We present first results from an ongoing study of a particular kind of ‘free’ or ‘independent’ infinitive (Visapää 2022; Wiemer 2017) in German and Polish talk-in-interaction. Data come from corpora of naturally-occuring interaction. Examples (1) and (2) provide illustrations of the kind of infinitive turn we are concerned with.

1. Laying table (Polish) 
01 Mum: 	To   brać          to (.) tam   ju:ż? 
		PTCL take.IPFV.INF this   there already 
		So (should I) take this over there already? 
02 Dad: 	No- 
		yes 

2. Clearing table (German) 
((Dad gets up from table, rest of the family has just left the table)) 
01 Dad:	 	so ↑tisch abräumen; 
		PTCL table clear.INF 
	        right, (can you) clear the table (please) 
02 Lia: 	>↑ja< mach ich; 
	 	yes, will do 

The infinitive turns in the first lines clearly accomplish very different social actions: requesting information (and possibly offering to do the work) in (1); directing others’ behaviour in (2) (Deppermann 2006). While we find infinitive directives in both German and Polish, infinitive questions are prominent in Polish, but not German talk-in-interaction. We are interested in the extent to which such infinitives can nevertheless be understood as being one and the same practice, with language-specific elaborations. The following formal and functional commonalities make a joint consideration of cases like (1) and (2) plausible: In both usages, the infinitive (a) is the only predicate in the turn; (b) participates in the coordination of here-and-now behaviour; and (c) would accomplish the same action if the entire utterance consisted solely of the infinitive verb (e.g., question: PL brać?, ‘take’; directive: DE abräumen!, ‘clear’).
Our analysis starts from the interactional-linguistic finding that language practices belong in specifiable sequential environments or positions (Schegloff 1996; Couper-Kuhlen und Selting 2018). We find that what free infinites have in common is that they orient to an external circumstance as the ground for making relevant the action articulated in the turn. Evidence for this analysis come from three sources: (1) Sequential analyses showing systematic differences between the contexts in which speakers select infinitives vs other directive and question/offer constructions; (2) Turn design analyses showing that expanded turn designs orient to the relevance of external circumstances; (3) Normative analyses showing that participants resist the use of infinitives in cases where the directive or question is not accountable in terms of external circumstances. In sum, the analysis uncovers shared principles of situated meaning-making underlying diversity in practices of speaking across languages.

In order to make sense of speech flow, listeners employ similar tactics as with other types of cognitive input: they chunk it into smaller units (Vetchinnikova et al. 2023; Christiansen & Chater 2016). In this paper I present the findings of my doctoral research on how native Finnish speakers segment natural speech flow online, as it is experienced – linear, instantaneous, transitory and irreversible (Auer 2009; Mauranen 2018). My research problem is psycholinguistic in nature: how do listeners chunk up speech in real time. To answer this question, I combine two different approaches to language and the methodologies they provide: those of interactional linguistics and Linear Unit Grammar (Sinclair & Mauranen 2006), as well as methodology from phonetic and syntactic research. The data was collected in a listening experiment from 51 native Finnish speakers, who listened to speech extracts from expert interviews and panel discussions, and were asked to intuitively mark boundaries between chunks. Statistical methods were implemented to define the unanimity between participants, but my approach is qualitative and data-driven.

In this paper I investigate multi-unit turns from the viewpoint of perceptual chunking, with a special focus on prosody and syntax. I will dig into the interplay between prosodic and syntactic units and their boundaries, showing how the occasional mismatch contributes to the make up of multi-unit turns as well as listeners’ perception of chunk boundaries. I also hope to bring a new perspective to the discussion of fuzzy boundaries (cf. Barth-Weingarten & Ogden 2021).

Our presentation concerns multi-unit turns (MUTs; e.g. Marian et al. 2021) in a narrative context. By multi-unit turns we mean long and complex turns consisting of several turn-constructional units (clauses, phrases). They have also been called never-ending sentences (Auer 1992; Laury & Ono 2014). They flout the conversational constraint allowing speakers only one TCU at a time. They occur in different types of long turns; tellings are one environment for them (see Marian et al. 2021).

Using conversation analytic and interactional linguistic approach, we focus on so-called emerging narratives, that is, tellings that emerge in and from social interaction. According to Jefferson (1978) stories can be “triggered” in the course of turns-at-talk, or they may be projected in advance by a preface through which rights to a telling are accomplished (see also Schegloff 2007: 41–42). In our presentation, we focus on stories that are “triggered” and form incrementally accreted multi-unit turns that emerge in the course of interaction in response to local contingencies.

Our data come from Finnish everyday conversations. In our analysis, we focus on linguistic and embodied resources participants employ to make room for longer turns that turn out to be narrative even though they are not projected as such at the beginning. These resources include grammar, semantics and prosody on the one hand, and gaze, gesture, and body posture on the other. We also analyze how the co-participants’ responses can make room for multi-unit turns.

Studies have shown that participants use turn-initial particles to tie their turns to the previous ones and also to foreshadow what kind of action they will build in the emerging turn (Heritage & Sorjonen, 2018). However, previous studies have focused predominantly on singular occurrences of such particles. In this paper, we build on a dataset of 8.5 hours of interactions among peers in upper secondary English as a foreign language classes in Czechia and Finland. Our collection includes 39 instances of turn-initial (yeah) buts which occurred in 12 task-focused episodes. Building on the principles of multimodal conversation analysis, we will demonstrate that these multiple (yeah) buts function as a resource for argumentative talk as participants use them to calibrate their answers while managing disagreement and as turn-entry devices. Our findings show that students express disagreement or alternative viewpoints quite directly: often latching onto or in (terminal) overlap with the current speaker’s turn, sometimes also with gaze directed at the co-participant. These findings indicate that the students’ task orientation overrides the preference organization (e.g., Kendrick & Holler, 2017), which is closely related to the nature of the L2 classroom tasks that our participants are engaged in, such as comparing different data, sharing personal viewpoints, or inventing a story. Our findings also contribute to the existing body of research on (yeah) buts as turn-initial particles used for expressing disagreement (e.g., Bjørge, 2012; Niemi, 2014, Pekarek Doehler & Pochon-Berger, 2011) by showing how speakers use (yeah) buts across multiple turns.

References

Bjørge, A. K. (2012). Expressing Disagreement in ELF Business Negotiations: Theory and Practice. Applied Linguistics, 33(4), 406–427. https://doi.org/10.1093/applin/ams015
Heritage, J., & Sorjonen, M.-L. (Eds.). (2018). Between turn and sequence: Turn-initial particles across languages. John Benjamins.
Kendrick, K. H. & Holler, J., (2017). Gaze direction signals response preference in conversation. Research on Language and Social Interaction 50(1), 12–32. https://doi.org/10.1080/08351813.2017.1262120
Niemi, J. (2014). Two ‘yeah but’ formats in Finnish: The prior action engaging nii mut and the disengaging joo mut utterances. Journal of Pragmatics, 60, 54–74. https://doi.org/10.1016/j.pragma.2013.10.008
Pekarek Doehler, S., & Pochon-Berger, E. (2011). Developing ‘Methods’ for Interaction: A Cross-Sectional Study of Disagreement Sequences in French L2. In J. K. Hall, J. Hellermann, & S. Pekarek Doehler (Eds.), L2 Interactional Competence and development (pp. 206–243). Multilingual Matters.