In studies of social interaction, displays, negotiation and transmission of knowledge and the emergence of common ground usually are researched on the level of sequences of interaction, e.g., question-answer, instruction-compliance, etc. However, it is evident that common ground between participants emerges over extended series of interactions as well. In addition to sequences, cross-sequential and cross-event relationships of adding to, updating of, and revising common ground play an important role for the emergence of shared knowledge, social relationships, and the accomplishment of joint action. However, interactional histories that span over series of interactions have only very rarely been studied yet, and they pose very specific requirements on data-sampling.

In my talk, I will report on a study of the emergence and change of common ground over the course of psychotherapy sessions. I will track how therapist and patient, starting with divergent understandings of the sources of patient’s problems, come to develop a shared view over iterated topicalizations of the same conceptual domains pertaining to the patient’s problems. I will show that the development of common ground not only has a prospective, emerging dimension, but also a retrospective dimension: Participants presuppose and refer back to concepts, positions, and agreements that have been established on prior occasions by various explicit and indexical means. In this way they make use of their shared interactional history as a resource of economical and partner-specific recipient design that rests on shared meanings, which at times can be intransparent to an overhearer who does not know their history. Data come from psychodynamic psychotherapy in German.

eben can be used in Standard German as adjective (meaning ‘flat’), as time adverb (‘just’), as intensifier, as answer particle, or modal particle (Felder 2017). As answer or modal particle, it can also be relevant for evidentiality, as it can express that a proposition is common knowledge or that it can be inferred from what has already been said (Brausse 1986, Karagjosova 2003). In these cases, eben links an utterance to its context on the level of evidentiality.

In our talk, we focus on interactional uses of eben relevant to the negotiation of evidentiality. We suggest a way to combine context-sensitive quantitative analysis with qualitative analysis on the microlevel of conversations within the context of oral school tasks. In our corpus of 180 videorecorded discussions (873 minutes in total) of Swiss German elementary school children (aged 7–12), eben occurs 900 times, of which 473 are turn-initial or constitute a single-word turn. We annotated the entirely transcribed corpus with context-sensitive codes for argumentative actions (e.g., oppositional/validating justification or (dis-)agreement in relation to a previous statement, etc.), 38 conversations have been annotated additionally with regard to different devices used for positionings or justifications.

Analyzing our data, we draw on the GLOBE-framework (“globality and locality in the organization of jointly constructed units“, Quasthoff/Heller/Morek 2017) for describing oral argumentation competence (Heller 2012). In this understanding, eben can be seen as a ‘linguistic form’ within different ‘pragmatic devices’ (like ‘giving a statement’ or ‘supporting the own positioning with a justification’, etc.) to process interactional jobs like ‘positioning’ or ‘justifying’. At the same time, eben can be related to negotiations of epistemicity, as it is often used at the beginning of turns to upgrade one’s epistemic authority by supporting a claim or making a conclusion explicit while affiliating with the previous speaker; a dual function of agreeing with the previous turn and at the same time pointing to an already displayed or postpositioned position (cf. Betz/Deppermann 2018).

In our talk, we will ask which correlations of turn-initial eben with our codings can be observed, how these observations can be related to different interactional uses of eben, and how their sequential embedding relates to that of the information they are intended to update.

Evidentiality refers to the grammatical, lexical, textual resources (for Italian, e.g. Pietrandrea 2007, Rocci 2012, Miecznikowski 2016) whereby the speaker categorizes his/her sources of knowledge for a proposition P in assertive utterances and, from an interactional perspective, manages his/her epistemic positioning in various sequential contexts (recently, Cornillie/Gras 2020).

In this paper, we take a step further towards an interactional account of evidentiality. Following recent proposals on the temporality of grammar (Mushin/Pekarek-Doehler 2021) and on categorization as an incremental and collaborative activity (Mauri et al. 2021), our goal is to investigate how the production of evidential resources in spoken discourse unfolds in time to achieve source categorization.

We use conversational data from the TIGR corpus (6h) and the Kiparla corpus (16h) of spoken Italian, and combine systematic annotation of evidential resources with sequential analysis of selected phenomena.

In this contribution, adopting the methods of Interactional Linguistics and Conversation Analysis, we scrutinize a subset of evidential resources (around 30% of the total dataset) which are produced within increments or self-repairs. These practices allow the speaker to modulate the degree of explicitness and specificity of P’s sources and to negotiate epistemic positioning, targeting and preventing possible troubles in the epistemic affiliation with the interlocutor.

Our findings overall suggest that the production of evidential resources is finely tuned to the sequential and epistemic organization of talk and interplays with practices in the incremental construction of the turn and the sequence.

Example

BO046: ma, maurizio quando parte?
when is maurizio leaving?
BO021: il venticinque[P].
on the twenty-fifth.
BO046: ah,
oh,
BO021: dovrebbe[E], se ho ben capito[E], perche' dovrebbe avere tipo un esame il ventotto o una cosa del genere[E]
he should, if I understood well, because he should have like an exam on the twenty-eighth or something
BO046: e giada resta?
and is Giada staying here?
BO021: penso[E] di si’[P]. (-) cioe' io da come l'ho capita[E], si’[P].
I think so. (-) I mean, as far as I understood, yes.

References

Mauri, C., Fiorentini, I. Goria, E. (2021). Building Categories in Interaction. Linguistic resources at work. John Benjamins.

Miecznikowski, J. (2016). An experience that apparently differs a lot from mine”. Evidentials in discourse: the case of gastronomic discussions. In S. Greco/M. Danesi (Eds.), Case studies in Discourse Analysis. Lincom Europa, 270-298.

Mushin I., Pekarek-Doehler S., (2021), Linguistic structures in social interaction: Moving temporality to the forefront of a science of language, Interactional Linguistics, 101(1), 2–32.

Rocci, A. (2012). Modality and argumentative discourse relations: A study of the Italian necessity modal dovere. Journal of Pragmatics 44(15), 2129-2149.

Pietrandrea, P. (2007). The grammatical nature of some epistemic-evidential adverbs in spoken Italian. Italian Journal of Linguistics 19, 39-64

Squartini, M. (2008). Lexical vs. grammatical evidentiality in French and Italian, 46(5), 917-947.

Depuis une dizaine d’années environ, se sont développées, dans le champ de la formation des adultes, des démarches de formation professionnelle, inspirées des principes de l’analyse de la parole-en-interaction, et qui placent au centre de l’attention le travail en tant qu’accomplissement collectif (Filliettaz et al., 2021). Exploitant la pratique analytique de la « data session » (Stevanovic & Weiste, 2017), ces méthodes permettent de développer chez les professionnels en formation une posture analytique descriptive, non jugeante, à même de cerner les composantes collectives et intersubjectives liées à l’accomplissement du travail.

L’objectif de cette communication est d’étudier comment la mise en œuvre de ce dispositif est susceptible de développer chez les participants des savoirs portant sur l’organisation de l’interaction et par quelles « méthodes », elles-mêmes accomplies au sein de la parole-en-interaction, ces savoirs sont susceptibles d’être mis en circulation.

Pour mener à bien cet objectif, nous nous proposons d’examiner comment ce dispositif de formation par l’analyse des interactions a été mis en œuvre dans un contexte particulier, celui de la formation continue d’éducateurs spécialisés actifs dans le champ du handicap. A partir d’une démarche inspirée à la fois de la sociologie des sciences (Knorr Cetina, 1999), de la sémiotique multimodale (Kress et al., 2001) et des théories épistémiques en analyse conversationnelle (Heritage, 2012), notre communication se propose d’exploiter un corpus d’enregistrements audio-vidéo de data-sessions en formation professionnelle, dans lequel des éducateurs spécialisés actifs dans le domaine de l’autisme apprennent à observer et analyser leur travail d’accompagnement.

A partir d’une analyse des traces enregistrées de séquences de co-analyse, nous nous appliquerons à montrer comment le dispositif de l’analyse collective des interactions permet aux participants de s’orienter vers la part interactionnelle de leur travail d’accompagnement. En particulier, nous nous centrerons sur le repérage par les participants d’une compétence interactionnelle spécifique consistant à endosser dans les entretiens d’accompagnement une posture active d’écoute. A partir du repérage des « mhmm » ou des « ouais » qui ponctuent régulièrement le discours des accompagnants, le groupe s’oriente progressivement vers la description de ces « petits mots qui valident », et s’engage collectivement dans un processus dynamique d’enquête portant sur les formats et les fonctions des régulateurs verbaux (Laforest, 1993). Ce phénomène local illustre, de manière plus générale, la capacité des participants à des data-sessions en formation professionnelle à élargir le répertoire de savoirs mobilisés dans la conduite des activités de travail.

Depuis une dizaine d’années environ, se sont développées dans le champ de la formation des adultes des démarches de formation professionnelle, inspirées des principes de l’analyse de la parole-en-interaction, et qui placent au centre de l’attention le travail en tant qu’accomplissement collectif (Filliettaz et al., 2021). Exploitant la pratique analytique de la « data session » (Stevanovic & Weiste, 2017), ces méthodes permettent de développer chez les professionnels en formation une posture analytique descriptive, non jugeante, à même de cerner les composantes collectives et intersubjectives liées à l’accomplissement du travail.

L’objectif de cette communication est de porter une attention particulière à la manière dont la transcription des données analysées par les groupes en situation de co-analyse est utilisée par les participant-es. En tant que production épistémique et multimodale, la transcription est susceptible de jouer un rôle déterminant dans la mise en visibilité des savoirs convoqués dans la situation d’analyse (Goodwin, 2000). Observer la transcription en tant qu’objet sémiotique et épistémique représente ainsi un moyen inédit d’accéder aux différents types de savoirs qui se développent et s’imbriquent lors des séances de formation (ten Have, 2002).

A partir de ce postulat, les questions posées sont les suivantes. Comment les participants en formation s’orientent-ils vers l’objet transcription dans les sessions d’analyse de données audio-vidéo ? En quoi la nature multimodale de la transcription en tant que production sémiotique agit-elle comme une ressource dans la pratique analytique du groupe ? Et quelles actions elles-mêmes multimodales permettent aux participants de faire usage des transcriptions dans les pratiques épistémiques en cours d’accomplissement dans les data- sessions en formation ?

A partir d’une démarche inspirée à la fois de la sociologie des sciences (Knorr Cetina, 1999), de la sémiotique multimodale (Kress et al., 2001) et des théories épistémiques en analyse conversationnelle (Heritage, 2012), notre communication se propose d’exploiter un corpus d’enregistrements audio-vidéo de data-sessions en formation professionnelle, dans lequel des éducatrices de la petite enfance apprennent à animer des séances d’analyse de leur travail en groupes. A partir du repérage des actions typiquement accomplies par les participantes au moment où elles font référence à la transcription (ex. (contre)argumenter un point de vue, réorienter l’analyse, promouvoir une transition topicale, etc.), il s’agira de mieux comprendre comment les usages épistémiques de la transcription agissent à la fois sur les rôles interactionnels endossés par les participants et sur les domaines de connaissances qu’ils convoquent dans la démarche d’analyse collective.

In this program slot, the organizers propose a roundtable with members of the Swissuniversities-funded project "Data-sharing skills in corpus-based research on talk-in-interaction" (https://www.chord-talk-in-interaction.usi.ch/) about the management and reusability of primary and secondary spoken language data, a topic that promises to be relevant for many KNOWINT 2023 authors working on talk-in-interaction.

Here is the abstract of the project:

=========

Audio-video recorded and transcribed corpora of spoken language in interaction are collected by a diversity of disciplines in the social sciences and the humanities. They are particularly complex and offer several challenges when it comes to practices that generate open research data (ORD). The project aims at identifying the conditions that must be fulfilled in order to prepare and treat audio-video corpora of spoken language in interaction as ORD in a meaningful way. The applicants describe, analyse and assess existing ORD practices related to audio-video interactional data, on the basis of experiences and expertise in the fields of interactional linguistics, conversation analysis and dialogue-oriented argumentation studies. They explore possibilities of improvement, with a particular focus on researchers’ data-sharing skills related to corpora of spoken language in interaction and on the Swiss situation. The applicants engage in a theoretical reflection on the challenges that these data raise in terms of ORD and on what this example reveals more generally about ORD practices as socio-technological practices in the scientific domain. They engage with the concerned disciplines and communities and evaluate procedures to disseminate relevant knowledge in these communities.

=========

The roundtable will be an occasion to present the aims and (past and further) activities of the project and to discuss the different topics addressed by the project with the participants to KNOWINT 2023, such as to encourage the sharing of best practices as well as reflexive thinking about the various challenges raised by ORD constraints and opportunities in the study of talk-in-interaction.

The workshops organized by the CHORD-Talk-in-interaction project address the following topics:

• Workshop 1 (28 April 2023, Lugano): presentation and discussion of the KIParla corpus of spoken Italian, with special attention to the management of personal data when studying talk-in-interaction, in a context where there is a tension between the promotion of ORD and the evolution of regulations about data protection.
• Workshop 2 (7-8 December 2023, Basel): aggregation of interactional corpora in larger data banks which can be (re)used for research in Conversation Analysis (CA) and Interactional Linguistics (IL).
• Workshop 3 (15–16 January 2024, Neuchâtel): design of annotation schemes (more or less standardized and adaptable to users’ needs) for long-term research in CA and IL.