Papers

‘Building sustainable communities through service encounters in Keech Charity shops’

by Ann Weatherall | University of Bedfordshire

The United Kingdom has agreed to the 2030 United Nations (UN) Sustainable Development goals.  They are an ambitious blueprint for a prosperous future for British citizens and planet Earth.  One aspect of the plan is a determination to promote sustainable consumption.  A rapidly growing sector that supports sustainable consumption is charity retail – the setting of the present research.  The participating organisation is Keech who provides hospice and palliative care funded, in part, through their second-hand shops.    We ask how the organisational values of inclusivity, sustainability and integrity are manifestly observable through member’s orientations and practices.    The data are video recordings of 200 interactions between servers and customers in two charity shops differentiated by the socio-economic status of their location.  Using a multimodal conversation analytic methodology, assessments emerged as phenomena of interest.  Mundane practices including remarks on the weather, on the royal minting of coins and on purchases, show the encounters are more than instrumental economic exchanges.   They indicate charity retail not only promotes sustainable consumption but also sustainable communities, where caring and shared good humour are a pervasive feature of social interactions.

‘Eating together? Co-eating practices in children’s preschool lunches’

by Sally Wiggins Young | Linköping University

Eating together is one of the most fundamental (social) human practices, yet we still know very little about what it means to eat ‘together’. While there is an extensive volume of research on concepts such as commensality, the empirical details of eating ‘with’ others are often lacking. Much research assumes an individual eater even when in social settings. This is particularly the case for children who, as dependent on others for food and care, are often obliged to eat with other people. In this presentation, I take up the topic of eating together and consider the contribution of interactional research within a much broader research landscape. Using video data from a research project on preschool (kindergarten) lunches in Sweden, as well as building on a foundation of EMCA/DP research on eating practices, I will demonstrate how interactional research not only enables us to empirically unpack the ‘togetherness’ of eating but also to define the analytical boundaries of the field. I furthermore propose that the seldom used concept of co-eating be respecified to define the field and foreground the interactional nature of eating together.

‘Headroom and Reset: Issues in the management of (family) conflict’

by Alexa Hepburn | Jonathan Potter | Rutgers University

Conflicts often follow a pattern of escalation, where participants gradually increase the intensity of their interactions. This escalation of conflict can reach a point where further intensification becomes difficult or impossible, a situation we term “running out of headroom”. This occurs when actions progress along a cline of increasing intensity, ultimately leaving little room for any further escalation.

One effective way of managing this is to perform an “interactional reset”. This strategy involves moving the conflict to an earlier place in the cline or changing the nature of the action sequence. We will illustrate the notions of headroom and reset using two examples: one where one parent transfers behavioral management responsibility to the other, allowing a reset in the intensity of the conflict; and another where a young child attempts to initiate a new action sequence with an audibly angry mother.

Through these examples, we will illustrate the value of the notions of headroom and reset within the dynamics of conflict. Additionally, we will explore the broader implications of adopting an emotionographic perspective (Potter and Hepburn, in prep), which emphasizes the role of emotional displays in shaping action formation.

‘Building trust to study trust: Paving the Way for Applied CA Research with vulnerable participants’

Mie Femø Nielsen | Ann Merrit RIkke Nielsen | University of Copenhagen

Keywords: Project design, applied CA, interventional studies, trust, trustworthiness

Accessing the field is a key challenge for any researcher collecting naturally occurring data. In applied CA (Conversation Analysis) studies, establishing trust becomes a crucial foundation for successful collaboration, particularly when working with participants in vulnerable situations. The TIES project is an interventionist applied CA research project (Antaki, 2011) focusing on interactional trust work in institutional encounters between social workers and vulnerable citizens in an open prison, a social psychiatric supported housing facility, and an NGO assisting current and former prisoners reintegrating into society. As trust plays a critical role in supporting individuals in vulnerable situations (Banks, 2004; Butler & Drakeford, 2005; Warming, 2019), and empirical research on trust in applied settings is scarce, TIES aims to achieve two objectives: first, contribute to an EM/CA-based conceptualization of trust as a practical, social interactional phenomenon(González-Martínez & Mlynář, 2019; Antaki & Finlay, 2013; Jørgensen, 2018; Watson, 2009); and second, in collaboration with practitioners, develop empirically-based reflective training material for social workers, helping them build trusting relationships more effectively.

In this paper, we detail the project’s design and our experiences in cultivating trust with participants, from the initial developmental stages of finding partner institutions to the measures taken at each step to build trust—particularly showing how we unintentionally oriented to components of both our own trustworthiness as researchers that of the participants (Nielsen & Nielsen, 2023). Furthermore, we discuss how this fostered collaboration with both practitioners and citizens and facilitated co-ownership of the project and how this enhances the impact of our findings. Finally, we address and discuss the ethical and moral considerations of conducting interventionist CA research in settings involving vulnerable populations (Fine, 1993; Müller et al., 2022; Nielsen, Nielsen & Due, 2022), as well as the implications of operating within contexts of heightened political attentiveness.

‘System Justification in Action: Discursive Constructions of the Status Quo in Parliamentary Debates Concerning Brexit’

Alex Hunt | Stephen Gibson | Heriot-Watt University

Keywords: Political Ideology, Status Quo, System Justification, UK Parliament

Background:

System Justification Theory (SJT) predicts that those aligned with right-wing ideology are motivated to uphold the status quo, while those with left-wing beliefs challenge it. However, there is limited research considering how system justification unfolds in social interactional settings. The present study aims to address this by exploring how speakers achieve rhetorical goals when they talk about the status quo.

Methods:

Our data were drawn from the official Hansard records of UK parliamentary debates, from which we selected material from the House of Commons relating to the ‘European Union (Withdrawal) Act’. We selected 22 political debates which we coded into 102 pages of data and analysed them using a discursive-rhetorical psychological framework to highlight how system justification is enacted in discourse.

Analysis:

The analysis demonstrated that politicians from left and right-wing parties constructed the status quo and depicted themselves as either preserving or challenging it to justify and criticise positions. Politicians notably criticised opposing positions for threatening to undo democratic systems, conversely, other politicians attempted to frame contrasting positions as seeking to uphold undemocratic systems.

Implications:

The reconceptualisaton of system justification in discursive-rhetorical terms provides a more nuanced understanding of how ideology relates to system justification. People who claim to justify or challenge the status quo can be understood as seeking to achieve social actions rather than passive beings who obey their ideological beliefs. The present study therefore shows that system justification can be respecified from a static cognitive-motivational construct to a dynamic process enacted in discourse.

‘Voices of the Future: Children’s Encounters with Trees’

Johan Siebers | Middlesex University London

Keywords: Treescapes; Climate Change; Ecolinguistics; Interaction among Children and Young People; Hope

Our woods, forests and trees belong to our children. However, the current long-term harming of our Treescapes is leading to loss of animal and plant life and lower environmental quality. Young people in the UK will be deeply affected by this, and they will be tasked with fixing these problems. And yet their voices are rarely heard in policy and practice, even with growing youth activism. The UKRI funded project Voices of the Future (2021-2024) linked young people’s hopes to ground-breaking science on how trees adapt to and limit climate change. In addition to this, researchers looked at how we can change the present and reimagine the future of treescapes for the benefit of the whole UK.

We brought together children, Natural England and Community Forests to look at how young people connect with our woods, forests and trees. We explored the ideas and experiences of children from early years up to secondary school age. This critically involved chidren with limited access to treescapes, such as those from racialized, migrant- background, and socio-economically disadvantaged communities.

The results will inform educational policies, teacher training, urban planning and treescape design. This project has shown that treescapes are not just environments to consume, but are spaces of connection, involvement and hope.

We collected extensive video and audio data of children interacting with each other in tree workshops, tree planting sessions and forest school lessons. We conducted a conversation analysis of some of these data to understand how connection, involvement and hope are enacted linguistically and rhetorically by the children in their interaction with each other and with trees. In this presentation we present some of the early results of this analysis.

‘Describing and categorising family violence in calls to police’

Emma Tennent | Victoria University of Wellington | Ann Weatherall  | University of Bedfordshire

Keywords: description; categorisation; police calls; family violence

Feminists have long drawn attention to the politics of labelling violence. For example, terms like “violence against women” highlight victims but obscure perpetrators, while “intimate partner violence” highlights the relational context but obscures gender relations. In New Zealand, police terminology reflects legislative change that replaced ‘domestic violence’ with ‘family harm’. However, the language used by institutions (and scholars) can be quite different from how everyday people describe their circumstances.

We examine how people describe and categorise violence in police calls. The data are 200 calls to the emergency and non-emergency police lines in New Zealand that were institutionally classified as ‘family harm’. This term encompasses issues within the family including violence. Recordings were produced by the police and the language is New Zealand English.

We use conversation analysis and discursive psychology to examine how people in real social interactions describe (and contest) events and experiences. The practices of description and categorisation are two fundamental ways participants present themselves, others, and the social world. Ordinarily, people have primary rights to their experience. However, many institutional contexts are marked by asymmetries where institutional representatives have greater rights to categorise laypeople and their circumstances. Police calls are an institutional context where call-takers have category-bound rights as institutional representatives to decide whether callers’ problems are legitimate for police help.

We examine how callers and call-takers build a joint understanding of circumstances that come to be classified as family harm. We find that callers use descriptions and categorisations to present their problem; and that categorisations are regularly targeted by call-takers who pursue descriptions. We argue that these patterns demonstrate an orientation to categorisation as an institutional entitlement.

‘Interactional challenges and practices in recommending aids and changes to patients’ activities of daily living in palliative occupational therapy and physiotherapy’

Elliott Hoey | Vrije Universitait Amsterdam | Ruth Parry | Loughborough University and Treetops Hospice, Derbyshire

Keywords: Conversation analysis; palliative care; communication skills, activities of daily living; occupational therapy; physiotherapy; shared decision making

Health and social care guidance and research about aids and adaptations to activities of daily living largely focus on physical design and uses. However, therapist/patient conversations about aids and adaptations entail social and relational challenges, and rejection of recommendations is common. Resistance jeopardizes therapeutic collaboration and can mean patients miss out on benefit now or in the future.

We set out to investigate and describe practices through which therapists circumvent or manage the social interactional challenges in recommending aids and adaptations.

We searched the VERDIS Allied Health Professions dataset for episodes wherein occupational therapists and/or physiotherapists recommended, or worked towards recommending, aids and adaptations. We found 39 episodes involving 39 patients, 18 family- members/carers, 5 occupational therapists, 3 physiotherapists. We used conversation analysis to investigate the structure and functioning of therapists’ vocal and embodied practices for introducing recommendations and responding to resistance.

We identified seven practices used to introduce interventions cautiously, and present recommendations as negotiable. These practices reduce pressure to decide immediately. Nevertheless, resistance (implicit or overt) was evident in 22/39 episodes. We identified seven ways therapists responded. Five entailed further talk (produced by the therapist or elicited from the patient) about the intervention or about the resistance to it, thereby keeping the intervention a live matter. Two involved accepting the resistance and moving on. The episodes we studied sometimes involved reference to patients’ deteriorating illness trajectory.

The practices we describe head-off or mitigate rejection of therapeutic recommendations, keeping interventions an available option for the future. Where recommendation episodes involved reference to future deterioration, the interactional sensitivity and effort involved were heightened. At the same time, talk on the topic of future deterioration presented opportunities for patients and family to convey their associated awareness and feelings.

‘Staging Conversation with AI: Human Demonstrators’ Interactional Work in a GPT-4 Live Demo’

Yujin Shin, | Yo-An Lee | Sogang University

Keywords: Conversation Analysis, Human-Computer Interaction, Live Technology Demonstrations, Turn-Taking

According to the scoping review by Mlynář et al. (2024) that covers over 50 artificial intelligence (AI) related EMCA research papers, the most frequently studied AI technology is robots, followed by voice user interfaces and virtual agents. Studies often involve participants interacting with AI in settings like public spaces, homes, or experimental setups, with data collection methods ranging from naturally-occurring data to lab experiments. Yet, there remains a gap in analyzing live demo videos that trace sequential progression of talk in real-time interaction. Live demos are staged interactions by developers and enterprises, designed to showcase and demonstrate state-of-the-art technologies. For this reason, they might be selective in what users say or even avoid practices that display disruptive repair sequences. This is why it is significant to trace the sequential courses of interaction to realistically represent real-time interactions and identify what is possible and what becomes problematic in interactional exchange.

Using conversation analysis as a methodological framework, this research examines how conversation with GPT-4 omni is constructed on stage by human demonstrators during a live demo. From a live demo video showcasing OpenAI’s LLM model, GPT-4 omni and its voice interaction capability (video open to public, see references), this research uncovered how the demo is organized to facilitate and highlight the machine’s competence from the perspective of sequential analysis.

The analysis uncovers three key patterns in members’ conduct that become visible through sequential analysis. First, opening sequences show human demonstrators compress the opening sequence into a turn or two in a packed format containing multi-unit turns (summons-answer, identification, greeting, and ‘how are you’ exchanges), which then become progressively simplified (in subsequent exchanges, some of these greeting sequences are bypassed to get to the reason for interaction). Similarly, closing sequences deviate from ordinary conversation—there are no pre-closing moves or co-achieved closings; instead, demonstrators unilaterally terminate the interaction by seeking a timing to press the ‘x’ button. Second, strict orientation to A-B-A-B speaker alternation is actively maintained through demonstrators’ careful coordination, where two human presenters collaboratively occupy position B in the turn-taking sequence, signaling to each other when to speak to preserve the alternating pattern. This is also observable when demonstrating a visual capability where the demonstrator has to verbally ask GPT when they finish writing. Third, laughter is handled differently: while it functions as social work among human demonstrators and audience, it is systematically excluded from system-directed talk, treated as invalid audio input.

While these patterns seem to arise from the staged nature of demos, similar ‘stage managed’ interactions occur, according to the literature, in domestic settings with smart home devices, where “people regulate their actions as part of the ongoing social order” (Reeves & Porcheron, 2023, p. 581). This suggests that when examined in close sequential detail, it is the human users who adjust and adapt to the parameters of the chatbot system, which limits the scope and manner of managing turn-taking.

This study contributes to the existing literature by documenting how established CA concepts like turn-taking and sequence organization are adapted when humans interact with state-of-the-art AI system. Ultimately, through Conversation Analysis, it poses a question as to whether recent technology in NLP and its engineering methods show a different analytic route than CA in replicating human interactional exchanges.

‘Preliminary results of interactional functions of expletives and swearing in the Italian instant messaging corpus WhaP!’

Maria Eleonora Sciubba | Tilburg University | Ilaria Fiorentini | Università di Pavia

Keywords: Italian, WhatsApp, digital CA, instant messaging, swearing

Swearing has often been connected to “bad behavior” and incivility in both online and face- to-face interactions and considered restricted to particular settings and categories of users, generally toward the lower end of register and class. Expletives have been found to sanction noncooperation in interaction, but participants can achieve sanctioning not only by framing a turn as “sanctionable” but also by co-constructing a shared jocular dimension where teasing is involved, and a problematic element becomes a “laughable”.

Therefore, swearing might serve a more positive social purpose of building intersubjectivity and affiliation in interaction or emotional management, and may also favorably affect hearer’s perceptions of the speakers proffering profanities. Studies in EMCA have mostly focused on the transgressive status and face-to-face interactional restriction around improprieties that may, or may not, positively affect intimacy.

Previous studies on swearing online have mostly focused on hate speech and aggression, but to our knowledge, there is no study of swearing on online platforms from an EMCA perspective taking into account the use of swearwords to build rapport. The present paper aims to help disambiguate the contexts in which swearing is disentangled from hate speech and aggressive behavior, by focusing on participants’ observable orientations to online swearing, and building on previous research on swearing, intersubjectivity and affiliation.

We will use digital CA to analyze if and how the affordances of instant messaging apps like WhatsApp constrain or not interactional practices around swearing. The data come from the corpus WhaP, containing 58 WhatsApp chats (ca. 22K words) and 211 voice messages (being transcribed), and the language is mostly Italian. The corpus spans 3 years of chats (2020-2023) and contains both two- and multi-party conversations among students.

This research is a work in progress.