens concern with maintaining a proper sense of decorum, overlapping talk seems to be pervasive in all-female Romanian conversational discourse 45. Yet this is rarely seen as disruptive or as a sign of conversational malfunction and getting the floor does not appear to be particularly problematic46. TEENAGERS In excerpts 101 to 103 taken from conversations among teenagers all instances of overlapping talk are realised as assessing comments47 or interrogatives that expand on the topic under discussion. The absence of any hitches, perturbations and recycling which is indicative of repairing proves them to be non-competitive and non-problematic.
Excerpt 101 (Constana corpus) 1 Raluca: io snt prieten cu irina deci noi sntem ca dou surori irina and I are friends; so were like sisters 2 i prinii notri nici mcar- a::: tiu una de alta da nu se cunosc and our parents dont even aaa, they know about us, but they dont know each other 3 sau dac se cunosc nu se salut
44

Cf. Tannen (1989). In addition to the frequent use of overlapping talk, this style is also characterized by a rapid pace, exaggerated intonation contour and frequent back-channel responses. This high involvement style has been shown to be characteristic of female conversational discourse in English-speaking communities. 45 This section on overlapping talk and anticipatory completion in Romanian conversational discourse is a significantly revised and extended version of Hornoiu (2004). 46 There is considerable evidence that in English-speaking communities women tend to make use of overlapping talk to indicate support, collaboration and solidarity. Coates (1989), in a study of conversations among a group of women friends, reported that simultaneous speech was very common, and that it normally consisted of work[ing] together to produce shared meanings (p.113), rather than attempts to take the floor from another speaker. Most commonly one speaker would make a comment or ask a question during another speakers turn, without in any way attempting to obtain the floor; or two or more speakers would contribute simultaneously to the same topic. Similar observations are made by other researchers. Edelsky (1993), in a study of faculty committee meetings, argued that two types of floor could be distinguished, singly developed floors and collaboratively developed floors. In single floors, in which the discussion was highly task-oriented (focusing on such matters as reporting on items), the one speaker at a time rule was followed and there were few interruptions. In collaborative floors, however, this rule no longer applied, and simultaneous speech was normal. In this type of floor, through talking simultaneously, participants developed an idea together, produced a joint answer to a question, or shared in joking. Edelsky notes that a high degree of involvement in the interaction characterized the use of overlapping talk in collaborative floor. She concludes that women participated more in collaborative floors than they did in single ones, whereas men favoured singly developed floors. In other words, this study implies that women are more comfortable talking when there is more than one voice going at once. Similarly McLachlan (1991) reported that when tackling a problem in which they were in agreement as to the solution (as determined by a pre-test), female dyads produced more non-disruptive simultaneous speech than did male dyads. 47 Assessing is an activity which may specifically be organized to occur simultaneously with the very talk activity that is being assessed. This type of overlapping talk is not perceived as disruptive; on the contrary, it shows that its producer is interested in the topic being discussed. For an insightful analysis of assessment in talk-in-interaction, including the use of concurrent assessments, see Goodwin and Goodwin (1987, 1992).

175

or, if they know each other, they dont greet each other 4 Mona: s tii c am i eu [o prieten foarte bun you know Ive got a very good friend too 5 Raluca [e o chestie ciudat its weird 6 Mona: care a venit smbt atuncea who came that Saturday 7 vreau s spun ca noi eram foarte bune prietene I mean we were very good friends 8 deci io cnd m-am mutat n constana [.] so, when I moved to constanta Excerpt 102 (Constana corpus) 1 Laura: i nu ia zis c merge la biliard la bowling la:: [c-au venit pe la dOU and he didnt tell him that he was going to play billiards, bowling, cause they came back around TWO 2 Anca: [da unde-i biliardu sta mare? but where is this billiards room ? 3 Laura: au fost n mamaia [primthey went to mamaia first 4 Anca: [a::: a ::: 5 Laura: i p-orm au fost la occident aiopt sau ai [nou and then they went to occident sixty-eight or sixty-nine 6 Anca: [ainou sixty-nine

Excerpt 103 (Constana corpus) 1 Laura: am ntrebat-o i pe elena ce i-a adus mou i-a zis ca nimica Ive also asked elena what Santa had brought her and she said nothing 2 i-a luat [tii i:: she had already bought herself, you know, and 3 Anca: [ce napa aa this sucks 4 mie nu-mi place s-mi iau aa i s tiu c nu-mi mai aduce nimica mo crciun I dont like to buy myself things and know that santa wont bring me anything else

ADULTS In excerpt 104 interruptions and overlaps, far from being disruptive, are used by both conversationalists to pursue the topic simultaneously.
Excerpt 104 (Constana corpus)

176

1 Iulia:

da bi uite avea i d-tia ca ai ti mari. yeah, look, they also had like these ones, like yours 2 <da acuma dac stau bine i m gndes <now if I come to think about it 3 s-ar putea s nu fi fost chiar aa materialu. the fabric might have been a bit different. 4 <tot cinci sute de mii erau i ia they were five hundred thousand as well 5 Maria: io p-tia am dat trei sute da m-neap nfiorto:r= I paid three hundred for these, but they itch terribly 6 Iulia: =i :ia erau C[RE:::m ] and those were CREA::M 7 Maria: [prin cio]RAP through stockings 8 Iulia: sta- DA:: s-i spun [erau un- nu crem] erau un be::j = wait YES, let me tell you, they werent cream, they were beige 9 Maria: [foa:rte mito snt] theyre so cool 10 Iulia: = i aveau ici colo cte un punct rou and they had red dots here and there 11 i c:nd i-am probat erau era un material splendid cnd m-am mbrcat and when I tried them on, they were- the fabric was great when I put them on 12 va::i nepeau- nepau groa:znic= Ay, they itched, they itched terribly 13 Maria: =e:::h ca tia Eh, like these ones 14 Iulia: i am zis nu pot s-mi iau io: aa hhh [cehhva <(s hh m::) and, I said, I cant buy something like this (to hmmm) 15 Maria: [deci m mnnc hhh de nu mai pot so its itching me, hhh. I cant stand it 16 Iulia : da da thats right 17 Maria: [de nu mai pot I cant stand it 18 Iulia: [pcat pcat such a shame

Excerpt 104 shows both speakers as initiating turns that either latch onto or intrude into the others turn. In lines 1-4 Iulia starts describing a pair of trousers that are similar to those of her friend, mentioning the quality of the fabric and the price. In line 4 Iulia seems to have brought her turn to an apparent grammatical prosodic and pragmatical completion, at which point Maria intervenes by producing her own turn, saying how much she paid for her trousers and pointing to a drawback of the fabric, which again seems to be finished in line 5. Iulia resumes her description in line 6 mentioning a detail: the colour. This strategy that she employs in line 6 renders her prior turn retrospectively incomplete, which in turn renders Marias contribution in line 5 interruptive. As Iulia comes to the possible completion of the turn constructional unit that makes up her turn in line 6, she apparently registers Marias gearing up to start a turn of her own (the turn is actually started and delivered in slight overlap in line 7) and therefore Iulia wants to signal that she is in the course of the production of an extended spate of talk; she does so by employing a sudden sharp increase in the volume and pitch of her talk, represented in the transcript by the capitalization and partial underlining of the word CRE:::m 177

in line 6. Notice that her sudden increase in volume and pitch occurs during the pre-onset phase (Schegloff 2000:15) of the overlap. By deploying such practices otherwise available for dealing with overlapping talk, the speaker may attempt to delay or interdict a possible start-up by another speaker. Marias talk, however, is launched in slight overlap in line 7. This first instance of overlapping talk appears to be problematic because: (1) of the practices Iulia employs in the pre-onset stage whereby she tries to interdict a possible start-up and to signal that she is in the middle of the production an extended piece of talk, an environment that renders possible overlapping talk as an extremely disruptive face-threatening move; thus on the one hand, Iulia tries to warn Maria against performing a face-threatening action, on the other, Iulias behaviour is equally face-threatening since trying to delay a possible start-up by another speaker may be equivalent to attempting to dominate the floor; (2) Marias contribution in line 7 is an add-up to her prior otherwise complete turn in line 5, turning thus Iulias turn in line 6 into a possible interruption. What stance do the parties take toward the occurrence of this first overlap? Neither speaker goes into competitive production by employing hitches and perturbations such as a sharp increase in volume and pitch, cut-offs, stretched sounds, etc., whose especially dense occurrence is characteristic of competitive overlapping talk. Iulias stretching of the vowel /e/ in CRE::m and her high pitch and loud volume while delivering this vowel as part of the overlap is a continuation of the strategy she employs in the pre-onset phase rather than a signal of her switching to competitive production. Notice that by the end of the overlap she moves to solo production mode. Marias increment in line 7 is also delivered in solo production. The hitches and perturbations in lines 7 and 8 (Marias high pitch and loud volume in the second syllable of the word cioRAP, Iulias cut-off at the beginning of line 8, the word DA::: delivered in high pitch loud volume and with vowel stretch) do not signal a fight for the floor but are normal occurrences in the aftermath phase of the overlap and their role is to signal that the overlap was taken notice of (Schegloff 2000a: 33). The second sequence of overlapping talk occurs in lines 8 and 9. In line 8 after the first overlap is resolved and taken notice of, Iulia carries on with her description giving details in lines 10-12 about the colour and the pattern of the fabric and pointing that when she tried on those trousers their fabric had the same drawback. Marias turn in line 9 which is an appreciation of Iulias possessions catering thus for the latters positive face begins in overlap with Iulias talk already in progress. However, its being delivered in solo production with no hitches and perturbations 48 renders this overlapping episode as non-competitive. Similarly, Iulias slightly overloud words overlapped by Marias turn erau un- nu crem and the cut-off seem to be less a shift to competitive production than a contrastive stress which implements the opposition to her prior description. A third episode of simultaneous talk occurs in lines 14 and 15. In line 14 Iulia concludes her story by saying that she eventually decided not to buy them while Maria in line 15 once again complains about her trousers. In line 15 Maria misprojects the end of Iulia s turn: she begins to speak at what she takes to be the end of Iulias turn overlapping the last word of Iulias TCU. Iulia however rushes through49 the juncture following her possible completion into a next TCU only to find herself talking in overlap with Maria. Again the overlap is quickly brought to its resolution phase in the third
48

According to Schegloff (2000a, 2002), any deflection of solo production suggests a shift to competitive production and thereby an orientation to possibly problematic overlapping talk. 49 See Schegloff (1982:76) where he defines a rush through as a practice whereby a speaker, approaching a possible completion of a turn-constructional unit, speeds up the pace of the talk, withholds a dropping pitch or the intake of breath and starts producing another turn-constructional unit in such a way as to bridge what would otherwise be the juncture at the end of the unit.

178

beat by Iulias dropping out and with neither of them upgrading to competitive production which would have signalled problematic overlapping talk and a possible fight for the floor. There is another point to be advanced here: Iulias rush through seems to point to a high involvement style rather than to her wish to grab the floor. This interpretation would be consistent with the rapid pace of this conversational exchange (notice that Iulia employs a rush through at places where nobody is contesting her right to occupy the floor) and the numerous instances of latching employed by both parties. In fact had Iulias rush through been meant to grab the floor for the next TCU, she would not have dropped out the moment she found herself talking in overlap with Maria. Finally, as Maria reiterates her complaint in line 17, it is Iulia who starts talking in overlap while sympathizing with Maria. Note again that both Marias and Iulias turns in lines 17 and 18 are delivered in solo production with no signs of competitive talk. In this conversational exchange, despite Marias and Iulias consistent use of overlapping talk, fighting for the floor does not seem to be an issue. These violations of the turn-taking rules of conversation are not meant to infringe the other speakers turn and reduce her to silence. Iulia and Maria develop their topic jointly and silence, which is a sign of malfunction in conversation, is absent. The two participants alternate their turns rapidly with no gaps between turns and these instances of slight overlap can be viewed as displays of involvement with the topic and with each other. If in the previous example both speakers made use overlapping talk in order to jointly construct a topic, excerpt 105 provides another example of collaboratively-oriented overlapping talk by means of which a speaker is invited to join in.
Excerpt 105 (Constana corpus) 1 Iulia: s vezi ce a fcut atunci, s vezi o sun tipa asta wait and see what she did then. well, this lady calls her 2 <e tot medic shes a physician as well 3 i o sun pe alina and she calls alina 4 ct era? unpe noaptea ct era? what time was it? eleven at night what was it? 5 Alina: unpe jumate doipe (nici nu tiu) half past eleven twelve (I dont know) 6 Iulia: da (.) yes 7 Iulia: [nnebunit ia] povestete-i tu= mad, come on, you tell her, 8 Alina: [speria:t c-] scared that 9 Alina: = n loc s-i pun picturi de nas cu Olinth instead of giving him Olinth nose drops 10 Alina: pentru bielu ei i-a pus ALte tipuri de picturi tot pentru nas for her little boy, she gave him ANOTHER kind of nose drops 11 da hhh cu hhh un hhh antihistaminic care era pentru aduli but hhh with hhh another hhh type of antihistamine which was for adults 12 Maria: [AOleu AY 13 Alina: [VA:i dispera:t sra:ca dispera:t AY, poor thing, she was desperate

179

14

la unpe jumate ce se poate ntmpla? at eleven thirty what could happen? 15 Ali:na de trei zile i pun din astea [((laughs)) Alina: Ive been giving him these for three days 16 Maria: [((laughs)) 17 Maria: a::: i nu i-a dat seama? a::, and she didnt realize? 18 Alina: nu i-a dat seama she didnt realize 19 i-n seara a:ia ce-o::: fi fcut-o s se uite [pe asta ] and that night what made her look at this 20 Maria: [>da bine c-<] well its good that 21 Alina: pe etichet pentru c ea pn atunci i punea i tia c avea flaconu la fel da eticheta pe etichet scria altceva= at the label. because she had been giving him before and she knew that the bottle was the same but the label - the label read something else 22 Maria: =va::I ay 23 Alina: i era disperat, disperat= and she was desperate, really desperate 24 Maria: =oribil horrible 25 Alina: bi:ne c [n-a: fost nimic] thank God nothing happened 26 Iulia: [n-a avut nimica] he was fine 27 Alina: se absoarbe de obicei it usually gets absorbed

In lines 1 to 3 Iulia starts telling her friend Maria a story about an acquaintance of her sister-in-laws who one night called them at a very late hour. In line 4 she asks Alina, her sister-in-law, what the time was when latters acquaintance rang them up. Iulias question is not meant to elicit information. Instead she uses the interrogative form to invite her sister-in-law to take part in the interchange, which she does in line 5. In lines 7 and 8 Iulias and Alinas contributions overlap: they both express their sympathy towards Alinas acquaintance. This first overlap is caused by a relatively straightforward turn-taking miscue: both Alina and Iulia self-select to take next turn, and thus start simultaneously. In this brief overlap neither of them shifts to competitive production, both deliver their turns in solo production showing thus that they do not find this instance of simultaneous talk disruptive; we can notice, however, two minor departures from solo production: Iulias nnebunit is given emphatic stress (this is indicated in the transcript by underlining placed under the second syllable of nnebunit) and Alinas stretching of the vowel /a/ in speria:t (the stretching of the sound is indicated by colons placed after it). These minor deflections of solo production mode rather than displaying an orientation to overlapping talk as possibly problematic are used with a view to drawing attention to the seriousness of the situation. The overlap comes to an end relatively quickly as Alina drops out: note that her sudden cut-off occurs after the conjunction c (because) which introduces a subordinate clause which would have stated the reason why her acquaintance was scared. Noticing Alinas intention to tell the story, Iulia invites her to do so as she emerges into the clear and carries on with her next TCU. Iulias intervention in line 7 is not disruptive for Alina as in line 9 the latter carries on with her turn which began in line 8: line 8 ends in a conjunction and her turn in lines 9-11 is made consistent with this 180

conjunction. The relatively quick resolution of the overlap, the absence of hitches and perturbations that might have pointed to competitive production and both speakers willingness to give up their turn so that the other might carry on with hers reflect a lack of interactional investment by both parties in securing the turn space. Thus fighting for the floor is not an issue here. Note that Alinas contribution to the conversation is an extended unit of talk with clear marking of its beginning (Iulias story preface50 in line7) and ending (Alinas and Iulias turns delivered in overlap stand as the conclusion to the story). Starting with line 9 as far as line 26, Maria orients herself to and exhibits an understanding that an extended unit of talk is underway by some other party and that it is not yet complete. She takes the stance that Alina, the producer of that extended unit of talk, should not only continue talking but she should continue that extended unit. Continuers such as mhm, uh huh, unh hunh yeah, oh, yes would exhibit this understanding and take this stance by passing an opportunity to produce a full turn at talk (Schegloff 1982:81). Thus in lines 12, 22 and 24 Maria confines herself to continuers which she delivers either by overlapping with or latching to Alinas story. These instances allow us to remark on several additional points which may shed light on the interactional texture involved here. First, note that the bits of behaviour produced by the recipient of an extended unit of talk may vary and usually do vary. Even when no more than continuer usage is involved the tokens employed for it vary since the occurrence in four or five consecutive slots of the same token may be taken to hint incipient disinterest, whereas varying the tokens, whatever tokens are employed, may mark a baseline of interest (Schegloff 1982:85). Second, in some of the positions in which some sort of continuer is relevant, the immediately preceding talk may be such as to invite some sort of reaction instead of or in addition to the continuer. Thus throughout extended units, stories being a case in point, markers of surprise or assessment such as oh my, isnt that weird/wonderful, and the like are frequently employed. Note that Alina packages her story in a very dramatic format with exaggerated intonation and directly quoted speech which is designed to be more appreciative than occurrences of a single continuer are. Thus at places where some continuer might be relevant, Maria uses, in lines 12, 16, 22 and 24, tokens of surprise AOleu, va::i, assessments oribil and a laugh token to signal that she finds the story interesting and assessable. Maria intervenes two more times in this conversational exchange and although on cursory examination both her interventions may appear problematic, they are actually designed to be supportive elaborations of and comments on the point Alina is making. In line 17 Maria asks a question at a point where the story is not yet over. This may appear as a face-threatening move on Marias part since interrupting the other gains its sharpest profile when a story or some other extended unit is in progress. However, since her question is designed to be in continuity with the topic currently discussed it does not embody the conventional sense of aggression or disruptiveness associated with the term interruption. This interpretation is supported by Alinas repetition in line 18 of Marias utterance ratifying thus listenership (cf. Tannen 1989) by agreeing with Marias understanding of the story. The excerpt ends with Alinas conclusion that eventually nothing serious happened, conclusion stated in line 25. While Alina is producing her turn in line 25, Iulia joins her in stating the conclusion to the story. Thus Alinas turn is partially overlapped by Iulia's contribution. Again neither in the overlapping sequence nor in its aftermath are there signs of competitive production. Simultaneous speech in all-female discourse can take the form of asking questions and making comments while another participant is speaking. Extract 106 occurred in the context of talk about complementary medicine and herbal remedies.
Excerpt 106
50

See Sacks (1992) for a discussion of the turn that includes a story preface.

181

(Constana corpus) 1 Alina: asear m-a sunat melania s-mi spun last night melania she called to tell me 2 c a luat pentru copilu ei tii that she has bought for her child, you know 3 c a luat un medicament m rog care avea that she has bought a medicine, anyway, which had 4 Iulia: aa yes 5 Alina: un lactobacilus i a dat a lactobacilus and she paid for it 6 tot chipurile naturist i aa supposedly a herbal medicine as well and so on 7 o sut de mii pentru la mic a hundred thousand for the kid 8 c-i sub form de mini [marieni cause its shaped like mini martians 9 Iulia: [da pentru ce-i? but what is it for ? 10 Alina: pentru reglarea florei intestinale to regulate the small bowel flora 11 Iulia: aha c el are probleme aha, cause hes got problems 12 Alina: da o sut de mii? [i zic da melania but a hundred thousand? and I say, but melania 13 Iulia: [melania are bani melania is well-off 14 Alina: da la nu prea se d sub trei ani tii but that isnt usually for kids under three, you know 15 da ea zice da m da altceva nu se gsete but she says that, well, you cant find anything else 16 c io spusesem s ia (ecoflorin) sau altceva cause I told her to give him (ecoflorine) or something like that 17 i n-a vrut nu n-a gsit-o and she didnt want no she didnt find it 18 i atunci i d marieni aia doi and then she gives him those two martians

In lines 1-3 and 5-8 Alina starts telling a story about Melania, an acquaintance of hers, who rang her up the previous night to tell her about some medicines for intestinal disorders that the latter had bought for her son. In line 9 Iulia begins asking her question (da pentru ce-i? but what is it for?) at the very end of Alinas turn, overlapping part of the last word. This overlap under the form of a question that elicits more information about the topic currently discussed follows Iulias minimal response in line 4 and reinforces the latters main function as a strategy for supporting the speaker in her choice of topic. In line 7 Alina mentions the price of the medicine ( o suta de mii one hundred thousand) which is then repeated in line 12. This self-repetition is meant to support the point she has previously made and to persuade her conversationalist partners that contrary to popular opinion some herbal medicines are quite expensive. The second instance of simultaneous talk appears in line 13 under the form of a supportive comment. This overlapping comment (Melania are bani Melania is well-off) is not an attempt to take 182

over since it does not prevent Alina from finishing her turn in lines 14-18. On the contrary it provides indirect evidence that backs up Alinas opinion regarding the price of some herbal medicines. By supplying the reason why Melania can afford to pay one hundred thousand for two tablets, Iulia may be taken to imply that she agrees with Alinas opinion that some herbal remedies are far from being affordable. In the examples examined so far as well as in other materials drawn from talk-in-interaction with which I am familiar it turns out with great regularity that while engaging in overlapping talk with female friends, Romanian women do not generally upgrade to competitive production. The vast majority of overlaps are resolved to a single speaker in the first or second beat while longer overlaps which do not extend for more than 4 or 5 beats are delivered in solo production mode displaying thus no hitches and perturbations characteristic of problematic competitive talk. Although pervasive in all-female Romanian conversational discourse, overlapping talk is not used with a view to gaining the floor, at least not in friendly talk. In the examples examined so far participants in overlapping talk do not appear to be or to conduct themselves as competitors fighting for the floor but rather as properly simultaneous occupants of the floor either because this is permissible (e.g. with overlapping continuers) or because it is mandated by certain rules of appropriate social conduct (e.g. expressing ones concern about or sympathy on somebodys misfortune, paying compliments, congratulating somebody in response to announcements of personal good news) (Schegloff 2000a; 2002). Anticipatory completion We have already seen that not all instances of overlapping talk are competitive and disruptive. One the one hand overlapping talk does not necessarily involve interruption or fighting for the floor and the excerpts above illustrate various forms of collaboratively oriented overlapping talk. On the other hand, interruption without overlap can occur when the recipient of a not-yet-complete turn designs the newly entering stream of talk to be in continuity and complementarity with the turn already in progress. This state of affairs has been referred to by Lerner as anticipatory completion and and is similar to a certain extent to but not identical with what Sacks refers to as collaborative sentence constructions51. This conversational strategy does not embody the conventional sense of hostility and disruptiveness associated with the term interruption. TEENAGERS In both extracts 107 and 108 the recipient of a not-yet-complete turn produces her own turn in lines 2 which is syntactically and semantically consistent with what went before anticipating thus the outcome.
Excerpt 107 (Constana corpus) 1 Irina: deci te-am pupat am trecut la chestii mai aa mai [((laughing)) so kiss you, we went on to things more more 2 Raluca: [mai intime more intimate 3 Irina: mai intime tii
51

For a detailed discussion, see, among others, Sacks (1992), Lerner (1991) and Lerner (1996).

183

more intimate, you know Excerpt 108 (Constana corpus) 1 Anca: andu-l ntreba pe el s-i zic lu maic-sa pe ce pe ce andu was asking him to tell his mother what what 2 Laura: pe ce osea a fost what street he took 3 Anca: nu pe::- da pe ce da pe ce osea am fost not on but what what street have I taken

Similarly, both recipients of these anticipatory completions (i.e. the producer of the not-yet-complete turn) agree to them by using ratifying repetitions at lines 3. This conversational strategy is similar to collaboratively built sentences in showing that participants in the conversation know what is on each others minds. ADULTS In excerpt 109 Maria and Iulia have been comparing McIntosh computers to IBM compatibles.
Excerpt 109 (Constana corpus) 1 Maria: pi da asta-i bine asta-i bine c de multe ori mi s-a-ntmplat s= well, thats good its good cause it happened to me many times to 2 Iulia: =s-nchizi din greeal to accidentally close 3 Maria: s le-nchid din greeal tii to close them accidentally you know 4 Iulia: da i porm= yes, and then 5 Maria: =nnebuneti s nu mai gseti youre going crazy if you dont find it anymore 6 Iulia: da s nu mai gseti yes if you dont find it anymore

In line 1 Maria starts explaining why she finds a certain characteristic of the programs running on McIntosh computers very useful. Line 1 ends in a subordinate conjunction. Iulia interrupts Maria and produces in line 2 a clause that is syntactically and semantically consistent with Marias turn in the previous line. Marias repetition with slight variation (changing from 2nd to 1st person) in line 3 ratifies Iulias contribution. In line 4 Iulia agrees with Maria and starts mentioning a related aspect ( da i porm yes and then). However, she is interrupted by Maria who makes use of the same strategy: her contribution in line 5 is syntactically consistent with the adverb phrase used by Iulia in line 4. Iulias exact repetition in line 6 ratifies Marias contribution. Sometimes both the producer and the recipient of a not-yet-complete turn end up talking at once and the two articulated completions may not be identically composed, even though they both are designed to be syntactically and semantically consistent with the previous turn. The following excerpt, which comes from adults conversational style, is a case in point. 184

Excerpt 110 is taken from a conversation where Anca talks about a murder that shocked the whole community and was in the headlines for several months.
Extract 110 (Constana corpus) 1 Anca: s-a dus la mtu-sa c i maic-sa i mtu-sa stau tot acolo (n preajm) he went to his aunts cause his mother and his aunt live nearby as well (close) 2 cic aia i-a luat cuitu l-a splat l-a schimbat p-sta that one supposedly took the knife from him, cleaned it, changed his clothes 3 i-a splat toate hainele de snge she washed the blood off all his clothes 4 nu tiu ce ca s-l ac[opere whatever, to cover him 5 Rodica: [ascund= cover 6 Rodica: =da bine-neles but of course

At line 4 Rodica intends to align with Anca by collaborating in the production of the latters turn and provides a completion of Ancas turn delivered in lines 1-3. Anca, however, completes it herself, at line 3, with the same upshot but different composition. Thus the two completions are delivered in overlap. However, this is treated by both conversationalists as non-problematic and non-competitive 52. This interpretation is supported by the absence of hitches and perturbations as well as by the agreement in line 5. 5.3. Conclusions Despite the universality of such issues as showing invlovement with or keeping the interlocutor at a distance, i.e. honouring their positive or negative face respectively (Goffman 1956; Brown and Levinson 1987), research has shown that societies can be grouped according to their ethos, i.e. according to the affective quality of interaction characteristic of members of a society (Brown and Levinson 1978: 248). Not only can societies be positive-politeness or negative-politeness oriented, but this differentiation also applies to various social groups within complex stratified societies. An orientation towards positive-politeness as opposed to negative-politeness linguistic behaviours marks one social group from another and also different kinds of social roles from one another. However, this does not mean that societies as a whole or that various social groups within a complex stratified society can clearly and exclusively be categorized as being either positively or negatively polite. Rather they can be categorized as relatively more positive politeness oriented or relatively more negative politeness oriented, according to the type of ethos which is given more play. Moreover, since this politeness orientation is relative rather than absolute, it should be added here that although two societies or two social groups within a given society may have the same politeness orientation, nevertheless they will not exhibit identical preferences (Sifianou: 1999:40), either in terms of the strategies their members may choose to use, or in terms of the frequency with which such strategies are employed. Thus, we hypothesized that Romanian women when engaging in verbal interaction with other women put more emphasis on serving the need for positive face i.e. honouring their interlocutors need
52

See also Tannen (1983).

185

for involvement. More specifically they have a tendency to use more elaborated positive-politeness strategies and with greater frequency than do men among men. The findings reported in this chapter emerged from empirical research meant to test this hypothesis. They are based primarily on the analysis of original data taken from ten hours of face-toface interactions recorded in both formal and informal settings. These interactions have been referred to as the Constanta corpus. The participants (primarily grouped in dyads) include twenty Romanian women living in Constanta, whose ages ranged from thirteen to sixty-four (including eight teenagers, six in their twenties, two in her thirties, three in their forties, one in her sixties) and and sharing a similar educational background with their interlocutor. The findings that emerged from the Constanta corpus were also illustrated by extracts from another corpus of spoken Romanian gathered by researchers from the Department of Romanian Language, Faculty of Letters, University of Bucharest and which we use as a control sample to prove that the linguistic preferences identified do not happen by chance but are caused by social and cultural factors and that it is highly probable that we can generalize these preferences for positive-politeness strategies to a wider population. The analysis of the extracts presented in this study is undertaken within the framework of conversation analysis, an approach to discourse grounded in ethomethodology, which we supplemented with elements from Brown and Levisons pragmatic theory of politeness. The analysis of our data has confirmed the hypothesis showing that the interactional strategies for which my informants manifest an interest can be grouped under the following headings: S claims common ground with the addressee and S fulfils the addressees want (for some X) , headings that label two broad mechanisms of positive politeness (Brown and Levinson 1987). Moreover, whenever linguistic forms (pragmatic particles and interactional strategies) polysemous so that they can be interpreted either as a display of solidarity or a display of dominance, our informants consistently show a preference for the function associated with positive politeness, in other words such linguistic forms are used as markers solidarity. To conclude Table 1 summarises our findings in terms of the conversational strategies which our informants consistently use when engaging in friendly talk. The linguistic variables given in bold complete the list of linguistic embodiments of positive-politeness strategies proposed by Brown and Levinson (1987:103).
Positive-politeness mechanism S claims common ground with H ground with the addressee ground with H S claims in-group membership S claims common opinions, Ways of activating the mechanism S conveys that he finds some want of H admirable/interesting interesting Positive-politeness strategy Notice/attend to H S exaggerates interest/approval/ sympathy S intensifies interest to H S uses in-group identity markers S seeks agreement with H Linguistic embodiments of positive-politeness strategies noticings about changes or possessions of H; compliments Extreme case formulations Historic present; constructed dialogue (i.e. directly quoted speech) slang/jargon; address forms; collaboratively built sentences Repetition as a way of ratifying listenership; Repetition as a way of

Claim

186

S ground with the addressee

attitudes, knowledge with H S avoids disagreement

S constructs shared knowledge with H S fulfils Hs want for some X S gives gifts to H: S gives symbolic gifts to H: sympathy, understanding, co-operation

showing participatory listenership; minimal responses as an indication of agreement token agreement; hedging opinion (the use of softeners); achieving alignment with H through the use of repetition and constructed dialogue/directly quoted speech Imagery and details; general extenders minimal responses as continuers; supportive overlapping talk; anticipatory completion; phatic questions

Table 1 The inventory of interactional strategies and their linguistic embodiments in female conversational style in Romanian Many of the interactional strategies for which Romanian women show a preference are associated with positive politeness, being thus ways of expressing similarity between the self and others by indicating the speakers appreciation of the addressees wants i.e., the addressees actions, opinions, acquisitions, etc. are thought of as desirable. However, given the fact that this solidarityoriented conversational style is acquired within the process of socialization and that many positive politeness manifestations are multifunctional in nature, we expect various segments within this gender group to differ in relation to their preference for certain linguistic devices. Thus in the next chapter we examine our ten-hour corpus of naturally occurring conversations in order to make observations of a quantitative nature, observations meant to reveal expected intra-group differences and the emergence of this solidarity-oriented conversational style.

187