Bringing identities to the table: Exploring conversational practices of vegetarians and vegans at flashpoints in interaction with meat-eaters

ABSTRACT

Although vegetarian and vegan (veg*an) diets can have various health, environmental and animal welfare benefits, they remain socially contentious. Despite the fundamentally social nature of eating, in situ investigations of the social-interactional elements of dietary identities have so far been lacking. Using a recently developed remotely moderated focus group design, we explore (across 25 discussions involving 122 participants) the discursive management of veg*an ‘identity flashpoints’ during discussion with meat-eaters. Our discursive analyses explore how these moments in conversation arise and are handled in real-time within the unfolding interaction. We demonstrate how two particularly interactionally consequential features of veg*an accounts of their practice become constructed within these encounters. Firstly, personal accounts of veg*anism are interactionally preferred to moral accounts. Secondly, demonstrating continued liking of meat was necessary to be met with group acceptance among majority meat-eating groups. These features shed light on the social interactional perils of veg*an attempts to engage in persuasion with meat-eaters, particularly those based on moral grounds. Our novel methodology affords detailed analyses of how veg*ans navigate the performance of their identities in their daily interactions with dietary outgroups, providing insight into micro-level processes that might underpin, or hamper, processes of societal change in the dietary domain.

KEYWORDS:

• Remotely moderated focus groups
• Identity
• flashpoints
• dietary
• moral

The consumption of meat and other animal products as a key part of one’s diet is a highly normative practice in many cultures throughout the world. Although meat consumption is justified and constructed by many as natural, normal, necessary and nice (Piazza et al. 2015), the normative and socio-political status of meat consumption (and its avoidance) varies across cultural contexts. In India, for example, meat consumption is often seen as a socially progressive act due to vegetarianism being regarded as an upper-caste practice that reinforces social hierarchies (Srinivasan 2023). Although internet search trend data suggests that interest in vegetarian and vegan diets are growing around the meat-eating majority world (Kamiński et al. 2020), within many cultural contexts, especially in the West, vegetarians and vegans (hereafter referred to collectively as veg*ns) are still likely to be in the minority in many meal settings. Indeed, recent surveys suggest that only two percent of UK consumers are vegan (YouGov 2024), and only six percent of Australians are vegan or vegetarian (YouGov 2023). Due to mealtimes frequently involving veg*ans making different food choices to others, or justifying their choice not to partake in a shared food provision, the identity of the individual as veg*n is likely particularly salient at mealtimes compared to other points during the day. Moreover, this is true in ways that are simply not the case for those in the meat-eating majority, whose dietary identity represents an invisible default that does not require comment or justification. The salience of minority dietary identities at mealtimes, thus, may create identity flashpoints, in two different senses of the term, a) as points in interaction where individuals are required to ‘flash’ their minority practice to outgroup members, and b) as places where tensions or conflict may easily ignite (Kurz et al. 2020).

Opportunities within interaction for veg*an identities to become visible to meat-eaters may have consequences for social change. Research indicates that a person’s meat consumption is considerably lower when a member of their household is vegetarian, and lower still when a person’s social circle includes a vegetarian friend or family member (Vandermoere et al. 2019). This relationship is likely bidirectional, as individuals choose friends with similar values. Nonetheless, these findings point to the potential importance of understanding how the everyday interactions that occur between veg*ans and meat-eaters unfold around the table (and elsewhere). Understanding how members of these dietary minorities manage such identity flashpoints has important implications for understanding broader processes of societal dietary shifts.

Although identity flashpoints present a potential opportunity for veg*ans to influence the majority, previous work suggests that they may represent risky business for dietary minorities because ‘outing’ one’s identity risks negative social evaluation (Corrin and Papadopoulos 2017; Rothgerber 2014). Interview studies report veg*n experiences of teasing, rejection, mockery and attempts by meat-eaters to change their behaviour (Lindquist 2013), with some individuals going so far as to make new friends and shift social groups after embracing veg*anism to avoid such social-interactional encounters (Greenebaum 2012). Surveys and scenario studies with meat-eaters provide further evidence of prejudice, suggesting that omnivore participants rate veg*ans more negatively than other dietary minorities (e.g. gluten-intolerants), and equally or more negatively than other commonly derogated groups, including immigrants and ethnic minorities (MacInnis and Hodson 2017). Although it is worth noting that recent research has suggested that such evaluations tend to be more negative of vegans than vegetarians (Nezlek and Forestell 2020), and recent work in some contexts, such as New Zealand, has shown attitudes towards veg*ans to be less positive (than towards meat-eaters) rather than explicitly negative per se (Judge and Wilson 2019).

Potential prejudice against those who avoid meat may be accounted for by the perceived moral nature of the practice. Minson and Monin (2012) coin the term ‘Do-Gooder Derogation’ to explain their finding that 47% of participants in a free-association task associated vegetarians with negative words, an effect positively correlated with expectations that vegetarians would see themselves as morally superior to meat-eaters. Further research demonstrates that veg*ans may ‘self-silence’ to avoid potential social awkwardness and stigma from meat-eaters (Bolderdijk and Cornelissen 2022), or potentially avoid moralised arguments surrounding their practice (Brouwer et al. 2022). Thus, identity flashpoints may provide a potent opportunity for minorities to ‘flash’ their identities and potentially engage in persuasive attempts and effect social change, but revealing one’s dietary identity may incur risk of negative social evaluation and being perceived to judge others. This may, in turn, elicit defensiveness and reduce the willingness of omnivores to reduce their meat consumption. Therefore, these points within interaction where a minority practitioner’s identity becomes a salient matter of conversational business also represent flashpoints in so much as (drawing on a physics metaphor) they represent a place where tensions or conflict may easily ignite if not managed appropriately (Kurz et al. 2020).

In response to these potential social-interactional challenges, qualitative interview studies have investigated the strategies veg*ns report adopting to deal with them. In Greenebaum’s (2012) interviews, veg*ns report using ‘face-saving’ impression management techniques to speak about their dietary identity without incurring alienation or confrontation. Greenebaum’s findings indicate that, although some veg*ns in early stages of their practice reported making attempts to describe to others the negative environmental and animal welfare impacts of meat-consumption, they report finding that these were largely unsuccessful, and most described instead using face-saving techniques. These included avoiding confrontation by downplaying their identity or changing the subject, waiting for specific times to broach the subject, leading by example in demonstrating the ease of their lifestyle, and emphasising health benefits rather than animal or environmental factors.

Thus, the literature has, to date, provided insight into what may result from a ‘successful’ or ‘unsuccessful’ identity flashpoint – that if the ‘do-gooder’ is ‘liked’ at this point the majority may be more likely to adopt similar behaviour, however if disliked or seen as a moral threat, they may be less likely than usual to do so. However, this literature has relied upon evaluations of hypothetical activists or do-gooders in vignette scenario studies, rather than studying actual social-interactional encounters. Similarly, qualitative work from a realist perspective has suggested strategies veg*ns may adopt during such flashpoints, but these rely on retrospective accounts provided within interview settings. As has been noted by researchers from within the discourse analytic tradition, there are many reasons to question the treatment of interview data as simply a direct pipeline into the cognitive or historical reality of interviewees (Potter and Hepburn 2005). Accounts provided in interviews are subject to revisions and distortions, and the re-telling of which itself may be influenced by the presence of the interviewer (typically a veg*n researcher to improve interview rapport).

Other research has used focus groups to explore factors influencing meat consumption and reduction, which offer the potential advantage that people may feel more able to speak freely within a larger group than one-on-one. For example, Graça, Calheiros, and Oliveira (2014) conducted focus groups about meat consumption with meat-eaters and found that, although participants described feeling a sense of personal duty towards the environment and animal welfare, they were reluctant to reduce their personal meat-eating. The authors suggested participants demonstrated avoidance of, and a disregard for, negative consequences of their behaviour, concluding their results provide support for moral disengagement theory (Bandura 2002). However, while such focus groups may be useful tools for soliciting views of meat eaters (or veg*ans) surrounding their own practices, the structure and setup of such focus groups is arguably less well suited for exploring identity flashpoints. The presence of a researcher directly asking about participants’ feelings towards animals, whether current meat practices may harm them and whether participants would change their habits to minimise harm (Graça, Calheiros, and Oliveira 2014) may make the research agenda too explicit, and lead participants to construct their accounts in response to this relatively contrived and directed situation. This structure inherent in most focus groups doesn’t allow for insights into the types of social interactions that may occur between those of different dietary identities (for example, during mealtime conversations). One study that has recently attempted to partly surmount this problem is Buttny and Kinefuchi (2020). In their discursive analysis of veg*n (intra-group) interactions, focus groups were conducted with no moderator present, with six veg*n participants talking through a list of questions provided on a sheet of paper, at their own pace. The results of this less artificial set-up provided support for previous work (Greenebaum 2012) in suggesting that veg*ns are aware of the possibility of being perceived to judge meat-eaters and take care to position themselves as being non-judgemental in the presence of meat-eaters, and has the benefit of speakers’ answers being uninfluenced by the presence of a moderator.

What is missing from the current literature, however, are actual analyses of observed conversations between meat-eaters and veg*ns as they unfold in situ, without the presence of a researcher or outside observer of any particular dietary identity. Such studies of talk-in-interaction also afford the opportunity to empirically observe how meat-eaters respond to (and potentially invoke) the conversational tactics deployed by veg*ans within these interactions. Ideal data in this scenario would, arguably, be purely naturally occurring: data that would occur or exist in the absence of the researcher (or indeed the research project itself) and with no researcher intervention (Potter 2002), such as audio recordings of talk occurring in a cafeteria within a work institution. However, there are both practical and ethical barriers to this suggestion. Practically, it would be difficult to capture veg*n-meat-eater conversation within a loud environment, and not being able to predict when identity flashpoints would occur would likely mean recordings would need to be obtained over weeks or months, leading to datasets of a size that would present practical challenges. Ethically, obtaining such data would constitute a severe invasion of privacy, and it is unlikely anyone would agree to provide their lunchtime conversation data over several weeks, or indeed have genuine/natural conversations in a cafeteria that was full of cameras or voice recorders ‘for research purposes’.

Hence, in the absence of fully naturally occurring data being feasible, we sought a methodological compromise somewhere between the controlled and artificial situations of survey research and researcher-led interviews and focus groups on the one hand and naturally occurring cafeteria conversation on the other. Some research has examined online forum discussions of vegetarianism, which could arguably provide insight into identity flashpoints to this purpose. For example, Wilson, Weatherall, and Butler (2004) analysed the data from online forum discussions on the question ‘Why be vegetarian?’. Their analysis revealed that ‘health’ reasons were used strategically as rhetorical devices to support or reject vegetarianism, providing useful critique on previous cognitivist work that assumes attitudes are straightforward causes of behaviour. Using a similar methodology, Sneijder and te Molder (2005, 2009) analysed threads on the Dutch Association of Veganism website discussion forums to explore the ways in which practicing vegans constructed their practice in response to questions posed by ‘novice’ or prospective vegans regarding matters such as avoiding gastronomic boredom or vitamin deficiency. However, as noted by these authors, despite some of the potential similarities with face-to-face interaction, online forums differ from more conventional forms of talk-in-interaction in several respects. Firstly, people have a lot of time to compose their responses. Secondly, individuals may say things online that they wouldn’t in ‘real life’, and thirdly, particular kinds of people may be attracted to view and comment on online forum discussions of veg*nism; perhaps those seeking an argument, or (in the case of Snijder & te Molder’s work) those who were already actively (considering) adopting the diet themselves. Hence, online forum studies may have some limitations in their direct application to everyday talk between dietary outgroups.

Quantitative meat-eater surveys, qualitative veg*n interviews, meat-eater focus groups and online forum studies have all contributed to the present knowledge base with regards to the psychology of dietary practice and identity. However, to understand the ways in which both veg*ans and meat-eaters discursively manage dietary identity flashpoints, and the implications that these discursive practices might have for the potential for social change, we wished to capture and analyse social interaction between veg*ns and meat-eaters directly, to observe (and analyse) it ‘as it happened’. This formed the aim of the current research. To this end, the present study drew upon recent developments in focus group methodologies that utilise remotely moderated focus group (RMFG) discussion (Prosser et al. 2024). We performed a discursive analysis of remotely moderated lunchtime focus groups where each group involved a meat-eater majority and a veg*an minority. Our aim was to investigate the ways in which identity flashpoints within veg*n-meat-eater interaction were managed and negotiated to understand the strategies used by veg*ns to navigate the presentation of their identities, how these strategies were potentially invoked by meat-eaters, and meat-eater responses to the use, or lack thereof, of these strategies.

Methodology

Creation of the data corpus

Data were collected via 25 remotely moderated focus groups (RMFGs), the majority of which (16) contained five participants, with 4 groups involving six participants, 3 involving four participants and 2 groups involving three. Participants were 122 University of Bath students (80 females, 41 males) aged between 18 and 31 years (M = 20.47, SD = 2.44). These included 85 meat-eaters, 19 vegetarians and 18 vegans. Each group included either 1 veg*an (13 groups) or 2 veg*ans (12), with meat eaters always being in the majority in each group. Participants were recruited to participate in a focus group about ‘food provision on campus’ through convenience sampling, which included recruitment posters on university campus public noticeboards, social network posts, and presentations at student advocacy and recreational clubs. Recruitment strategies did not mention the specific recruitment of veg*ns to prevent demand characteristics, but otherwise provided sufficient information on the discussion topic of food so that participants could make an informed choice on whether to participate.

All participants completed an online pre-screen questionnaire that asked them to indicate their dietary identities, allergies and lunch meal preferences before participating. Participants were selected based on their declarative identity and received a meal during the focus group consistent with that identity (e.g., falafel wraps for vegans, egg and mayonnaise sandwiches for vegetarians, chicken and bacon sandwiches for meat-eaters).

On entering the research space, participants received study information and provided written informed consent in an adjoining waiting room, before entering the discussion room, seating themselves around a table, and receiving their lunches. The focus group was guided by questions on a PowerPoint presentation displayed on a large wall-mounted monitor in front of participants (see Figure 1). Focus groups lasted for 45.04 minutes on average (range = 30–67).

Figure 1. A photographic image of the remotely moderated focus group setup taken from the audio-visual recording of one of the discussions.

Participants were told these questions advanced automatically; however, they were actually advanced manually, according to the flow of the conversation, by the researcher monitoring an audio-visual feed of the interaction from a room next door. See Prosser et al. (2024) for a detailed explication of the RMFG method, including its implications for turn-taking and division of conversational responsibility.

To build group rapport, focus group discussion questions began broadly, by asking about participant dietary requirements and how easy they found it to find food options on campus, before slowly leading into more targeted questions designed to elicit identity flashpoints, e.g. ‘Do you think people across Britain should drastically reduce their meat consumption?’ and ‘For those of you who eat meat, have you ever considered removing meat from your diet?’ (see Supplementary Materials A for full focus group schedule).

After completing the focus group, participants were debriefed, given the opportunity to ask further questions, re-signed the consent form and then received £5 or course credit for their participation. All sessions (18.76 hours total), were audio and video-recorded and subsequently transcribed (using a simplified version of the Jeffersonian notation system (Atkinson and Heritage 1984) – see Supplementary Materials B). The study was approved by the University of Bath Psychology Research Ethics Committee (reference number: 19–024).

Analytic approach

We drew upon the theoretical and methodological underpinnings of a discursive psychology approach (Edwards and Potter 1992; Wiggins 2016), which interrogates ‘psychological’ phenomena (e.g., attributions, memory, evaluations of social objects) not as cognitive states but rather as matters that are constructed and negotiated within social interaction. Here, we were not focussed on examining the cognitive effect of moralised vs non-moralised accounts of veg*an’s practice on meat eaters, or the quantitative levels of ‘liking’ subsequently reported towards veg*ans who spoke in particular ways. Rather, we were interested in exploring how different kinds of veg*an accounts of their practice were taken up and responded to by meat-eaters within actual social interactions between members of these two identity groups, as well as the consequences for the subsequent unfolding interaction and for the discursive problematisation (or not) of a meat-eater identity in this context. To this end, we assembled a body of instances within the talk where veg*an participants either spontaneously (or by invitation) presented accounts of their own dietary practice, their reasons for abstaining from meat/dairy, the historical trajectory of their adopting of said diet, and their levels of intention to maintain their diet. We then examined the rhetorical features of these accounts and the ways in which the responses to different types of accounts were (collectively) taken up and responded to by meat-eater participants within subsequent turns, with a view to examining the social-interactional implications of different accounts on the trajectory of the unfolding conversation.

Analysis and discussion

Our results present an analysis of the strategies used by vegetarians and vegans to navigate identity flashpoints with a dietary outgroup, with an emphasis on how meat-eaters respond to these strategies and the nature of the interaction that unfurls as a result. In the extracts used, the notation M, V and Ve denotes meat-eater, vegetarian and vegan participants respectively, with numbers added if there is more than one per group. Two key types of flashpoints were identified within the analysis: a) moments whereby vegetarian and vegan participants were prompted to discuss their reasons for their dietary practice, and b) moments when they were asked whether they would ever consider reintroducing meat into their diets.

Why did you go vegetarian? The privileging of personal accounts

The question of why veg*ans stopped eating meat was not actually one of the questions directed to the participants via the screen, however, was nonetheless asked of veg*ans by a meat-eater in almost every group where that vegetarian or vegan did not offer ‘their reasons’ for their dietary practice themselves first. Although there was some variation between participants, what was common in the vegetarian and vegan replies was a tendency to privilege accounts couched in relation to health, cost, or other personal reasons, over those framed in terms of animal welfare or environmental reasons. Thus, these accounts were constructed around notions of ‘self-benefit’ rather than ‘other-benefit’, often focussing on circumstantial reasons for veg*nism, and in doing so downplay any other motivations. Two examples can be seen in Extracts 1 and 2.

Extract 1.

Extract 2.

In these extracts, in each case the meat-eater asks about the veg*n’s own reason for being veg*n, for making that decision. However, the responses given distance the self from that decision and place the responsibility onto someone else: their mothers. The vegan in extract 2 (after switching pronoun from ‘I’ to ‘my mum’ – line 17) does make hints towards her own agency by describing personally disliking meat but is silent on any other underlying motivation behind this decision. She is also notably silent on her reasons for becoming a vegan – if going vegetarian is attributed to her personal distaste of meat, going vegan is accompanied by no mention of not liking dairy, and the agency here is attributed to her boyfriend. Any hint of reasons of health, the environment or animal welfare is conspicuously absent from this account. In extract 1, reasons for vegetarianism (other than the influence of figures in one’s own life) are also absent – there is a hint towards personal reasons ‘that vaguely aligns with what I think’, but exactly what she thinks is left unstated. In essence, there was a discursive attribution of agency away from the self, with other people in one’s life often constructed as the active agents in the decision to become a veg*n.

The second means by which the privileging of personally specific reasons over moralistic reasons was evident in these flashpoint interactions was through a tendency for veg*ans to package their accounts in a form where personal reasons preceded moral reasons, before then completing the account with a return to personal reasons, as can be seen in the extracts below.

Extract 3.

Extract 4.

By wedging environmental and animal welfare reasons between personal benefit reasons, the vegetarians in both of these extracts above package any potentially morally implicative motives within less interactionally ‘difficult’ ones, completing a shielding manoeuvre, or a form of ‘defensive sandwich’. Specifically, the first item on a list of reasons may be taken to be the most important, the main reason – the reason the speaker wants you to know first (Wiggins 2016). As to the content of this first reason, a decision based on personal factors of health or money may be more difficult to argue with and less contentious than an environmental or animal welfare reason because it does not imply that everyone should be morally obliged to follow the same behaviour, or that the vegetarian might judge them for failing to do so. The positioning of a personal reason first here therefore establishes the vegetarian (in interaction) to be primarily concerned with their own lifestyle, and therefore not likely to judge others who do not share their dietary practice. The potential for a moral reason to be treated as a dispreferred response (Rendle-Short 2015) is clearly demonstrated in Extract 4, where we see that while the first, personal reason (‘cheapness’) was met with interest and agreement by the group (lines 18–22), the second, environmental reason gains an ‘mhm’ from only one meat-eater, which is followed by silence; the five-second pause suggests hesitation on both sides to further discuss this potentially controversial issue, and is only broken by the vegetarian returning his account to a personal reason. The fact that in both extracts the vegetarians end their accounts on personal reasons suggests there is something problematic in completing their account of their motivations with moral reasons, which is avoided by a return to a socially ‘safer’ position. Personal reasons are thus placed to appear the most important, with the moral reason constructed as almost ‘incidental’.

This finding provides further support to previous (experimental) work showing that meat-eaters will dislike a vegetarian to the extent they believe they are being morally judged by them (Minson and Monin 2012) and qualititave work in which veg*ans report changing their stated reasons from moral to personal to fit their audience (Greenebaum 2012). The present analysis shows that, in a group of meat-eaters, vegetarians and vegans give greater interactional space to personal reasons compared to moral reasons. Moreover, we demonstrate the ways that meat eaters respond, in situ, that help to position more moralised accounts as interactionally problematic. Such an interactional move on behalf of veg*ans can work to defend against appearing judgemental, and in doing so may ensure continued wider group membership. Although, of course, it also discursively positions meat eaters within the interaction as, by extension, not being morally accountable for their choices to consume meat.

The interactional significance of the privileging of individualised accounts of meat abstinence is brought into particularly sharp relief by examining two deviant cases identified in the data. In the following extract, a participant who is actually a meat-eater attempts (initially) to construct his decision to remove beef and dairy (specifically) from his diet (while continuing to eat other animal products) as being environmentally motivated.

Extract 5.

Extract 5 empirically demonstrates the social-interactional difficulties of discussing environmental (moral) reasons as primary justifications for not eating meat or dairy. When M1 invokes this reason for no longer eating beef or dairy, there are a number of discursive features indicative of interactional trouble. Firstly, he begins his account by immediately changing footing to the passive voice (Goffman 1981) and thereby laying the responsibility for his decision with his housemate: ‘I was persuaded – by my housemate’ rather than ‘I decided to stop’ distances him from his account, creating lessened ownership of his words and placing all moral conviction (and potential ‘judgementalness’) with his housemate. The repeated ‘um’s during this part of the interaction (especially as they are not a usual part of M1’s speech in the rest of the interaction) further support the difficulty of making this point (Pomerantz 1984).

Secondly, despite minimisation of both problem and action in further elaboration (‘it’s not particularly sustainable’, ‘better do my bit’ – rather than ‘it is unsustainable so I stopped’) the group remain silent during M1’s account. Conversation is usually punctuated with agreement tokens, in the form of ‘mm’s, ‘mhm’s ‘yeah’s and nods (Heritage and Raymond 2005), however during M1’s environmental explanation the group show none of these, which can convey social disapproval or discomfort (Pomerantz 1984). M1’s moral account, unlike the vegetarians in extracts 1–4, could suggest judgment of meat/dairy-eaters who do not follow his example. Therefore, it is notable that, following this, M1 then rapidly switches to a personalised account of how it was particularly easy for him to reduce his consumption because of his particular taste – to which the group begins demonstrating agreement tokens once more (lines 31–39). This switch to the personal can be read as a face-saving attempt: if giving up beef and dairy had been extremely difficult for M1 but he had accomplished it through the strength of his moral conviction, this then opens up potential for his account to imply a judgment of others without similar willpower. By portraying his change as ‘not really much of a sacrifice’ for him personally because he didn’t like/eat many of those foods anyway, a space is created in which anyone without these mitigating circumstances could be excused for not following his example. The group agreement tokens after his professed dislike of dairy at line 31 onwards indicate his attempt is successful, however his continued justification of his choice with reference to personal reasons indicate the extent to which personal reasons are preferred to moral reasons. If moral environmental accounts were entirely socially acceptable justifications, there would be nothing wrong with leaving his explanation at the environment. Instead, he continues to run through an entire list of foods he dislikes to further minimise the moral strength of his action and demonstrate that he ‘only’ had to really give up very few. He is challenged by M3, who reminds him ‘beef as well’ (line 42), to which M1 responds that beef was ‘only last November’, and that he didn’t eat much beef anyway (further minimisation); at which point M3 provides a ‘rescue’, handing him another personal reason in the form of health (line 51), which then serves to re-open group conversation. This extract provides a vivid demonstration of the social difficulties inherent in discussing morally motivated reasons for meat reduction (let alone elimination).

This social difficulty becomes even more apparent when a vegetarian outright professes her belief against killing animals for meat:

Extract 6.

Here, as in extracts 3 and 4, the vegetarian attempts to preface her animal ethics motivations with claims of personal preference (‘I enjoy being vegetarian, I like a lot of vegetarian food’), and also disclaims the possibility she might be judging others (‘I just don’t personally, I know it’s different for everyone but I personally’). However, despite such efforts, the starkness of ‘don’t like killing animals for the sake of meat’ serves to undo these previous attempts at interactional smoothing. The active ‘killing’ draws attention to the violence of the act, whilst the ‘sake of meat’ implies this is not a good enough reason for ‘killing animals’. Therefore, despite all of her attempts to avoid the appearance of judgment (emphasising her personal preferences; stressing how it’s ‘different for everyone’) her concluding reason for not returning to meat-eating carries a distinct possibility of judging those who practice it.

The meat eater response to this is a telling example of the interactional trouble veg*ns face when talking about animal ethics as a motivation for veg*nism. What follows her ‘killing animals’ declaration can best be described as a dedicated search for hypocrisy – if she has indeed been ‘totally committed’ these last three years, she must account for her behaviour before then (if she was fine with it then, what happened?). Having established she had long desired to give up meat, a new line of investigation begins – a person who eats those ‘fake, wannabe meats’ must be on some level a hypocrite. When she establishes that she does love Quorn, she is swiftly rebutted with a new charge of hypocrisy (‘No but like, why do you want to be flirting with meat if you don’t want to be with it’). Crucially, it is hard to imagine this same line of questioning would have followed had she simply stuck to her first response of liking vegetarian food – as indeed the other veg*ns who stuck to personal reasons were not questioned in this manner. Instead, it reads most convincingly as a meat-eater attempt to ‘catch out’ a veg*n who implies moral judgment of meat-eaters’ behaviour.

In sum, health, cost, personal taste, life circumstances, the environment and animal ethics are all discursive resources that are culturally available for accounting for dietary practice. However, personal reasons (health, cost, life circumstances or taste) seem to occupy the status of what conversation analysts would term the ‘preferred response’ (Bilmes 1988; Kitzinger and Frith 1999) in comparison to moral reasons (animal ethics and the environment). Where moral reasons are presented as sole reasons, the veg*an speaker appears to be required to perform laborious interactional work to defend their own position and save the face of others who have not made a similarly morally motivated choice.

Would you ever go back? The need for continued liking of meat

The second flashpoint revolved around the question of whether veg*ns would ever reintroduce meat to their diet, a question that was asked within the focus group slides presented on the screen by the researchers, or, in some cases, pre-emptively by a meat-eater participant. This question invites veg*ns to share their current thoughts or feelings about meat as part of why they would or would not consider its reintroduction to their diet.

A response commonly observed here was for veg*ans to say no while demonstrating a continued liking of, or temptation towards, meat (or cheese), as we see below in Extract 7.

Extract 7.

The vegetarian here begins by confirming that he used to be a meat-eater, establishing prior ingroup membership, and still claiming a casual membership by disclosing occasional meat-eating. This prefaces his overall answer that he would not go back; a conclusion framed in ways that steer away from absolute certainty: ‘I think I would struggle to’ and ‘I don’t know what would persuade me’, all ways of softening a simpler ‘no’. However, despite this, there is no group acknowledgement to the conclusion of his account (at line 439). Receiving no acknowledgement (for three entire seconds), he proceeds to reassure the group that he does still like eating meat and is then acknowledged and agreed with (lines 441–443). This assurance of psychological warmth and gastronomic inclination towards meat works to, by extension, be received as warmth (or at least non-judgment) towards the meat-consumption of meat-eaters.

Other veg*n participants, by contrast, did not have occasional meat-eating to use as proof of non-judgment, and instead spoke about still missing or being occasionally tempted by a meat or dairy product:

Extract 8.

The vegan in extract 8 presents her desire for chicken nuggets to mass group approval, as demonstrated by repeated meat-eater laughter. This account prefaces her response that she would not reintroduce meat – again softened (‘I’ve not really felt like’ as compared to a more straightforward ‘I wouldn’t’) and again without group acknowledgement. This silence in response to the expressed unwillingness to reintroduce meat can be understood as a ‘dispreferred response’, one that does not meet with social approval (Bilmes 1988; Pomerantz 1984). Being unacknowledged (with the group all averting their gaze), the vegan here goes on (in a similar way to Extract 7) to talk about being tempted by cheese, a ‘preferred response’ that meets with laughter and social approval (reminiscent of Docherty and Jasper’s (2023) recent account of the ‘Cheese Paradox’ in vegetarian talk about veganism). The laughter in Extract 8 can be read as social approval, and perhaps a release of group tension. Indeed, there is something about a vegan saying they occasionally miss nuggets that attracts more laughter within the groups than when a meat-eater says they couldn’t give up meat because of bacon. Both are speaking of personal taste; however, the vegan’s comment has the power to release any tension of meat-eaters in the presence of someone who could potentially judge them, but then appears to be ‘human’ after all. The vegan’s occasional wish for nuggets and cheese would appear to excuse the meat-eaters consistent consumption, giving them common ground and assuring the vegan’s non-judgment. This has the effect of constructing the vegan as ‘not too moral/morally righteous’.

Again, the veracity of this commonly observed pattern in our data is best demonstrated by observing the social interactional result of a deviant case, as we see below.

Extract 9.

Similar to other vegetarians and vegans, the vegan’s use of the preferred response (the temptation of a meat product) gains laughter in response. However, her ultimate distaste of meat does not gain a favourable group reaction. When describing that she would probably ‘throw up’ if she ate the same pepperoni she was tempted by, whilst the meat-eater (M2) who laughed at her pepperoni temptation continues to engage with her, M3 (who has previously described himself as trying to increase his chicken breast consumption) shows limited approval and appears to try to change the direction of the conversation: ‘this is much more time filling when they had several vegetarians’ – implying the current vegan’s account must now be finished. However, the vegan does not follow this potential hint to stop, and proceeds to tell a narrative of feeling sick and stressed when she thought she had inadvertently eaten a piece of meat. Like normal conversation, narrative stories are usually filled with reactions from the listener, from agreement to surprise or even simply nodding (Potter 1996). However, as soon as the vegan begins to talk about her shock and her disgust reaction towards meat, the group’s silence, lack of joining in with her own laughter (at lines 602, 607 and 609) or any non-verbal acknowledgement of her story show that demonstrating visceral dislike of meat is not an account that the group is willing to align with. The vegan faces social ostracism for her disgust account, and a full eight seconds of silence is left (line 611) until the researcher changes the slide show to the next question.

This example can similarly be explained in relation to the interactional construction of moral judgment – if a vegetarian or vegan presents themselves as disgusted by meat, this implies that same reaction might occur to those who consume it. Conversely, where meat-eaters do not respond favourably to disgust accounts of meat, this suggests they may feel blamed for meat-eating. The particular use of the ‘being sick’ imagery may be intended to construct the situation as simply a physical reaction, however it can perhaps unavoidably also be read by the meat-eaters as yet stronger moral judgment: if eating a burger is ‘sick-making’, what is their dietary practice to her?

Conclusions

Our findings here suggest a dilemma for those who might wish to see shifts in societal dietary patterns away from meat-based diets. Firstly, because they imply judgment, moralised accounts appear to provide interactionally problematic vehicles for veg*ans to try to engage in any form of persuasive attempts regarding others’ dietary practices. While the potential social costs of veg*ans presenting as a moral ‘do-gooder’ have been previously highlighted via more quantitative/experimental approaches (Minson and Monin 2012), what we provide here is a detailed analysis of the ways in which moralised accounts are demonstrably dispreferred by meat-eaters (via silence and a failure to align with these accounts through agreement tokens), in situ, within actual social interactions, and the interactional pivots that veg*ans perform within the unfolding interaction in response to this. Our analysis highlights just what interactionally difficult identities veganism and vegetarianism are to perform within a social-interactional context in which others’ identities as meat-eaters are potentially problematised. Furthermore, veg*ns face a context in which their social acceptance and approval is contingent upon appearing to continue to like/crave meat and their abstinence being mostly motivated by their own personal lifestyle, risking rejection or ostracism if they fail to do so. Thus, a difficult position is created from which a veg*an speaker might attempt to influence others or advocate behavioural change. The very nature of (and reason for) their identity as a veg*an ends up being interactionally produced in ways that construct meat-eating as both desirable and morally unproblematic.

In addition to its theoretical and empirical contribution, this study also provides an interesting further demonstration of the potential utility of the recently developed remotely moderated focus group methodology. Prosser et al.’s (2024) explication of the method presented evidence for its efficacy in producing useful and rich qualitative data in the context of intra-group discussions (where the discussion group members share a relevant identity). Here, we extend the application of this methodology to create inter-group discussions between participants from different identity groups. We demonstrate its ability to generate a type of social-interactional data that is fundamentally different in nature to that which would be afforded by a traditional moderated focus group where the researcher drives and manages the discussion, arguably approximating more closely (albeit, by definition, not achieving) a more ‘naturally occurring’ form of conversational data (Potter 2002). This methodological ‘model system’ for generating cross-identity talk between those who hold quite oppositional views on an issue and/or engage in divergent sets of behavioural practices may well hold great utility in the examination of other such socially consequential contexts. Although work within the tradition of Discursive Psychology (Edwards and Potter 1992) has typically privileged the analysis of truly ‘naturally occurring’ data, what we hope to have demonstrated here is the ways in which RMFGs can generate data of a nature that is also highly amenable to these forms of analysis. As such, we suggest that they represent an interesting and potentially fruitful method by which to discursively examine other forms of ‘difficult’ intergroup interactions (i.e., identity flashpoints). Examples might include (amongst numerous others), cross-racial/gender identity discussions around racism/sexism or interactions between climate activists and climate sceptics. Although, researchers would have to carefully consider the ethical implications of encouraging these interactions and formulate ways to protect participants should discussions get too heated or confrontational.

The current research is not without potential limitations, however. Some could question whether a sample of exclusively university students may mean that findings may not generalise to the general population. However, given previous findings indicate vegetarians to generally be highly educated (Cramer et al. 2017; Pfeiler and Egloff 2018), this may mean that the university population are generally more sympathetic to the practice than would be a more diverse community sample. As such, this arguably makes it all the more striking that our vegetarian and vegan participants in this context demonstrably adhered to the social-interactional requirements for identity presentation that we identifiy in our data here, even within this potentially more understanding setting. Furthermore, university represents a ‘moment of change’ in life – with many living alone and cooking for themselves for the first time (Verplanken and Roy 2016). Thus, university may be a point where eating habits are particularly open to change, making it a particularly interesting stage of life with which to observe the kinds of interactions that our dataset here affords.

Due to the practical challenges involved in recruiting a large number of participants who were vegan or vegetarian (given they represent a small proportion of the population and we wished to avoid marketing the study as ‘about veg*anism’), our discussion groups involved only one or two veg*ans, and they were always clearly in the minority. This raises the interesting question though of how interactions might unfold in a majority veg*an context where a meat eater is very much in the minority themselves. Whether veg*ans would still demonstrate adherence to the interactional patterns that we have mapped in our data here, and how minority meat-eaters would respond to accounts of veg*an practice in this interactional context represent interesting (albeit practically challenging) avenues for future research.

To conclude, a discursive analysis of identity flashpoints within remotely moderated lunchtime discussions involving a veg*an minority revealed two key patterns within these interactions in terms of identity management practices: the privileging of personal over animal ethics or environmental reasons for meat-free lifestyles, and the display of continued liking of meat. These findings mark a difficult social context from which vegetarians or vegans might attempt to directly advocate behaviour change towards meat reduction or removal and open up new methodological avenues through which to further investigate this societal conundrum.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/14780887.2024.2359423

Additional information

Notes on contributors

Marike Fordonnell

Marike Fordonnell is a clinical psychologist in training in the Department of Psychology, at the University of Bath. Her research interests include clinical health psychology and qualitative methods, and her research explores trauma-informed practice in mental health services, factors which promote engagement in mindfulness interventions, and the impact of communicating epigenetics on health behaviour motivation, stigma and policy support.

Annayah M.B. Prosser

Annayah M.B. Prosser is a lecturer (Assistant Professor) in the University of Bath School of Management. Her primary research interest concerns how individuals and groups respond to societal crises, such as the climate and ecological emergency. She explores how morality, identity and ethics might influence group interactions and societal transformations. She is also interested in methodological innovations and open research, particularly with regards to advances in qualitative methods.

Tim Kurz

Tim Kurz is a Senior Lecturer in the School of Psychological Science at the University of Western Australia. His research focuses on the social psychological dimensions of attempts to foster more environmentally sustainable societies, with a particular focus on the role of morality and social identity and the ways in which both play out within social interaction.