Abstract
Social theory usually theorises the “social” as people encountering each other and interacting with each other. Some of these interactions are pleasurable and beneficial, some less so. Social theory has developed a great range of concepts for describing social interactions, especially the interactionist school associated with Mead, Goffman, and Blumer. What has not been conceptualised is all the coordinative work that goes into not having interactions at all. Interactionist theories, in particular, struggle to capture these kinds of living coordinations, even though, once seen properly, they probably are vastly more important than actual social interactions. The ethnography of hair pulling makes this unusually visible. People who pull out their hair conceal the behaviour and its bodily traces, abandon disclosures that have gone badly, avoid settings in which their hair will be open to comment, and live in families where some questions are never asked, while parents learn, often painfully, that care may require not asking, not watching and not correcting. Reanalysing through Goffman’s theory of stigma the ethnographic material we first published in 2018 (Bradley & Ecks 2018), we argue that noninteraction is a positive, multimediated achievement, which we call organised noninteraction. His theories of stigma, face and backstage life repeatedly approach the phenomenon but continue to conceptualise it through the interaction anticipated or avoided. Living Value Theory’s distinction between nonrecursive, selfrecursive and interrecursive counterparts, together with its concepts of recursive availability, opening and closure, shows how unwanted interrecursivity is prevented, suspended and selectively reopened, and why a theory of living coordination requires an account of its selective suspension.
I. Introduction: The Social Work of Making Interaction Not Happen
In her twenties, Heather read a magazine article about trichotillomania and realised that the thing she did to her hair had a medical name. She telephoned her father, a doctor, and said that she was feeling brave: “you know I do this thing?” He said yes. She went on: “well, it’s got a name, can you do some research for me, can you ask some people?” A few days later she rang back to ask whether he had looked into it, and he said: “no, no, no, no, your mum says not to indulge you.” Her parents had never spoken with her about her hair, not even in the period at primary school when the crown of her head had gone bare. Heather’s conclusion was brief: “So I don’t talk about it with them. And I don’t [now] and I don’t think I will [again]. There’s no point”.
The conventional sociological question about this episode is what happened in a failed interaction. An interactionist account can say a great deal about it: a request for recognition met with a refusal, a medical authority declining to lend his authority, a mother’s phrase relayed through a father, a daughter’s face threatened at the moment she had made herself most open. The question pursued here is what happens afterwards, when an entire domain of family life is reorganised so that this interaction will not happen again. Heather concludes not that the conversation should be conducted better but that it should not be conducted at all, and the conclusion is acted on for years.
Our 2018 article showed that hair pulling is entangled with family life in complex ways, that parental responses can deepen suffering, and that support networks can transform how the condition is lived. We read these findings through stigma, kinship and relatedness, and that reading stands. Here we reanalyse the same material through Goffman’s theory of stigma and the concepts around it, and read it again in the light of Living Value Theory (LVT). The reanalysis shows a further phenomenon that runs through nearly every case. Many participants try principally to prevent interaction around their hair pulling altogether, and only secondarily to improve it. They conceal the pulling and its traces, withhold it from parents, give up on topics after a failed attempt, and keep away from activities in which their hair would be exposed. The fieldwork, carried out by one of us (Bradley) over sixteen months in 2015 and 2016, drew on interviews and participant observation with people living with body-focused repetitive behaviours (BFRBs), parents, partners and professionals in the United Kingdom and the United States (Bradley and Ecks 2018). Because we are reanalysing our own article, and every participant quotation below comes from it, we give no page references to it. The fieldwork was not designed to study noninteraction, so the argument below is a reanalysis, and wherever a claim rests on implication rather than on what we reported, we say so.
The central distinction is between an event and an achievement. Interaction is an event. Noninteraction can be an achievement, and often a sustained, skilled and costly one. Organised noninteraction can be defined as the positive coordination of bodies, materials, spaces, symbols and relationships so that particular recursive encounters are prevented, deferred, limited or terminated. The definition excludes isolation, the absence of being-with and the absence of coordination. A person who conceals hair pulling from a mother with whom they share a house, a table and a history remains thoroughly related to her. What has been arranged is a selective closure of interrecursive availability, so that one domain of the person cannot be taken up and answered by this particular other.
Interactionist sociology is exceptionally good at analysing encounters once they occur. It is much weaker at analysing the work through which people arrange their lives so that particular encounters never occur in the first place. The remainder of the article develops that claim through the trichotillomania material, beginning with Goffman, whose theory of stigma remains indispensable and whose vocabulary nonetheless exerts a pull toward the encounter.
II. From Stigma Management to the Organisation of Noninteraction
Hair pulling becomes consequential in large part through anticipated judgements of abnormality, self-control, femininity and bodily appearance, and Goffman’s theory of stigma (1963) is the natural starting point for understanding why. In our 2018 article we extended the theory by showing that stigma is managed within domestic life as much as in public, and that for these participants “managing the stigma of hair pulling emphatically starts at home”. Goffman’s distinction between the discredited, whose difference is already known or evident, and the discreditable, whose difference can still be kept from others (Goffman 1963: 4), also fits the material closely, since most participants lived as discreditable persons to their closest relatives.
Goffman’s vocabulary nonetheless carries a gravitational pull toward the encounter. Passing, covering, information control, face-work and impression management all ask how a person manages what another person perceives. Backstage, the region in which the front put on for an audience can be dropped, is defined relative to a given performance and the audience for it (Goffman 1959). Goffman did not ignore avoidance. In his essay on face-work he describes avoidance processes through which people steer clear of situations in which face would be threatened (Goffman 1967: 15–18), and in the study of impression management he discusses the tact with which audiences overlook what would discredit a performer (Goffman 1959: 229–237). In each case the avoidance is analysed in terms of what the parties do within or toward an interaction that is under way or expected. This leaves a different question unasked: what if the person is organising the world so that there will be no audience?
Three distinctions open that question. The first is between interaction avoidance and organised noninteraction. Avoidance suggests a psychological reaction, a flinching from something feared. Organised noninteraction is a mesocosmic arrangement, and it includes bodily habits, material instruments, spatial and temporal routines, secrecy, selective disclosure and the use of institutional boundaries. The second is between anticipated and actual interrecursivity. A person can organise hours of daily conduct around an encounter that never takes place, so that the absent encounter is causally important without being empirically present. Heather’s silence with her parents is shaped by a conversation that has not happened for years and, on her own prediction, will not happen again. The third is between being-with and interrecursive availability. One can live intimately with another person without making every aspect of oneself available to be taken up and answered by them. The interactionist temptation is to read intimacy as increasing recursive openness, and the ethnography shows that intimacy often requires zones within which the other does not enter.
LVT’s three types of recursive counterparts give these distinctions an exact form. They describe relations in which a recursive being engages something, and they do not classify persons or situations. A nonrecursive counterpart is one that does not orient back, whatever else it may do. A selfrecursive counterpart obtains when a person engages some aspect of their own ongoing state and alters in relation to it. An interrecursive counterpart is another recursive being that responds to one’s responses. Hair pulling involves all three. There is selfrecursive coordination with urge, sensation, tension and relief. There is nonrecursive coordination with hair, fingers, mirrors, wigs, rooms and the things that might hide or reveal. And there is interrecursive coordination with parents, partners, peers and clinicians. The tradition inherited by sociology treats the third as the only properly social domain, and the first two as private habit or physical circumstance. Our argument is that what people do in relation to the first two is largely organised in view of the third.
III. Hair Pulling Before the Other: Embodiment and Selfrecursive Coordination
An account of noninteraction has to begin before stigma, with what is being concealed. Our 2018 article contains an observation that its own framework did not fully exploit. Participants, we reported, “unanimously rejected the description of ‘self-harm’” and described their behaviour “as pleasurable and as making them feel relaxed and happy”; they rarely called it painful, and the act itself “is more often a source of pleasure” than the results it leaves. Several would not choose a “magic bullet” to remove it: “I wouldn’t want to change it”. Elsewhere we speak of the “secret pleasures” the behaviour affords.
Consider what hair pulling is before anybody sees it. From an interactionist perspective its sociological significance begins when hair loss becomes visible and available for interpretation. From an LVT perspective the act already has a complex coordinative structure. It involves fingers, hair, skin, texture, tension, visual attention, bodily position, time and repetition, and the participants’ own descriptions speak of felt qualities, relaxation and happiness. Much of this appears to belong to the L1 and L2 registers, the zone of coordination that proceeds without explicit articulation and the zone of felt misalignment and its easing. We also reported a sudden urge, and an inability to stop despite the will to stop. This is a selfrecursive configuration in which a person engages their own ongoing state while articulated intention, at L3 or L4, does not govern the outcome. None of this requires another person, and an account that locates the sociological interest of hair pulling wholly in the responses of others misses the part of it that is lived in coordination with one’s own body.
The contrast that matters is between pulling as lived embodied coordination and pulling as recursively exposed conduct. Once another person becomes aware of it, a different process begins. The same act can become evidence of illness, of lack of self-control, of bodily damage, of failed femininity or of parental failure. Our cases supply each of these readings. Camila asked herself “what did we do wrong?”; Mark came to believe “I did this to my daughter”; Mandy and Camila both worried about their daughters’ appearance in “critical years as a girl”; Dhaya’s mother treated hair pulling as equivalent to the cutting of hair forbidden to a Sikh and punished it severely. The shift from the first mode to the second transfers the act from selfrecursive and nonrecursive coordination into interrecursive exposure, so that an act whose felt meaning was relaxation becomes available for interpretation by people who cannot feel it.
The discussion of diagnostic naming belongs here. For people living with BFRBs, learning that their behaviour “has a name” and is shared by many others is “a major life event”, and the exclamations recorded in the United Kingdom are consistent: “I couldn’t believe it was a thing!”, “It had a name!”. Naming has two effects that are easily run together. It opens interrecursivity with a new community, as the later sections show. It can also reduce exposure to hostile interpretation, by supplying a deposited symbolic category, so that a person no longer has to explain from the beginning why they do something that lay people tend to misrecognise “as a personal quirk rather than as a health problem”. The category can therefore both open and close. Heather’s case shows the risk on the other side, since the name she brought to a father with medical authority produced a closure rather than an opening.
IV. Concealment Is Not Merely Stigma Management
The empirical centre of the article is the finding that most participants “tried to keep their hair pulling a secret from everybody, not only those encountered in public spheres but even close family members,” and that some concealed it so systematically “that not a single other family member was aware of it”. Our 2018 interpretation ran from stigma to concealment. The reading proposed here is more specific: anticipated harmful interrecursivity leads to the organisation of noninteraction. The difference matters because the second formulation directs attention to the mechanisms through which concealment is accomplished, and these are mediationally varied.
The first mechanism is the prevention of recursive opening. If the condition remains invisible, no question is asked, no explanation is required, no diagnosis can be contested, no promise to stop has to be made, no disappointed parent has to be reassured, and nobody says “why can’t you just stop?”, the phrase we identified as the characteristic derisive comment of parental failure. Concealment protects a settled zone from becoming askable.
The second is material. We mentioned wigs, shaved heads, and the absence of eyebrows and eyelashes, and noted that visible hair loss carries particular weight for girls and women, since “when they see a girl with her head shaved … it’s a whole gender thing”. Mark describes how he has stopped looking at his daughter and seeing only “the bald spot, or … the wig, or … the no eyebrows or no eyelashes”. Such materials are nonrecursive counterparts through which exposure is regulated, and their significance extends beyond impression management. A wig can prevent the encounter in which an impression would have to be managed, because it removes the occasion for the question. Hairstyle, clothing and cosmetics are plausible further instruments, but we did not describe them, and they are mentioned here only as an implication.
The third is temporal. Disclosure is postponed, sometimes for years: Heather had lived with her parents’ silence throughout childhood before she tried once, in her twenties, and several participants told Bradley things they said they had never told anyone, including their closest relatives. An encounter can be deferred indefinitely, and the deferral is a form of work.
The fourth is symbolic. Before participants learn the word “trichotillomania” the activity can remain below explicit family articulation. Heather’s opening sentence to her father, “you know I do this thing?”, shows how little language was available. Once a name exists, disclosure becomes possible without becoming obligatory, and the name can be used to open a topic or kept in reserve.
The cases of Heather, Olly and Siobhan are three variants of the pattern. Heather attempts a disclosure and recloses. Olly’s father responded to the news that he was still pulling out his hair with “What? What’re you doing that for? That’s stupid,” while his mother, present, neither intervened nor offered support; asked whether he ever talked with his parents about it, Olly said “no”. Siobhan reports that her family do not know what she means and “never ask me about it”; at her sister’s wedding, while being prepared with the other bridesmaids, she had to disclose the state of her hair to her sister’s friends, and her mother said: “please don’t make a scene!” Siobhan now avoids conversations about her hair with family members. We stated that the desire to conceal is “based on well-founded fears about revealing it”, and the point can be taken literally. Concealment in these cases is evidence-based anticipatory coordination. Participants have learned what happens when the interaction occurs, and they have arranged their conduct accordingly. That the arrangement also costs them, in loneliness and in the lost possibility of the care they wanted, is the difficulty the next section takes up.
V. Silence: Failure of Care or Successful Boundary?
Our 2018 article recorded an ambiguity that is theoretically productive. Participants frequently experienced their relatives’ silence as inaction and lack of care, and as a source of “deep isolation and embarrassment.” Yet we also noted that keeping the hair pulling secret is the participants’ own way of “reducing the stigma inflicted, or feared to be inflicted, by family members,” and that silence and secrecy are ordinary features of domestic life, which can even be more harmonious when protected by secrecy. The two observations seem to contradict each other until it is seen that interaction has no single value. Sometimes asking is care and sometimes not asking is care. Sometimes silence protects and sometimes it wounds. The relevant question is therefore not whether there is interaction or none, but which recursive openings sustain living coordination and which damage it.
Four kinds of silence can be distinguished. Protective silence is the withholding of disclosure because disclosure would expose a person to predictable shame, which is the pattern of most participants. Dismissive silence is a refusal of engagement in reply to a request for recognition, and Heather’s father’s refusal belongs here. Respectful nonintrusion is the situation of a relative who knows but does not continually monitor, question or correct. This form appears only later in our cases, in the learning of parents. Mutual closure is a tacit agreement from both sides that a topic will remain unavailable, which can stabilise family life while leaving suffering unrecognised. Siobhan’s household, in which questions are not asked and she has stopped raising the subject, resembles it, although her mother’s “please don’t make a scene!” shows that the closure was at least partly imposed.
Silence of any of these kinds is still read. Our reference to “silent memories” (Kidron 2009), through which parental silence acts as “a form of communication between parents and children, an expression of how hair pulling is viewed in the household”, establishes that an encounter that never occurs can remain interrecursively consequential. A child reads a parent’s silence and the parent, presumably, reads the child’s. Organised noninteraction therefore closes a particular encounter while leaving being-with in place and the relation fully recursive, with each side interpreting what the other does not say. This is why the same silence can be experienced as protection and as abandonment, according to who is reading and what they had hoped to be asked.
The interactionist assumption that this exposes is that more articulation is epistemically or therapeutically superior. A theory centred on communication tends to treat the unspoken as a deficit to be remedied. The ethnography shows that successful life does not require maximal recursive openness. People need zones that are not continually available for interpretation, and closure is not pathological by definition. Without closure every bodily habit, feeling, decision and identity could remain perpetually available for interrogation, and that would be unliveable. The qualification is equally important. A closure that a person cannot reopen when they wish to, as in Heather’s case, or a closure imposed on them by others, as in Siobhan’s, damages coordination of another kind. What distinguishes a protective closure from a harmful one is who controls it and whether it can be reopened on terms the person can bear.
VI. Parents Learning Not to Intervene
The most conceptually productive material in our article concerns what parents learn. We interpreted it largely through relatedness, care and a “mutuality of being” in Sahlins’s phrase (Sahlins 2011). Read again, the material shows parents learning something more difficult, that care can require withdrawal from interrecursive control.
Parents begin from a model in which care means intervention. They notice, ask, instruct, correct, monitor and try to stop. We listed the repertoire of negative responses, which includes “censoring verbal reactions relating to appearance (shaming in front of others), physical punishment, and derisive comments” such as “why can’t you just stop?”, and added that these attempts at caring “fail because they increase tensions and, subsequently, they increase pulling behavior”. Camila describes the oscillation: “you want to talk to them about it, but they don’t want to talk about it. And you want to help them and you want to make them stop, so it’s like you go between being soft and gentle and ‘what can I do for you?’ to being angry because you’re like ‘why can’t you stop?’” Mark recalls that “you just wanna fix everything, so for me to look at her and say, ‘Hey Tiff stop doin’ that,’ that should be enough. But that’s not the way it works”. The model fails because the child’s body is not recursively controllable by the parent. Monitoring places a recursive burden on both sides, since the parent must watch and the child must be watched, and each reply from the child feeds the parent’s next attempt.
What the parents who have found their way to the TLC Foundation, a US organisation for BFRBs, describe in its place is a relation that remains in place while command is relinquished. Mandy came to see that when a child is “adamant and defiant about receiving help … it’s their journey to be on, it’s not ours.” Camila arrived at the same realisation, and we described the process as “disconnecting from the subjective experience of their child,” which “became part of the support process”. Mandy’s account of the early period shows what the earlier model cost. She would brush her daughter’s hair each night before bed and sob while doing it, and she says she now realises that this “might have made Frankie feel even more ashamed and guilty”. A routine of bodily care had become a nightly exposure of the loss.
Five positions need to be kept apart, because the ethnography passes through all of them and a theory that collapses them loses the learning. Abandonment is the withdrawal of care and attention. Nonintervention is a decision not to act on the other’s conduct. Nonintrusion is a restraint from questioning, watching and correcting. Withdrawal is a loss of contact. Respectful availability is the situation of someone who refrains from asking while remaining ready to respond if the other chooses to open the topic. Mandy, who attends the conferences with her daughter, administers a parents’ group and speaks of the daughter’s journey as her own to take, appears to occupy the last position. Someone can refrain from asking and remain ready to answer, and that is a particular architecture of being-with. A change in the way a parent looks belongs here too. Mark says that he no longer looks at his daughter and sees only the bald spot, the wig or the missing eyebrows, but sees “all of the wonderful things that are her”. The change concerns how much of her he takes up when he looks, and the difference is a reduction in what the child has to be available for.
Our account of fathers supplies a counterexample to the assumption that care is best measured by interactional intensity. Mandy says that her husband is “just not an active force in any of this” emotionally, and in the same breath: “He’s the first one to get online and book our flights for us. And that’s his way of supporting us”. We read this earlier in terms of gendered caregiving, with fathers as material providers, and the reading stands. A further point can be added. Support need not take the form of intense interrecursive disclosure. He coordinates through money, bookings and logistics, which are multimaterial and multisymbolic supports that allow other forms of relationship to occur, since the mother and daughter reach the conference because he arranged the journey. Whether this is sufficient is a question the participants answer ambivalently, but it shows that support and interactional intensity vary independently.
VII. Stigma, Face, and the Right Not to Become an Interactional Problem
Stigma theory is powerful because it shows that visible bodily difference can reorganise encounters. It begins analytically, however, once the potentially discrediting attribute has entered the field of another person’s attention. The ethnography suggests a prior politics, which might be described as a claim not to have one’s body taken up recursively by others at all. The word claim is chosen with care. Participants do not formulate it as a right, and they would be unlikely to use the term, so it is an analytic description of what their arrangements enact, and it is not a report of their idiom.
Goffman came close to the claim in his account of the territories of the self. The information preserve, the set of facts about oneself over which a person expects to control access in the presence of others, can be violated by questioning, prying or exposure (Goffman 1971: 38–41). Participants’ arrangements are efforts to keep a part of themselves within such a preserve. Goffman analyses the violation, however, as an offence committed by others within an encounter, and what his concept does not capture is the long-term architecture through which a preserve is maintained so that the occasion for violation does not arise.
The standard repertoire of stigma management has the same orientation. Passing, which succeeds when others do not detect the attribute, still takes the audience as its explanatory centre. The new question is how people construct conditions in which their hair never becomes an interactional object at all. In the same way face-work explains how people repair threatened encounters, whereas participants may prefer arrangements in which face never has to be repaired. An avoided interaction need not be a failed one. It can be a successful outcome, and when it is, nothing is visible to the analyst.
This is where the idea of an interactional non-event is useful. A successful day may contain nothing ethnographically dramatic because no one comments on the hair, no one asks, no one watches and no one corrects. Our article gives a trace of one such non-event. Tiffany, at ten, took the unusual step during the conference of going swimming with other children, “an activity she usually avoided due to her trich”. The usual days, on which she did not swim, are the ones in which an entire category of encounter was prevented, and we recorded them only by contrast with the exceptional one. From an interactionist viewpoint there is little to analyse in such days. From the perspective developed here, the absence is the achievement.
The sociological claim to be drawn is general. The most consequential effects of social organisation may sometimes consist in events that never happen. A child who is never asked, a conversation never started, a pool never entered and a gaze never met can be the principal products of a social arrangement, and a sociology confined to the occurrence of events cannot see them.
VIII. Support Groups: Selective Intensification Rather Than More Sociality
The TLC material at first appears to refute the argument. The conferences are intensely interactive. People hug, disclose, talk, cry and quickly form relationships. Participants and professionals call the community “like one big family,” and a conference rule is quoted: “There are no handshakes in this community, only hugs”. Tiffany sits slumped over her plate at breakfast on the first day, and by the second has made friends and hurries away from her mother in the morning to meet them. If noninteraction were the central achievement, one would not expect the most beneficial setting to be the most interactive.
The material shows instead that interaction is not inherently therapeutic, and that selective interrecursivity is. A great deal of dangerous recursive work has already been removed from the conference. The setting plausibly makes unnecessary the questions that elsewhere accompany hair loss: what trich is, whether it is real, why a person cannot stop, whether it is self-harm, laziness or a failure of discipline. Pulling does not have to be translated into any of these. The BFRB category, shared by the participants and the organisation, has settled many questions beforehand, and we recorded that for most of them the revelation is simply “knowing I’m not alone any more” and “realising there are others out there just like me,” which a researcher found more valuable to participants than the scientific findings he had come to present. The environment closes some recursive possibilities in order to open others.
Hence the support group should be understood through its distribution of askability, which differs radically from that of daily life. Some questions become unnecessary and some disclosures become safe. The conference’s rule that people hug rather than shake hands can be read in the same way. A handshake is a measured contact, whose depth each pair negotiates in the moment, and a standing rule that everyone hugs deposits a stance of openness in the setting itself so that it need not be negotiated one pair at a time. The point is an interpretation, since we reported the rule and not its use, but it fits the pattern.
The intensification is also temporary. On the last day the ballroom is full of people “hugging, and crying, sad to be returning to daily life without this support and freedom,” and Tiffany wrote afterwards that she was “pulling a bit more,” which she attributed to not being “around my TLC family anymore,” adding “I just wish there was someone closer to me who has trich”. The architecture that made openness safe stays behind at the conference, and the questions that were closed at the conference reopen in daily life. This connects to our original finding that recovery often means acceptance rather than the complete cessation of pulling, and that belonging to the TLC family reduces shame and isolation even when pulling continues. The therapeutic achievement on this reading is a better fit between openness and closure, with the aim of reaching neither the maximum of interaction nor its minimum.
IX. Relatedness Without Constant Interrecursivity
In our 2018 article we proposed that support groups create kin-like ties that can prove more supportive than immediate family relations, and chose “relatedness” over “social kinship,” “fictive kin” or “biosociality” as the term best able to carry the point. That argument is retained here. One assumption implicit in much kinship theory deserves criticism, namely that relational intensity is demonstrated by reciprocal engagement. When kinship is defined as a “mutuality of being” in which people share one another’s experiences, the definition invites the inference that closer kin know more, ask more and are told more.
Family life is made equally from what people do not ask, do not reveal, do not insist upon and do not reopen. This allows a different account of relatedness, in which recursive restraint is part of the relation. A mature relationship may be one in which the participants know which domains they are entitled not to enter. The point has wide application, and it is mentioned here only to show generality. Privacy in marriage, the autonomy of adolescents, medical confidentiality, professional boundaries, friendship, grief, sexuality, religious difference and political disagreement all involve zones into which a person admits others on terms and at times of their choosing, and in which the refusal to be entered is compatible with, and sometimes a condition of, closeness. A relationship in which everything is recursively available would not necessarily be more intimate. It might be intolerable.
The trichotillomania material gives this general point a specific content. The parents who learned to say that it was their child’s journey revised the form of their relatedness and did not reduce it, moving from a form that required the child to be available for correction to one that allowed the child to decide when to be available at all. The participants who gave up on a failed topic maintained a relation on terms they could survive, and did not withdraw from their parents. We can therefore propose that good relatedness requires not only capacities for recursive engagement but capacities for recursive abstention. The latter may be harder to acquire, as the parents’ learning shows, since they have to be learned against the strong inclination to help.
X. Institutions and the Positive Production of Noninteraction
The argument can now be scaled up from family life. Institutions often create arrangements precisely so that people do not have to negotiate every boundary personally. Confidentiality rules, closed doors, appointment systems, separate support groups, private online forums, diagnostic labels and professional jurisdictions all stabilise selective noninteraction. A confidentiality rule means that a patient does not have to ask each listener whether they may be trusted. A clinic door means that nobody has to be asked to leave.
The TLC provides the positive institutional example. The conferences, the Facebook groups and the organisational infrastructure allow participants to control who enters the interactional field. The parents’ group that Mandy administers is described as “a safe environment where they’re not feeling judged”, and it is restricted to parents. The conference organizers scheduled a workshop for “men only” because fathers might have additional struggles. The label BFRB itself helps create a community while reducing the need for repeated explanation, and we described its adoption by advocacy groups dissatisfied with psychiatric classification, which we related to Hacking’s looping, a change of a diagnostic label in response to the concerns of those living with it. The label is a deposit. It carries a settlement, achieved by participants and advocates, into encounters in which it need not be argued afresh.
This produces a contrast with coercive institutions. In Goffman’s account of the total institution (1961), the arrangements that remove interaction, the locked doors, schedules and records, are controlled by staff and impose closure on inmates, who are cut off from encounters they might have wished for and exposed to those they would have preferred to avoid. Institutional noninteraction can protect autonomy, as at the conferences, or suppress it, as in the ward. The analytical question is never simply whether interaction has been removed. It is who controls the closure, what forms of interrecursivity are prevented, and whether the resulting arrangement sustains viable coordination. Our material adds a third case, in which medical institutions offered no such arrangement. Participants perceived “a lack of empathy from medical professionals who were often said to ‘not understand’ and therefore be ‘unable to help’”, and the interaction that the clinic might have provided was precisely the one they did not expect to find.
This keeps the argument from celebrating withdrawal. The same arrangement can be a refuge and a cage, according to who holds the key and whether it can be opened from the inside.
XI. Toward an Anthropology of Noninteraction
Anthropology has traditionally privileged observable interaction, because ethnography itself depends heavily on encounters, conversations, interviews and participation. This produces a methodological bias. We see what happens and hear what people say, and we are less well equipped to observe the question never asked, the conversation avoided, the room not entered, the disclosure postponed, the person not contacted, the argument prevented, the gaze averted, the invitation declined and the category never activated. Yet these non-events may organise much of everyday life.
Our 2018 article itself contains an instance of the problem. We argued that structured assessments of family functioning “only work in cases when the condition is clearly diagnosed and when other family members are aware of it,” and that they “do not work when the condition is undiagnosed and when sufferers successfully conceal the condition from others in the family”. Instruments register what has entered recursive availability. Where noninteraction has been achieved, the instruments register an absence, and the absence can be read as the lack of a problem. Our own position illustrates the same point from the other side. One of us (Bradley) lives with trichotillomania, and participants told her things they said they had never told anyone, including family members. Her presence was a selective opening, a setting in which an otherwise closed topic became askable.
Four methodological questions follow. Which interactions are people actively trying to prevent? Through which embodied, material, environmental and symbolic arrangements do they prevent them? What work is displaced when interaction is successfully avoided? And under what conditions is noninteraction protective, coercive, caring or isolating? These are questions about conduct that leaves few traces, and the evidence for them may lie in contrasts, as with Tiffany’s swimming, in the accounts of participants who have learned what happens when interaction occurs, and in the points at which an arrangement fails.
The theoretical advance can be stated briefly. Interactionism begins with the encounter. An anthropology of noninteraction begins with the distribution of recursive availability. Who can ask whom what? Who must answer? Which bodies can be inspected? Which actions are monitored? Which questions are closed? Which encounters are prevented by architecture, law, habit or etiquette? The questions apply well beyond trichotillomania, to any social world in which a person’s body, history or conduct is more available to some than to others, and in which the pattern of availability is itself the product of work.
XII. Conclusion: A Good Life Requires Not Being Available All the Time
Beyond managing stigma in interaction, the participants in the trichotillomania ethnography were continuously deciding when to disclose and when to conceal, whom to tell, when to stop explaining, when to accept help and when to reject intervention, when to seek others who already understood, and when to create the conditions in which the hair could simply cease to be a matter for recursive attention. In 2018 we saw all of this through relatedness. One of the fundamental achievements of relatedness, the present reading adds, is knowing when not to interact. Noninteraction stands in no opposition to relatedness and is one of the conditions that make relatedness livable.
Interactionist theories have taught us how people manage encounters. The ethnography of trichotillomania reveals something equally fundamental, that much of social life consists in managing the conditions under which encounters will not occur. People build lives not only by responding to others but by arranging bodies, materials, spaces, symbols and institutions so that some responses never have to be elicited. A theory of living coordination therefore requires not only an account of interrecursivity but an account of its selective suspension.
A viable social world depends not only on being able to reach others, but on being able not to be reached by them.
References
Bradley, Bridget, and Stefan Ecks. 2018. “Disentangling Family Life and Hair Pulling: Trichotillomania and Relatedness.” Medical Anthropology. https://doi.org/10.1080/01459740.2018.1476509.
Goffman, Erving. 1959. The Presentation of Self in Everyday Life. Garden City, NY: Doubleday Anchor.
Goffman, Erving. 1961. Asylums: Essays on the Social Situation of Mental Patients and Other Inmates. Garden City, NY: Doubleday Anchor.
Goffman, Erving. 1963. Stigma: Notes on the Management of Spoiled Identity. Englewood Cliffs, NJ: Prentice-Hall.
Goffman, Erving. 1967. Interaction Ritual: Essays on Face-to-Face Behavior. Garden City, NY: Doubleday Anchor.
Goffman, Erving. 1971. Relations in Public: Microstudies of the Public Order. New York: Basic Books.
Hacking, Ian. 1995. “The Looping Effects of Human Kinds.” In Causal Cognition: A Multidisciplinary Debate, edited by Dan Sperber, David Premack and Ann James Premack, 351–394. Oxford: Clarendon Press.
Kidron, Carol A. 2009. “Toward an Ethnography of Silence: The Lived Presence of the Past in the Everyday Life of Holocaust Trauma Survivors and Their Descendants in Israel.” Current Anthropology 50 (1): 5–27.
Sahlins, Marshall. 2011. “What Kinship Is (Part One).” Journal of the Royal Anthropological Institute 17 (1): 2–19.