‘The stuff that happens while we’re having CAke'

CA Day, held at Loughborough University each year (barring a Covid break) means Conversation Analysis day. This time round, it also meant Charles Antaki day, celebrating his 'pivot to Emeritus Professorship'.

Expertly organised by Loughborough pair Saul Albert and Elizabeth Stokoe, the interdisciplinary and hybrid event was very much in Antaki's image. Keynote speaker Steven Bloch (University College London) thanked him for all those 'opportunities for collegial exploration and discussion', noting 'you just always seemed to have an opinion about things, whatever they may be'. So much of collaboration, Bloch noted, is 'the stuff that happens while we're having cake… that's where the work gets done, that's where interaction really happens.' And that is CA Day (there's even a CAke Off).

Affiliation and knowledge

First presentation was Catrin Rhys, Maria Erofeeva (Ulster University), and Bethan Benwell (University of Stirling), analysing how complaints are handled on the phone. In ordinary conversation, complaints usually prompt 'affiliative uptake': we take a stance that matches the teller's stance towards the event(s) they are describing. Or we might offer 'informational uptake', which might also serve to 'decouple the hurt from the blaming'. If we fail to do either – for example, if a complaint is followed by a rather noticeable pause of 1.5 seconds – there can be an escalation of blame. The call handler might just be following their own institutional goals, but not taking up that stance is likely to lead to heightened emotional valence for the caller. Implicit hurt quickly becomes explicit hurt, and implicit blame turns to explicit blame. Instead, it is often possible for a call handler to justify the hurt as a minor incident with major consequences; or to use 'reasonableness as a missing link between the hurt and the blame'.

Next up, Yarong Xie (University of Edinburgh) was granted permission by Mumsnet to analyse 72 posts from their discussion forum. This showed that in reporting racist incidents via Mumsnet, what counts as 'being a good mum' is constantly negotiated and mobilised. Complaints about child's behaviour quickly becomes complaint about other mum's behaviour ('I'm guessing that she's heard some prejudice at home…'). The suffering of one's hurt child becomes a license to post racist problems on Mumsnet, with educating children about race/ism and protecting children discussed as mother's duty.

Complaints were also the data for Andrea Rodriguez and Valeria Sinkeviciute (University of Queensland), drawn from 'casual family phone conversations'. Their analysis was grounded in the epistemics of social relationships: broadly speaking, who knows what. If we complain about a family member, our expectations around whether or not the other person will agree with us are shaped by these epistemic imbalances.

Knowledge was also the fore in a talk from Matthew Butler (University of York), who showed footage of both Tommy Robinson and Matt Hancock being interviewed on breakfast TV. The presenters 'recruited common knowledge' in response to evasions or nonanswers; for example, Piers Morgan chivvying the conversation along with 'Yes we all know what's been happening'. Matt Hancock's attempt to account both for not answering, and for his inability to answer – 'I don't know anything about the party' – was challenged by Susanna Reid with 'Don't we all know everything about the party now?'.

Power

Knowledge is power, or so the saying goes, and a few of the presentations explored how that manifests in conversation. Kate Steel (University of the West of England) studied police-victim interaction during first response call-outs to domestic abuse incidents, via audio recordings extracted from body-worn video footage. For various reasons, she only had a 24-hour window with the footage, and her description of poring over the recordings to get as rich and multimodal transcription as possible was a vivid reminder of the value of data in this field, and the rare access to it. Steel's examples showed how the conversations 'co-constructed blame and responsibility', and also co-constructed police expertise. Phrases such as 'are you together or not', while aimed at establishing facts that are necessary for police procedure, can be taken as problematising a victim's claim. The victims themselves orient to both self-blame and blame from others, and may address this in their talk ('he just came out of nowhere'). Steel illustrated how the power relationship between speakers is 're-negotiated on a moment-to-moment basis', to either compound victims' vulnerability or to encourage their confidence in police support.

Police support was starkly lacking in a conversation analysis examination of bystander interventions in the murder of George Floyd, from Chris Walton (Lancaster University). 'We've got a lot of theories about bystander intervention, but what we're in search of is a good description,' he began. So, how do you stop the police from killing someone? 'I won't be playing any of the video and audio data,' Walton said: 'I've sat and listened to those things enough' – a reminder of the often sensitive nature of conversation data. (Bloch said later, 'a lot of the work we do is very emotional, and has a moral dimension'). Snippets from the crowd that gathered around Floyd and the police officers spoke volumes: declarative statements such as 'You can't win brother', assessments like 'he's not responsive right now bro', interrogatives asking 'does he have a pulse?', imperatives, insults and more.

One of the first interventions was from Charles McMillan, telling Floyd 'the man he gonna win'. 'The man' in this situation was very much not Floyd, who responded 'Man I know… I don't wanna try to win'. McMillan was projecting the outcome of the interaction as already known and inevitable, yet Walton pointed out that the analysis shows that although the absence of response from the officers 'is marked', some – Thomas Lane in particular – were at times responsive to the interventions. 'It could have gone different,' Walton concluded. The success of any verbal bystander intervention, he said, depends upon its power to significantly reshape a wider sense of context and to effectively overcome alternatives. Walton noted that we did see examples of this, for example Genevieve Hansen chipping in with 'I'm a Minneapolis…' before correcting to 'I'm a firefighter from Minneapolis': an 'appeal to epistemic status', highlighting that she has the knowledge necessary to intervene. Ultimately, the data 'doesn't speak for the power of the group in the way we might want it to [based on previous work]', Walton admitted.

On a lighter note – at least as I experienced it – Cat Holt (University of Exeter) delivered the splendidly titled 'You're an idiot': Using insults and harsh criticism to foster the development of staff in high quality leader-member exchange LMX relationships. This was a wonderful window onto how the construction industry are managing via Zoom, with conversational extracts such as 'consider that as a verbal slap', 'is that your girlfriend who's walked in? Tell her I've just given you a telling off', swearing, joking, informal language, and chat about home and family life. Holt showed how a leader would tend to admit and explain the harshness, as 'it's for your own development'. 'I found evidence of "toughness" and "love" – the followers accepted the harsh criticism because they understood that the leader supported them and had their best interests at heart.'

The ordinary becomes extraordinary

What has always pulled me to Conversation Analysis is that it's about noticing the significant in the mundane. As Bloch said of working with Antaki, it 'has taught me compassion in understanding how people do everyday things, and the complexities of what they are doing'. That moment of what Bloch called 'our first noticing' must be a magical thing, like mining precious stones.

Antti Kamunen, Tuire Oittinen, Iira Rautiainen and Pentti Haddington (Oulu) worked with people who seem to be professional noticers, on the UN Military Observer course. Their question was 'when does the analysis begin in conversation analysis?', showing the role of 'accumulated ethnographic knowledge in the recognition and analysis of interactional phenomena'. Contextual knowledge, including a researcher's own subjective experience of the studied context and events, can be 'proto-data', working new aspects – in this case, the 'unmotivated looking' the observers do – into the process. 'Recordings are not the whole truth', they warned.

Doctor-patient consultations are another context where power, knowledge and morality collide, and are thrust to the surface in words as mundane as 'So I can go to work?' Aija Logren (Tampere) studied video data to determine who initiates talk about sick leave. 'Windows of opportunity to bring up sick leave'. For the patient, important 'windows of opportunity' are early in the consultation – around their reason for visiting, their history, the physical exam. (As an aside, I was struck a few times during CA Day that if a person begins to talk about sleep, and sleep disruption, you know the chat has shifted up a notch.) Doctors tend to bring it up in the later stages, for example around treatment delivery. The patient manages their stance to avoid social/moral inferences: 'They tend not to give the impression that getting sick leave was on their agenda. Instead it's a necessity, to be deduced out of the information they provide, rather than them explicitly requesting it.'

The second keynote, from Leelo Keevallik (Linköping), was a fitting closing example of the ordinary, that most people – most psychologists, even – might pass over. Have you ever noticed how people can sound empathetic in their interactions with you, even without saying words? Me neither. Keevallik has, and she has videos from infant feeding, primary care consultations and yoga instruction to unearth this. In 'non-lexical vocalisations in interaction', or 'sounding for others', one person vocalises to enact someone else's ongoing bodily sensation. It's a 'collaborative performance of sensorial experiences': during infant feeding (Sally Wiggins' work) that might include sucking lips in, putting the tongue out, vocalising 'mmm'. When discussing pain (Ann Weatherall), a GP may use a sharp inbreath or even an empathic pain cry. These sounds may subsequently be elaborated in words, so it's an illustration of what Keevallik called the 'fuzzy boundary between language and non-language'.

Doing being helpful

Back to Steven Bloch, whose keynote wove together how staff on a Parkinson's helpline do 'being helpful', and personal reflections on how Antaki has done the same for him and so many other academics.

Bloch recounted hearing that an acquaintance worked on a Parkinson's helpline: 'Data!', he thought. 'It's about getting a foot in the door, and showing the value of stuff we do.' Granted access to 30 calls, Bloch noticed that the way callers present problems is a socially organised practice. Stating the impact of the problem, or symptom(s), is one way to signal that a problem presentation is complete. There may be an 'upshot (pivot-point) statement', such as 'it has disturbed my sleep to such an extent…', which is calling out to be met with an affiliative response such as 'So these symptoms are getting in the way of life, aren't they?'

The nurses on the helpline, though, are in a tricky position. They're not really there to give medical advice, but they tend to manage this by 'adopting a low deontic stance': marking their own authority as being in some way diminished or constrained, for example by saying 'I'm not a doctor, and I advise you to go and see your doctor'… but then, crucially, following this up with 'What the doctor might tell you…' The nurse may also try to manoeuvre an objection (to perceived advice) onto grounds where their own expertise will win the day. For example, 'I don't see the consultant very much actually,' may be met with 'no I think where there's a nurse I mean often the nurses can provide just as much if not more than the consultant…'

It struck me that many of the people found in the CA data are validating, warm and kind – maybe with the exception of the 'you're an idiot' guy, but even his heart was apparently in the right place. And those are terms used by Bloch about Antaki, who in his typically humble way was keen to 'invert the credit' in that 'beautiful but largely untrue narrative'. In closing, Jonathan Potter paid warm tribute to Antaki, noting that he has always been egalitarian, never trailing behind the big stars… he is always connecting people, nurturing the marginal so that they became less marginal. 'He loves to gossip, but is never cruel. Always a humanitarian. Charles has always cared about people and felt for people.'

He is also, it was noted, 'a bit of a wag'. Stokoe recounted how she was sent into a panic on the eve of a British Psychological Society accreditation visit by a request to translate all materials into German, from one Professor Hermann Zermitz. In a parade of video messages, CA luminaries thanked Antaki for being, in the words of Barbara De Cock, 'a man for all seasons'.

Conversation Analysis is often about voice, and as Bloch noted, 'Charles' writing is extraordinary… it has a voice, and I now recognise that voice.' It's a voice that permeates an entire field, and beyond. It's the voice of an 'intellectual superstar', but perhaps more importantly it's the voice of a man who has for a long time been, in the words of Mick Finlay, 'kind, and thoughtful, and funny'.

Photo: Charles Antaki (centre) with the Loughborough Discourse and Rhetoric Group.

Watch videos from the day via the timetable links here.