1 Grammatical Constructions in the Context of Conversational Interaction Linköping Workshop April 23-24, Organisers: Per Linell, Mathias Broth & Jan Anward, Department of Culture and Communication, Linköping University Abstracts Arnulf Deppermann, Mannheim: Retrospective linking as projection of response obligation: The German modal particle "denn" In German conversation, the use of the modal particle "denn" (which has no lexical equivalent in English) is pervasive in questions. Examples: "Wer ist denn Peter?" (Who is Peter?) "Wie war's denn?" (What happened?) Drawing on a collection of more than 300 instances from German interactions (private, institutional, mediated), I will present some preliminary findings concerning sequential organisation, turn-design and discourse function of "denn" as a modal particle. "denn" is a cohesive device which locates the question with respect to the prior turn of the addressee or to some feature of the situation which is held to be known in common and to be salient in its relevance for the conversation. "denn" thus signals topical coherence and reference to common ground which serves as a legitimate reason for the question. In addition to this retrospective function, I will try to show that "denn" is used as a display of the speaker's assumption that the hearer knows that s/he is normatively required to give some account which expands on his/her prior turn and which is still missing. "denn" thus seems to work as a marker of normative implications of posited common ground and as such is a rhetorical device for self-legitimizing a question and for strengthening its conditional relevance. The use of "denn" can be seen as a starting point to think about the relevance of interactional retrospection and the assumption of common understandings for interactional projection and the (rhetorical) navigation of joint projects.
2 Susanne Günthner, Münster: Projective constructions in German interactions Based on a corpus of conversational German, I will analyze complex constructions in spoken German interactions. I shall argue that forms and functions of pseudoclefts, N-be-thatconstructions (e.g. die Sache/das Ding ist 'the thing is/the point is ') and extrapositions with es ('it') are closely connected to their on-line processing as well as dialogical negotiation in the course of interaction. Furthermore, the empirical study shows that the standard assessment of these constructions as "bi-clausal sentence patterns" cannot be supported by data from spoken interactions. Instead, participants use these syntactic patterns as 'projecting constructions', with the first part projecting an upcoming focal stretch of discourse. Due to the force of the first component in projecting "more to come" on various levels, these constructions are predestined to expand complex arguments over longer sequences of talk; at the same time, they work as a floor-holding device. I shall argue that due to the formal and functional similarities of these syntactic patterns, one could speak of a family of constructions i.e. the family of projecting constructions. Wolfgang Imo, Münster: Markers of emphasis in German and English The "change-of-state token" oh is described by Heritage (1984) as part of a structure interactants use to signal the receipt of new (and potentially interesting) information. When I was looking at German equivalents of the English oh (Imo 2007), I not only came across structures that more or less exactly mirrored those described by Heritage (1984) but also found structures that resembled change-of-state sequences while showing some striking differences. These patterns do not involve "neutral" expressions such as oh, aha, ah, ach or ach so but among others words such as wirklich?, ehrlich? or echt?, (all three could be translated by really?). They are used to signal not so much the receipt of new information but the involvement of listeners in what a speaker just said and the expression of astonishment or emphatic alignment with the speaker. As these patterns make use of and exploit the pattern of repair sequences without being "real" repairs, I will call them "show repairs" (stressing the conceptual similarity to what Antaki/Wetherell (1999) named "show concessions"). In my presentation I will illustrate amd compare the sequential structure, functions and lexical material that forms the basis of the "communicative project" (Linell 2005) of "show repairs" in German and English everyday conversation. Antaki, Charles und Margaret Wetherell (1999) "Show Concessions". In: Discourse Studies 1 (1), 2-27 Heritage, John (1984) "A change-of-state token and aspects of its sequential placement". In: Atkinson, J. M. und John Heritage (eds.) Structures of Social Action: Studies in Conversation Analysis. Cambridge, Linell, Per (1998) Approaching Dialogue. Benjamins.
3 Susanna Karlsson, Göteborg/Potsdam: On question intonation and the intonation of questions The general notion of questions and the intonation of questions in Swedish is complicated: Swedish has question syntax, thus not necessitating a question prosody, but if there is question prosody, it is a final rise (SAG Huvudsatser 3). On the other hand, some declarative utterances are heard and treated as questions. Here, rising intonation, question intonation, is often said to play a role in the analysis of the utterance as a question. It has, however, been shown that a lot of declaratives that are heard as questions do in fact not display a final rise (Engdahl to appear). Furthermore, in conversation, the phenomenon of final rises on declaratives that are not treated as questions is frequent. Some researchers have proposed that questions with a final rise are perceived as friendlier and more oriented towards social questions, while earlier peaks are generally perceived as information seeking (House 2005). In this presentation, some prosodic and interactional aspects of questions and final rises will be discussed. Literature Engdahl, Elisabet (to appear) Frågor i NoTa. In Johannesen, Janne Bondi & Kristin Hagen (eds) Språk i Oslo. Ny forskning omkring talespråk. Oslo: Novus forlag. House, David Phrase-final rises as a prosodic feature in wh-questions in Swedish human machine dialogue. Speech Communication 46: SAG = Teleman, Ulf, Staffan Hellberg, and Erik Andersson Svenska Akademiens grammatik. Stockholm: Norstedts Ordbok. Jan Lindström, Helsinki: Verb-first conditionals in spoken Swedish and German The purpose of this presentation is to compare spoken Swedish and German uses of the verbfirst conditional construction, i.e. sentence formations where the conditional protasis (antecedent) has the syntactic form of a polar question. An example is given from Swedish, a context where an environment advisor (A) explains to a farmer (F) what she is inspecting in his farmlands: A: Vi tittar ju på hur närings- hur mycke näring de ä: i marken, Wir sehen nach wie Nähr- wie viel Nährstoff es im Boden gibt F: Aja.
4 A: Äre väldigt näringsrikt så äre ju så att nåra få: växter klarar sej vä:ldigt bra: å konkurrerar ut dom andra. Ist es sehr nahrhaft ist es so dass nur wenige Pflanzen sehr gut überleben und die anderen besiegen. F: A The construction is here understood as a development of a dialogical sequence of the type question answer which has emerged and grammaticalized into an integrated sentential unit. The description of this development will be discussed from the point of emergent and dialogic grammar and (more tentatively) construction grammar. The discourse functions of the construction will also be discussed. It seems that spoken Swedish is more productive in the use of V1 conditionals than German, for example allowing a greater variety of verbal forms. In spoken German, the construction is fairly rare and restricts to irrealis contexts, often with the verb form hätte. In this sense, the situation in present day German is quite reminiscent of the use of conditionals in English, and this suggests that German has gone longer in the grammaticalization and divergence development of conditional constructions than Swedish. Per Linell, Linköping: Reactive constructions in Swedish Some grammatical constructions are formally designed to be used in utterances that are responsive to particular types of prior utterances, and/or projective of particular types of possible next utterances. What I call reactive constructions are a subclass of (formally designed) responsive constructions. By the use of a reactive construction the speaker reacts to (or against) a particular word or expression (X) as used in a prior utterance by self or other. The speaker reacts to this usage by a repetition of the word/expression X, usually reaccentuated, in the pre-front or front field of a new utterance or turn (TCU). Most reactive constructions put this repeated X in a grammatical frame of some kind, and this is often followed by an account of why the speaker regards the situated use of X, in the previous utterances, as more or less inappropriate. There are a number of special reactive constructions in Swedish, and presumably in other languages (e.g. German) too. One which has been fairly thoroughly researched is the so-called x-och-x ( x and x ) construction (Lindström & Linell, 2007). Here is an example (x in bold): (1) FLYTTA Å FLYTTA (SAM:V1: 989ff) 1. G: sen så beslagtos huse å (0.5) dom flytta tilbaka then so confiscate+pret+pass house+def and (0.5) they move+pret back then the house was confiscated and (0.5) they moved back
5 2. ti (0.7) ti Hamburg (å) to (0.7) to Hamburg (and) to (0.7) to Hamburg (and) 3. M: nå flytta å flytta men ja menar va (.) PRT move+pret and move+pret but I mean+pres what (.) well (nå), moved and moved but I mean what 4. fan kan du göra the devil can you do(inf) (.) the hell can you do In this paper I shall explore another reactive construction, which I tentatively call X aux X aux (Linell & Norén, fc.). It consists of two copies of a finite auxiliary or pseudo-auxiliary (= X; without any interjacent and ), and occurs in the front field of an utterance following another utterance where this X has occurred (as an auxiliary) in the same tense form. Here is an example: (2) VÅGAR VÅGAR (A and B have had a leakage from a pipe which resulted in a flooding of the flat they have rented. They have installed radiators to dry the floors out. In the morning of the following day, they are going on an excursion with C, who has come to pick them up:) 1. A: vågar vi lämna elementen på när vi åker bort? dare+pres we leave radiators+def on when we go+pres away do we dare to leave the radiators on when we go away 2. C: vå gar vågar ni väl men de behövs inte dare+pres dare+pres you PRT but it need+pres+pass not dare you do I s pose but it is not necessary In my presentation I shall discuss the syntactic, sequential and semantic-pragmatic properties of this construction, and compare it to other reactive constructions in Swedish, notably x-ochx. I also hope that we will have an opportunity to discuss what the German counterparts of these constructions might be. References: Lindström, J. & Linell, P Roli å roli : X-och-x som samtalspraktik och grammatisk konstruktion. I: Engdahl, E. & Londen, A.-M. (utg.), Interaktion och kontext. Lund: Studentlitteratur Linell, P. & Norén, K. 2008/fc. Vågar vågar ni väl men : Initiala finita dubbla hjälpverb och andra reaktiva konstruktioner i svenskt samtalsspråk (och annat interaktivt språkande).
6 Niklas Norén, Linköping: Apokoinou in Swedish talk-in-interaction To be supplied Camilla Wide, Helsinki: Freestanding that-clauses in Swedish In my paper I discuss the use of free-standing that (att)-clauses in Swedish conversations. The use of subordinate clauses initiated by the subjunction att that in contexts where no preceding matrix clause can be found has been discussed by Lehti-Eklund (2002), Anward (2003) and Lyngfelt (2003). The following example (1) from Anward (2003) illustrates the phenomenon: G: =a: de har gått rätt- han har blivi rätt yes it has gone rather he has become rather stor nu överhuvudtaget tycker ja, big now generally think I V: för ja tyckte han va rätt stor då? because I thought he was rather big then G: a[: PRT _ V: [att han liksom dro iväg that he like pulled away According to Lehti-Eklund (2002), free-standing att-clauses are used primarily in elaborations and conclusions. Lehti-Eklund focuses on Finland-Swedish conversations, where the use of free-standing att-clauses seems to be particularly frequent. The focus in my paper is on Sweden-Swedish conversations, but I also comment briefly on the typically Finland-Swedish uses of free-standing att-clauses. The main purpose of my paper is to describe the contexts in which the freestanding att-clauses are used and discuss the communicative function of the clauses in these contexts. In addition to this, I want to raise the question of whether and how the free-standing att-clauses in Swedish could be analyzed from a Construction Grammar point of view. Anward, Jan, att. Språk och stil Lehti, Eklund, Om att som diskursmarkör. Språk och stil 11: Lyngfelt, Benjamin, Samordnande att en talspråklig sambandsmarkör sedd från ett syntaktiskt perspektiv. In Texten framför allt. Festskrift till Aina Lundqvist på 65-årsdagen den 11 september 2003, Göteborgs universitet: Institutionen för svenska språket.