Stigma: The Shaming Model

According to a dominant view of stigma, a person is stig-matized within a community if sufficiently many people within that community hold a bad view of her. I call this the ‘ Bad View Model ’ . In this paper, I argue against the Bad View Model on the grounds that such beliefs are neither necessary nor sufficient for stigma, and that the account cannot explain the distinctive phenomenology of stigma, including certain vulnerabilities to shame. I then develop an alternative that explains these features of stigma, which I call the ‘ Shaming Model ’ . On this view, a person is stigmatized within a community if she is shamed by members of that community, and this is explained by their belief that she has deviated from some social norm and/or standard.


| INTRODUCTION
Say if I walked out of class because I could feel myself going into a panic attack, I'd get screamed at by the teacher.There'd be other kids out bunking class and if they saw me having a panic attack, they'd throw stuff at me.They'd throw bottles at me.They'd throw chewing gum at me.Throw their drinks at me.Dexter, 17, London (Testimony quoted in YMCA (2016, p. 33))   Dexter is stigmatized for their mental health difficulties within their school community.But what does this consist in?When we 'zoom-out' from the particulars of this example, what are the general features of stigma?
A natural starting point is the Bad View Model.Most of us are the targets of unfavourable attitudes held by at least some other people.What distinguishes the stigmatized person, on this view, is a critical mass of such opinion in some community to which she belongs.Stigmatic attitudes may also have a particular contentperhaps the stigmatized person is thought of as an inferior being.But the basic idea remains the same: we get stigma by multiplying instances, within some community, of people holding a bad view of an individual (perhaps as a member of some group they hold a bad view towards in general). 1argue that the Bad View Model is false.First, it is not necessary for an individual to be stigmatized within some community that its members hold any particularly negative attitudes about her.Second, such attitudes are not sufficient either.A mere distribution of attitudes will not explain the distinctive phenomenology of stigma.Stigma exerts a kind of 'normative' or 'evaluative' pressure in the stigmatized person's community.I do not mean that the stigmatized person does, or should, accept the stigmatic attitudes that members of the community express about her.But part of what distinguishes incidental acts of derision from those that form part of stigma is that the latter cannot be as easily brushed away. 2 Relatedly, stigmatized people are vulnerable to shamedespite often rejecting particular stigmatic assessments of themselves or the criteria to which such assessments appeal. 3,4To explain this, we must think of stigma not as an aggregate of unfavourable attitudes held about the stigmatized person, but rather as belonging to a social structure in which people's attitudes are in various ways interdependent.Thus, I also advance an alternative account of stigma that explains this phenomenologythe Shaming Model.An individual is stigmatized in a community if she is shamed by members of that community, and this is explained by their belief that she has deviated from some social norm and/or standard.I begin by reconstructing Goffman's (1963) account of stigma.This is for several reasons.First, his view instantiates the Bad View Model.I think this is, pretheoretically, a quite natural way to understand stigma. 5Second, Goffman's view is flexible and can accommodate plausible claims about 'passing' and stigma's relativity to social groups.Any account of stigma, including the Shaming Model, needs to accommodate these.Finally, the view is popular.As one author put it more than 40 years after the publication of Goffman's book: 'Even today, virtually all social scientists accept the broad definition of stigma developed through his work' (Lenhardt, 2004, p. 817).
In Section 3 I set out the objections to the Bad View Model.I then offer the Shaming Model as an alternative (Sections 5 and 6). 6Along the way, I provide an account of the characteristic functions of shaming (Section 4) and address shaming and stigma's relationship to attributions of responsibility (Sections 4 and 6).

| GOFFMAN
Stigma is a paradigmatic case of social hierarchy. 7Stigmatized people relate to some others as their social inferiors.
There are many contingent features associated with particular stigmashousing segregation may be such a consequence of the stigmatization of Blacks in the United States. 8These features can be of the utmost moral importance.
But we must set them aside in order to address stigma's essential features.
Goffman claims that 'a stigma… is really a special kind of relationship between attribute and stereotype' (1963,   p. 14).What is meant by 'stereotype'?They are 'the means of categorizing persons and the complement of attributes felt to be ordinary and natural for members of each of these categories ' (1963, p. 11). 9A social identity is thus imputed to us in a given context on the basis of such expectations (1963, p. 12).Stigma involves the deviation of some feature of ours (which may also be socially constructed, for example, our race 10 ) from this identity.
Goffman's use of 'stereotype' does not precisely track its ordinary usage today.Take the stigmatization of Blacks in American society viewed as a whole.For Goffman, this is understood in terms of their deviation from a stereotype of Americans as whitethe default state imputed to Americans.Though we are familiar with other concepts that cover similar territory, for example, the 'white gaze' (see [Yancy, 2008]), it is perhaps not so common to talk in terms of 'stereotypes' here.A more common use of 'stereotype', which does accord with Goffman's account, would for example pick out more particular social expectations placed upon Blacks in certain contexts.Take the expectation that Blacks should be particularly deferential to whites when occupying customer service roles. 11Being taken to deviate from this expectation may certainly incur stigmaand of course the expectation itself is presumably tied to some more general stigmatization of Blacks.
This last remark connects to another way in which Goffman's notion of 'stereotype' departs from ordinary usage.We talk, for example, about the stereotype of Black women as aggressive.But it is being taken to conform with that 'stereotype', rather than deviation from it, which we more often associate with stigma.The more general category of pejorative 'stereotypes' thus present a challenge for Goffman's account if we are to understand deviation (or perceived deviation) from a stereotype as both necessary and sufficient for stigma.Following Goffman's usage, then, we must say there is no stereotype of Black women as aggressive unless deviation from that expectation would in some sense be viewed as undesirable.The stigmatization of Black women that we often associate with (what we call) the aggressive 'stereotype' would have to be unpacked in other termsfor example, those pejorative attitudes are quite plausibly associated with some more general stigmatization of Black women.
Goffman's account instantiates the Bad View Model.This is because he appears to explain the mechanisms of 'categorization' involved in stereotypes by appeal to the way individuals 'buy-in' to the representations in question.
He says, 'By definition, of course, we believe the stigmatized person is not quite human ' (1963, p. 15, my emphasis)   and that we believe they are 'of a less desirable kindin the extreme, a person who is quite thoroughly bad, or dangerous, or weak ' (1963, p. 12). 12The 'buy-in' can also extend to the stigmatized person herself: 'The stigmatized individual tends to hold the same beliefs about identity that we do; this is a pivotal fact ' (1963, p. 17).So, it seems that stereotypes consist of nothing more than the readiness of large number of people in the community to assess others, and perhaps themselves, against certain expectationsthey think that people should conform to the relevant stereotypes, in the strong sense that they think these are good expectations to hold them to, and so form negative attitudes towards those who they believe fail to conform.This is perhaps not the only way to construe the account.
But it seems the obvious way to do so in light of the text.I will return to these affinities with the Bad View Model when I criticize it in the next section.
Let us put Goffman's account in more formal terms so that we may develop its core elements.'A' stands for a person, 'G' for a community to which A belongs, and 'T' for some set of traits. 13

Goffman stigma
A is stigmatized in G if, and only if: (1) A possesses T; (2) It follows from A possessing T that A deviates from the stereotypes available to A in G.
A noteworthy feature of the view thus stated is that both people who are (in Goffman's terminology (1963,   p. 14)) 'discredited' and people who are merely 'discreditable' are stigmatized.If A possesses T, and it follows from this that A deviates from the stereotypes available to A in G, A is stigmatized whether or not people know that A possesses Tthat is, whether or not A 'passes' as conforming to the relevant stereotypes.
Merely discreditable individuals do not escape all costs associated with stigma.A faces costs on account of deviating from the stereotypes available to A in G, that is in virtue of possessing T, even if members of G do not believe (because they do not know that A possesses T) that A deviates from the relevant stereotypes.As Goffman notes, for the merely discreditable person too, 'Shame becomes a central possibility, arising from the individual's perception of one of [her]   own attributes as being a defiling thing to possess and one [she] can readily see [herself] not possessing ' (1963, p. 18).
The prospect of discrediting itself, manifest in the kinds of attitudes held about discredited people (and those 'like them') in the community, and even potentially by the discreditable person herself, can be shame inducing. 14spite this, it is not clear that people who pass as conforming to the relevant stereotypes are in fact stigmatized.Intuitions are fuzzythis is to some extent a stipulative move.But the fact that stigmatization involves publicly recognized standings of inferiority gives us some reason not to count individuals who pass as conforming to the relevant stereotypes as in fact stigmatized.If members of G do not know that A possesses T, and so do not know that A deviates from the stereotypes available to A in G on account of possessing T, then A cannot have a public standing that makes reference to A's possession of T. 15 If we accept this reasoning, a simple amendment suggests itself.

Goffman stigma*
A is stigmatized in G if, and only if: (1*) A is taken to possess T by members of G; (2) It follows from A possessing T that A deviates from the stereotypes available to A in G.
Condition (1*) should not be read as requiring that all members of G take A to possess T. We must work with the intuitive idea that 'enough' members of G take A to possess T, or that members of G 'often' take A to possess T when they encounter A. We should not try to be more precise.
Finally, we should note that A is stigmatized relative to Gsome group to which she belongs. 16This allows for the possibility that whilst A is stigmatized relative to G, she may not be stigmatized relative to some other community or (communities) to which she belongs, call them 'G 0 , G 00 …'.This is illustrated by an example of Goffman's involving 'the prostitute who, although adjusted to her urban round and the contacts she routinely has in it, fears to 'bump into' a man from her home town… In this case her closet is as big as her beat, and she is the skeleton that resides in it ' (1963, pp.99-100).The sex worker is not discredited or even discreditable in her life in the cityshe conforms to some stereotype available to her within that community.But she also belongs to another communityin her home townin which all possible ways for her to conform to a stereotype in her community include that she is not a sex worker.Discrediting there remains possible.
Can A be stigmatized relative to any G? Goffman sometimes appears highly permissive about what counts as stigma. 17So, there is some indirect evidence for interpreting his account as answering in the affirmative.
Still, our ordinary concept of stigma yields counterexamples to this claim.Suppose G is an office that A only attends once a month as part of A's job.Suppose further that A is taken to possess T by members of G, and that A consequently is taken to deviate from the stereotypes available to A in G. Finally, suppose the stereotypes that are available to A in G are highly local relative to the rest of A's social worldthere are no similar stereotypes operative in the rest of the communities to which A belongs (G 0 , G 00 …) that take A's possession of T to be discrediting.I think we would be reluctant to grant that A is stigmatized relative to G on account of being taken to possess T. A is merely in bad standing in G.
It is no accident that the kinds of social identities that we often cite as examples of stigmatized classes are categories that have a degree of salience in many contexts, for example, racial, gender, and religious identities, sexual orientations, having a disability and so on. 18This is because it is a necessary condition for A to be stigmatized relative to G that G comprises a large enough part of A's social world.Or to be more precise, that there is some set of communities to which A belongs (G, G 0 …) that comprises a large enough part of A's social world.
A can then be stigmatized relative to each of the communities belonging to this set (relative to G, relative to G 0 …) in case there is some T that is similarly discrediting for A across the members of the set.Compare a national political community, which comprises a large enough part of A's social world on its own, with a set of smaller (sub-) communities to which A belongs (a school, a workplace, a sports club…) which, taken together, comprise a large enough part of her social world.A could then be stigmatized for her racial identity, say, relative to the national political community if there are stereotypes operative at that level which take A's racial identity to be discrediting.A could also be discredited relative to each of the smaller communities to which A belongs if there are stereotypes that operate independently in A's school, workplace, sports club and so on, that all take A's racial identity to be similarly discrediting.When I use 'G 0 in formulating accounts of stigma henceforth, I refer to a community to which A belongs that satisfies the condition above. 19his appeal to the idea of 'a large enough part of A's social world' allows for two possibilities.First, A can be stigmatized relative to G even if there are some other (sub-) communities to which A belongs, such as a friendship group, workplace, ethnic enclave and so on (G 00 , G 00 '…) in which the T which is discrediting for A in G is not here discrediting for A. Second, A can be stigmatized within a friendship group, workplace, ethnic enclave and so on, when the T that is discrediting for A within such communities is not discrediting for A within some wider community (G 0 ), a national political one, say, to which A belongs.
Borderline cases are possible.There are cases (like the office example) where the contexts in which T is discrediting for A are too fleeting or spontaneously entered into for A to plausibly count as stigmatized in that context on account of being taken to possess T. Equally, whilst within a racist society members of certain racial groups may find some respite amongst each other from stereotypes that are operative in the wider political community, those stereotypes often still loom large enough in their social world that this is a paradigmatic case of stigma.Suppose, however, some quirk of mine is discrediting within my family network.Is this a possible case of stigma?It depends on how large a part of my social world this family network comprises.Equally, if the quirk is discrediting within some wider community, but admired within the family network, the question of whether I am stigmatized within the larger community rests partly on how significant a presence that wider community exerts in my social world.Thus, greater insularity both reduces the risk of stigma within some wider community and increases the risk within the subcommunity itself.
I doubt we can decide borderline cases.But this is not a problem.Vagueness may be a feature of the phenomenon itself. 20What we can say is that there is a range of cases that clearly count as stigmas, and within that range those cases in which A's possession of T is discrediting in larger parts of A's social world strike us as more deeply hierarchical. 21

| AGAINST THE BAD VIEW MODEL
Goffman's account is flexible enough to accommodate some plausible claims about stigma.But, as an exemplar of the Bad View Model, it is vulnerable to serious objections.On Goffman's account, there are a set of stereotypes available to A in G.This just amounts to the claim that members of G expect A to conform with a set of expectations, and thus stand ready to form a series of dehumanizing beliefs about A if she deviates from these expectations.The stigmatization of A in G then simply involves the forming of such dehumanizing beliefs, on account of A being taken to deviate from the stereotypes available to A in G.
The first objection to make against this account is that it is not necessary for A to be stigmatized relative to G that members of G hold any particularly negative beliefs about A. It is enough that they express the bad view of A, even if most members of G do not 'buy-in' to the relevant judgments about how members of G like A ought to be.
An example may help. 22Suppose there is a social norm in a community that requires men to beat their female partners.And suppose further that males who openly defy this norm are subject to derision by their peers.It is still possible that very few males actually conform with the norm.Indeed, it is possible that many of them think it is a bad norm.The pressure to express negative judgements about group members who openly deviate from the norm, then, would not come from a sincere conviction that they merit such judgements.Rather, it would come from a desire to uphold appearances, believing they too would be subject to treatment of this kind if their lack of commitment to the norm were known.In this case, most males would be ignorant of their peers' lack of commitment to the norm and record of nonconformity in private.This ignorance allows for a pattern of males insincerely expressing negative judgements about their peers who openly defy the norm.The intuition I am pressing is that the lack of sincere negative evaluations of such people does not mean they are not stigmatized.
In light of this argument, it may still be sufficient for A to be stigmatized in G that members of G hold a bad view of A. But I reject this too.First, it is hard to imagine a community in which such beliefs are pervasive and yet are given little or no expression.It is harder still to uncover any firm intuitions about whether this would count as stigma.This is particularly so since stigma involves publicly recognized standings of inferiority.This requires at least some coordination and expression of the attitudes in question.But if paradigmatic cases of stigma involve members of G giving expression to a bad view of A, then why retain any direct focus on their beliefs about A? Perhaps members of G really do hold a bad view of A, and people like A. Or maybe they do notand simply act is if they do for some other reason.
There is a deeper reason to hold, against Goffman, that members of G possessing certain dehumanizing beliefs about A is not sufficient to account for the stigmatization of A in G. Stigma has a distinctive phenomenology: it exerts a kind of 'normative' or 'evaluative' pressure in the stigmatized person's community.The widely accepted association between being a stigmatized person and a vulnerability to shame is strong evidence for this claim.The Bad View Model cannot explain this phenomenology.
Goffman shows some sensitivity to this phenomenology when he notes that stereotypes are 'righteously presented demands ' (1963, p. 12). 23We might wonder why his view cannot accommodate this datum.After all, it is fairly plausible empirical assumption, which has done some work in contemporary political philosophy, 24 that our selfesteem usually depends upon the good view that our peers take of us.So, when a bad view of a person is pervasive in her community, she is vulnerable to internalizing that attitude, often resulting in shame.This is correct.But it cannot do all of the work of explaining the 'normative' or 'evaluative' pressure that a stigmatized person comes under.Internalization is part of the explanation why stigmatized individuals are vulnerable to shame.But much contemporary philosophy of shame also begins from the assumption that some stigmatized individuals experience shame associated with their stigma despite not accepting the stigmatic attitudes in questionor even the criteria on which these attitudes are based. 25possible response is that all the Bad View Model needs here is a revised account of shame, so that it is better placed to accommodate experiences of shame where the agent does not judge herself negatively. 26I do not rule out a resolution of this kind.But since I do not want to rest my account on any controversial claims about the nature of shame, it is worth pursuing an alternative model of stigma, to see whether it performs better at explaining this datum in a way which is ecumenical between competing accounts of shame.If the reader is not satisfied with this response, we should recall that there are other reasons still on the table for thinking that it is not necessary (and probably not sufficient either) for A to be stigmatized relative to G that members of G hold any particularly negative beliefs about A. This provides enough reason to look for an alternative to the Bad View Model.
These criticisms not only show that we must reject the Bad View Model, but also point to a way forward.If stigma is not explained by an aggregate of individual beliefs, and our account of stigma must supply some resources for explaining the 'normative' or 'evaluative' pressure that attaches to it, then we will need to appeal instead to a phenomenon that is both complex, and in a certain sense weighty.I suggest that a kind social structure fits the bill.An individual is stigmatized in a community if she is shamed by members of that community, and this is explained by their belief that she has deviated from some social norm and/or standard.Social norms and standards are themselves understood in terms of individual attitudes, but not as a mere accumulation of them.Rather, the relevant attitudes are in various ways interdependent, involving conditionalized preferences to conform with rules.

| SHAMING
Before describing this account in more detail, we must say more about its key component: shaming.Shaming involves 'the communication of a negative evaluation of a person' (McDonald, 2021, p. 139). 27The evaluation is of a personnot merely of some act of hers.The target is presented as shamefulthat (as Williams says about the emotion of shame) it would be appropriate for the target to feel that her 'whole being' is 'diminished or lessened ' (1993,   p. 89). 28In this way, we are able to account for some of the sense that a shamed person (and indeed a stigmatized one) is 'marked' by the evaluation of her.This also accounts for some of the difference between shaming and blamingwhen I blame someone, I may only communicate that she should feel guilty about some act of hers. 29I will continue to talk about shaming as 'expressing negative evaluations of persons'but the reader should take this as referring to the more specific form of evaluation outlined here.
Shaming has the characteristic function of 'shoring up' social norms and standards.In order to serve this function, it must have a certain expressive content. 30Shaming serves to 'shore up' social norms and standards by expressing that the target has violated the norms and/or standards in question. 31To be clear, this is not a claim about all shaming.Some social acts may count as shaming mainly because they involve an intention to induce shame in the target, even when they possess no broader social function. 32I set aside such shaming in order to focus on the kind that is key to the explanation of stigma. 33e claim that at least some shaming has the function of enforcing social norms has a firm footing in the literature. 34McDonald suggests that some shaming sanctions us 'with the weight of the shamer's own reproach and the… reproach of the masses' (2021, p. 151).Simply finding myself an object of disapprobation can be enough to hurt meperhaps even inspire shame. 35This gives the audience a reason to refrain from violating norms backed by such sanctions.In order to make salient such reasons the act needs to have a certain expressive content.This is the significance of the point that shaming expresses that the target has fallen foul of (in this case) a social normthat is part of how it can serve the function of enforcing that very norm.This mechanism depends partly for its success on members of the community endorsing the norms that are enforced by such means.Such endorsement may be necessary to induce certain negative self-assessments or other painful attitudinal responses. 36But shaming can also serve to emphasize how transgression will harm our reputation.This, as well as being important in its own right, can affect how people interact with us in various detrimental ways.This kind of sanctioning does not similarly depend on our endorsement of the norms.
Billingham and Parr claim that: '… public criticism can increase our common knowledge of (a) what the norm is; (b) what it demands in particular situations; and (c) when it has been violated.In these ways, public criticism enables us collectively to reaffirm our endorsement of the norm, and of the values it promotes or respects' (Billingham &   Parr, 2020, p. 1000).This indicates that shaming has an informational role as well as a sanctioning oneand that the transmission of such information can itself have a hand in bringing it about that social norms are more likely to be followed.The quotation also hints at the possibility that shaming can function as a kind of manifestation, even ritual celebration, of our ideals that invites the audience to rejoice and take pride in them.We could continue to talk about such mechanisms as enforcing social norms, in some extended sense of 'enforce'.But since 'enforcement' may imply a closer connection to sanctioning than is intended, I prefer to say that such shaming 'shores up' social norms.
The preceding discussion reveals that there are no essential mechanisms by which shaming functions to shore up social norms.Rather, shaming can induce a range of cognitive changes (forming new beliefsor recalling old onesabout the norms and how they interact with our interests) or affective changes (fear, shame, pride) in the audienceand any of these can help to shore up social norms.
Should we stop here?Why not think that shoring up social norms is the characteristic function of shaming? 37r one thing, it is not always obvious that the main relation in which shaming stands to social norms is one of shoring up.Rather, shaming is often what the norms require.
Consider this example from Dumont's study of the Indian caste-system: In the north-west of India, men of similar status are wont to meet around a hookah (…) which is smoked in turn… a cloth is interposed between lips and mouthpiece… Higher castes share the pipe with almost all castes excluding, apart from the Untouchables (…), only four other castes… In some cases, a different cloth must be placed between the pipe and the lips of the smoker.(Dumont, 1980, p. 84)   On one intelligible interpretation (no doubt there are others), the following of rules around smoking shames those with certain caste identities (as well as those people deemed 'untouchable').Notably, we might think they shame members of those castes who members of higher castes do not share the pipe withperhaps communicating the message that such sharing would 'pollute' members of higher castes.For similar reasons, the practices may shame the members of those castes for whom a different cloth is deemed necessary.Certainly, this reading is imaginable. 38at is notable for our purposes is that such shaming would not be a response to any infraction of social norms around smoking.Such shaming would occur even if the norms were adhered to perfectly.Such shaming is what the norms require.It is perhaps possible to expand the meaning of 'shoring up' social norms to accommodate this.If shaming is what the norms require, we may say that it 'shores up' those very norms in virtue of being a case of continued conformity with themthough this seems a stretch.
There is in any case another decisive reason to reject the view that the characteristic functions of shaming are exhausted by their role in shoring up social norms: some shaming would not even make sense if this were the case.Shaming, like stigma, does not simply present the target as 'bad'but rather as having violated some expectation. 39is is straightforward when the agent is targeted for (alleged) transgression of some norm.But in the case above no norm has been violatednor does anyone think this has occurred.Everybody knows that the shaming itself is simply more norm-required behaviour. 40at, then, is the violation that the shamed person is taken to have committed?Any plausible interpretation will allow that they are shamed (if indeed they are shamed) for their caste identities.Might we then say that they do, after all, violate a normnamely, a norm prohibiting belonging to their castes?This also seems a stretch.There is, plausibly, a conceptual connection between the idea of a norm and attributions of responsibility: if I take someone to have violated a norm, this implies that I take them to be responsible for having done so.But caste identities are usually unchosen features of personsnot something they can be held responsible for.Perhaps we could say that the people who engage in this practice simply have incoherent beliefs about the things people are responsible for.But, whilst this may sometimes be the case, it seems an uncharitable interpretation of such practices in general. 41, we must concede that some shaming does not have the function of shoring up social norms, but rather of shoring up social standards.Social standards, like social norms, are a kind of abstract object.Like social norms, they are rules that we can conform or fail to conform with, which exert a kind of 'pressure' when they are present in a community.But unlike social norms, taking someone to have violated a social standard does not imply that we take them to be responsible for the violation.This tracks our ordinary usage of 'standard' as in 'beauty standard'.There are many 'beauty norms'people are held responsible for 'upholding appearances'.But there are also features of people that are just deemed ugly, that nobody believes they can do anything about (being 'too short' or 'too tall' might be examples).People who have such features are held to violate a 'beauty standard'a rule that prohibits possession of these features. 42w might we understand the role of shaming in shoring up social standards?It follows from what we have said already that such shaming does not work by sanctioning people who violate the standards.As McDonald writes: 'A person watching a peer being shamed for being disabled will not think 'I had better avoid being disabled' (2021, p. 151).Still, everything which we said about the informational and celebratory roles of shaming in relation to social norms seems to hold true also for the relationship between shaming and social standards. 43Shaming in response to being taken to have committed a standard violation can make those standards salient to the community.It may even serve to invite the audience to rejoice and take pride in those very standards.In both cases, shaming serves these functions in part as result of its expressive content: it expresses that the target has violated the standard.
Shaming can perform functions relating to social norms and standards together.The shaming of a person for being disabled might remind us of ableist standardsbut they might also remind us of ableist norms governing interactions between disabled and non-disabled people.Likewise, shaming that sanctions violations of those norms can also remind us of ableist standards. 44

| THE SHAMING MODEL: FIRST PASS
We return now to elaborating an account of stigma in terms of shaming.Of course, we can have shaming without stigmatization. 45This can happen if the social group in which such shaming occurs does not have the properties of a social group (or set of social groups) that a person can be stigmatized relative to.And it can happen if such shaming is too infrequent to constitute much of a pattern.Stigma is a certain pattern of shaming A in G. 46 Let us develop our explanation of how the Shaming Model accounts for the 'normative' pressure that stigma may exert in the stigmatized person's community.In my view, this 'normative' force that stigma possesses is accounted for by the way in which the pattern of shaming A in G is explained by a particular connection to social norms.We still need to unpack this connection.But we can already see how the explanation gets going.One of the characteristic functions of shaming is to shore up social norms.So, shaming itself, when it is tethered to social norms in a community, receives a kind of 'authority' from the 'authority' of the social norms that it seeks to shore up.On my account, stigma is constitutively related to shaming, so can avail itself of much the same resources.
According to one proposal, social norms are rules requiring certain behaviours and attitudes that sufficiently many people in a community 47 prefer to conform with conditional upon their belief, first, that sufficiently many other people in their community will conform with them too; and, second, their belief that sufficiently many other people in their community believe they ought to conform with them. 48Imagine a community in which there is a social norm that people do not sell sex.In this community sufficiently many people prefer that they not sell sex, conditional upon their belief that sufficiently many other people in their community will also not sell sex, and their belief that sufficiently many other people in their community believe that members should not sell sex.
Given the normative expectations that people have, when a social norm is 'operative' as opposed to 'latent' in a social context, 49 being taken to have violated the norm is predictably met with the expression of disapproval. 50Because sufficiently many members of the community believe that members believe they ought to conform with the rule, there is already some incentive to engage in practices that shore it up, for example through expressions of disapproval that purport to demonstrate one's commitment to the rule.Often, these expressions of disapproval are shaming.Importantly, such expressions of disapproval may or may not be genuine.A social norm can be operative in a community, generating these incentives to shame, even when the relevant normative expectations are falsewhen, that is, contrary to our belief, sufficiently many people within the community do not believe that members of the community ought to conform with the rule. 51The Shaming Model can thus explain, unlike the Bad View Model, how a person could be stigmatized within a community, even when people within that community do not hold any particularly negative attitudes towards her.
The shaming to which stigmatized people are subject accounts for the hierarchical nature of stigma.Shaming expresses that the shamed person has fallen foul of a social norm.This explanation of stigma's hierarchical nature deepens our understanding of the stigmatized person's vulnerability to shameanother place in which the Bad View Model was found wanting.A stigmatized person may internalize certain stigmatic attitudes, perhaps on account of their pervasiveness within the community.But such 'buy-in' to stigmatic attitudes is not necessary for stigmaand this is no less true of the stigmatized person herself.What might explain her continued vulnerability to shame, despite rejecting the stigmatic attitudes in questionand even the criteria on which they are based?Part of the story, on my view, appeals to the fact that the stigmatized person is, or at least is taken to be, in violation of a social norm.Such norms carry a kind of 'weight' that is independent of the agent's assessment of whether she has good reasons to conform with them.Thus, violation (and perhaps even merely being taken to have committed a violation) can induce the kind of shame occasioned by falling short of some normeven if we reject the content of that very norm. 52re is a first pass at a more formal account of stigma.'S' designates a set of social norms.

Stigma
A is stigmatized in G if, and only if: (1) Members of G believe that A has violated S; (2) Members of G shame A; (3) ( 2) is explained by (1).
This account is flexible enough to cover cases in which shaming that is consequent upon being taken to have committed a norm violation either is or is not itself required by social norms.Consider a case in which A is a wheelchair user and S is a set of rules in G about how wheelchair users should conduct themselves in relation to nonwheelchair users (e.g., not demanding 'special treatment').Suppose A is taken to have violated S by members of G.There may be another set of social norms, call it 'S 00 .S' could require (at least some) members of G to shame A. Equally, such sanctioning may not itself be norm-required, but simply a predictable consequence of being taken to have failed to comply with S (recall, sufficiently many members of G believe that members believe they ought to conform with the rule, and so there is already some incentive to engage in practices that shore it up).The account allows for either possibility.This result conforms with our intuitions about paradigmatic cases of stigma.The social norms of G may even be hostile to the shaming of Athis does not disqualify the pattern as a case of stigma.

| THE SHAMING MODEL: AMENDMENT
What about stigmas where no norm violation is even thought to have taken place?Take 'period stigma'.On one interpretation of this phenomenon, nobody diverges from expectations laid down for them.People who have periods go to great lengths to conceal evidence of them. 53Other members of the community (and perhaps also people who have periods) enact 'disgust scripts' about the idea of menstrual bleedingshaming people who have periods.The shaming is not meant to sanction those who have deviated from a rule.The shaming itself is simply more normrequired behaviour.The formulation Stigma cannot explain this.This suggests that an amendment is needed to accommodate the wider class of cases in which A is not shamed by members of G for violating S, but rather where the expectation is that A will be shamed for being a certain kind of member of G and doing exactly what is expected of such members. 54This may be the right characterization for many stigmatized occupations, subordinate caste identities, and so on.
To put the point another way, shaming shores up both social norms and standards.But so far, our account of stigma has only mentioned social norms.It thus does not yet have the resources to account for a kind of pressure that may be exerted by stigma which is 'evaluative' rather than 'normative' and which does not involve attributing responsibility to its targets (e.g., for having periods).'E' designates a set of social standards in the formulation below.

Stigma*
A is stigmatized in G if, and only if: (1*) Members of G believe that A has violated S and/or believe that A has violated E; (2) Members of G shame A; (3) (2) is explained by (1*). 55,56e formulation Stigma* explains the stigma in the example of the wheelchair user in the same way as Stigma.
(2) is explained by (1*) when the shaming of A (the wheelchair user) by members of G is a predictable consequence of their belief that A has violated S (a norm requiring that they not demand 'special treatment' from non-wheelchair users).( 2) is also explained by (1*) when such shaming is itself a norm-required response to such violations.Both explanations may hold in a given caseso the stigma may be overdetermined.
But unlike Stigma, the formulation Stigma* can explain the example of period stigma on the suggested reading, where no norm violation is even thought to have taken place.One possibility here is that (2) is explained by (1*) in the following way.There are a set of social norms (again, call them 'S") that require (at least some) members of G to shame those who violate a social standard against having periods (E).The pattern shaming A in G is thus explained by the belief on the part of members of G that she has violated E (she is taken to have periods).Because they believe she has violated E, they also believe that that are required by social norms (S') to shame her (without thereby attributing responsibility to her for the violation).
Another possibility is this.Just as shaming can be a predictable consequence of being taken to have failed to comply with a social norm, without itself being norm-required, so too shaming can be a predictable consequence of being taken to have violated a social standard.So, suppose there are no social norms requiring (at least some) members of G to shame those who violate a social standard against having periods (E).Members of G might still shame someone (A) who they take to violate the standard.This would be a predictable consequence of such a belief, assuming that the operative standards within a community are underwritten by certain kinds of social expectation. 57So (2) would still be explained by (1*), and the case would still count as stigma on my viewwhich is, intuitively, the correct result.
Again, this explanation of ( 2) by (1*) may coexist with the previous one, so that the stigma is overdetermined.
And further overdetermination may occur when an individual is stigmatized simultaneously on account of being taken to violate social norms and standards, which may well be connected to each other.

| CONCLUSION
This view of stigma is complex.But that is as we should expect.A mere distribution of negative beliefs about a person is not necessary for stigma and cannot account for the 'normative' or 'evaluative' pressure that stigma exerts.
We must appeal instead to social structures.These, like much of fabric of our social worlds, are complicated indeed.

Euan Allison
https://orcid.org/0000-0001-5913-225X ENDNOTES 1 Kolodny has a parallel view about social inequality.He claims that a kind of social inequality 'seems to have to do with… Some having attributes (for example, race, lineage, wealth, perceived divine favor) that generally attract greater consideration than the corresponding attributes of others ' (2014, pp.295-96, emphasis original).Though he does not say this, it is easy to imagine that Kolodny would interpret stigma as a social inequality of this kind.It seems the most likely candidate of the three kinds of social inequality mentionedthe others are 'Some having greater relative power […] over others' and 'Some having greater relative de facto authority […] over others ' (pp.295-96, emphasis original).
of hierarchies like stigma in a similarly distributional spirit, that is, certain positive or negative attitudes to some people as the targets of such attitudes more than others.
personsalso 'stigmatization'.I do not offer an account of stigma in the previous sense.Those attracted to Goffman's account might focus on the 'T' element in some of the Goffman-inspired analyses below.We often talk about an act being stigmatizingwhich is again not my focus here (see (Garfinkel, 1956) for relevant discussion).
8 See (Anderson, 2010). 9There is much recent philosophical work on psychological mechanisms such as implicit bias that often underwrite group generalization.See (Beeghly & Madva, 2020).My focus at this point is the role of stereotypes themselves in constructing stigma, not the psychological mechanisms underwriting them, whether explicit prejudice or implicit bias or whatever else.
11 An illustrative example from Malcolm X's autobiography: 'It didn't take me a week to learn that all you had to do was give white people a show and they'd buy anything you offered them.It was like popping your shoeshine rag.The dining-car waiters and Pullman porters knew it too, and they faked their Uncle Tomming to get bigger tips.We were in that world of Negroes who are both servants and psychologists, aware that white people are so obsessed with their own importance that they will pay liberally, even dearly, for the impression of being catered to and entertained ' (2001 [1965], p. 161).
person.This is unconventional in, for example, literature on discrimination.
14 How does Goffman explain the stigmatized person's vulnerability to shame?Since his account exemplifies the Bad View Model, a salient possibility is that the stigmatized person herself is simply one more person who takes a bad view of her.This is internalization of the stereotypebecause she 'buys-in' to it, she judges herself negatively for deviating from it and is thus vulnerable to shame.This story will differ slightly depending on your preferred account of shamebut internalization is surely part of any complete account of the stigmatized person's vulnerability to shame.It will feature less prominently if we accept an account of shame that takes the beliefs that other people hold about a person to figure in her experiences of shame in a particularly direct way (see Calhoun, 2004; O'Brien, 2020).Still, I doubt it can provide the whole explanationand on this count the Bad View Model is found wanting in Section 3.
are socially situated.This will depend on how discrediting T is.On the Bad View Model, this is a matter of just how badly people are judged for being taken to fail to conform to the relevant stereotypes.On the Shaming Model, this is a matter of how stringent the social norms and standards that the stigmatized person is taken to have violated are.
are issued.
25 See endnote 3 for references.
26 Calhoun (2004) and O'Brien (2020) defend an account of shame that takes the beliefs that other people hold about a person to figure in her experiences of shame in a particularly direct way.Webster uses Velleman's (2001) account of shame to explain experiences of shame in response to racism as 'prompted by [the target's] inability to choose when her stigmatised racialised identity is made salient ' (2021, pp.535-36).See also (Velleman, 2001, pp.45-47).For accounts which hold that a person experiences shame only if she makes a negative evaluation of herself, (O'Hear, 1977;  Rawls, 1999, pp.388-91; Taylor, 1985, ch.3).
27 See also (Billingham & Parr, 2020, p. 1000) in connection with 'public shaming' and (Thomason, 2018, p. 181).Much work shaming penalties in the law focuses on their communicative aspect.Kahan (1996) praises their 'condemnatory' force and Massaro (1997, p. 649) notices that they express 'disgust'.We often need to glean contextual meaning in order to establish whether the relevant evaluation has been expressed.The expressive acts will not necessarily be verbaland their expressive content will not necessarily be conveyed intentionally.See (Anderson & Pildes, 2000).some hand in shoring up norms and standards.Rather, it is both necessary and sufficient that they are the kind of act whose performance generally or usually or on the whole serves this function.Though we are good at making these judgements in practice, providing a deeper explanation of them is more elusive.I do not attempt one here.
31 Shaming thus has a binary quality of the kind Tilly (1998) takes to be characteristic of 'categorical inequality'.The same is true, on my view, of stigma.
33 I expand on this below.In brief: the wider social functions of the relevant kinds of shaming are essential to explaining the 'normative' or 'evaluative' pressure that stigma exerts in the stigmatized person's community.By contrast, shaming that simply aims to induce shame in the target lends itself to being occasioned in highly incidental and bespoke ways that typically do not carry the distinctive force of stigma.The kind of shaming that interests us here may, or may not, also aim at inducing shame in the target.
always lead to actual shame, but that is its intent ' (2004, pp.203-204).It only commits us to the idea that there is some, not wholly contingent, connection between being subjected to acts of shaming and a vulnerability to shame.One resource from this paper which we might appeal to here is simply the claim that the shamer presents her target as shameful.See (Bartky, 1990). 36Billingham and Parr note: 'Members of the group generally take the norms to be authoritative for them.They accept or internalize these norms, such that they consider themselves duty-bound to comply with them and are likely to feel guilt or shame if they violate them' (2020, p. 999).If this is correct, we need not worry in general that members of the community may not endorse the social norms that are enforced by acts of shaming.The claim, however, rests on a particular conception of social norms (see for example Hart (2012 [1961]) who argues that in order to make sense of social rules as imposing a kind of 'obligation' we have to suppose that enough people in the community take up internal point of view' with respect to those rulesa kind of 'endorsement' of them).It does not follow from the account of social norms defended by Bicchieri (2017) and van Wietmarschen (2021).According to such views, it is possible that a social norm can be 'operative' in a community despite the fact that no members endorse it.It would suffice that enough members of that community prefer to conform with a certain rule, conditional on their beliefs (which they in fact have) that sufficiently many other members of the community will also conform with it and believe they ought to conform with it.That the second of these beliefs is false does not affect the claim that the social norm is 'operative' in the community.This is connected to the argument made against Goffman in Section 3, that no particularly negative attitudes about the stigmatized person are necessary to account for stigmaand this point extends to the stigmatized person (and so, on my view, shamed person) herself.To what extent is the threat of shame, as a sanctioning mechanism of shaming, compromised by this finding?As I mentioned in endnote 35, I think it is of some significance that the shamer presents her target as shameful.I provide some further resources for making sense of the shame of stigmatized subjects in Section 5.
38 I am not committed to any particular reading of this example.Perhaps adherence to these norms does not shame members of certain castesit may simply connote the absence of a kind of esteem.
39 I take this to follow from the claim made earlier that the shamer indicates that it would appropriate for the target to feel that her whole being is lessened or diminished.The idea of violating an expectation is also a key feature of the examples of shaming discussed so far that shore up social norms.This too gives us some reason to expect further cases of shaming to replicate this feature.Manne (2019, pp.53-54) claims that misogyny, which serves to 'police' patriarchal norms, can be visited upon people who do not actually deviate from patriarchal norms nor are taken to have done so.For the reason mentioned here, we should not understand such forms of misogyny as involving shamingthey are simply acts of terrorism.
'non-agential' forms.The latter enforces 'social standards' and does not involve holding a subject responsiblemore on this in a moment.We 'non-agentially' shame a target by expressing 'objective' attitudes such as disgust and contempt, that involve, in some ways, not engaging with a person in their capacity as an agent (2021, p. 150).When we 'agentially' shame a target, by contrast, we express 'reactive' attitudes, that are responses to a person's will, and which thus involve the imputation of responsibility to the target.Such shaming enforces social norms.One quibble I have with McDonald is that she only mentions 'non-agential' shaming in connection with hierarchy (2021, pp.153-55).But, as the discussion below shows, a person can be stigmatized when they are taken to have violated a norm. 42We need a more detailed account of social standards.I do not commit to one here, but they might be modelled on existing accounts of social norms.In the spirit of Bicchieri (2017), we could think of a 'beauty standard' pertaining to height as follows: there is a rule against being 'too tall', and sufficiently many people in a community prefer to conform with it (prefer not to be 'too tall'), conditional upon their expectation that sufficiently many other people in the community will conform with it (that is, the belief that there are sufficiently many people who are not in fact 'too tall' by this measure) and believe they ought to conform with it (that there are sufficiently many people who believe members of the community ought not to be 'too tall').The 'sufficiently many' clause that attaches to the 'empirical' expectations here may not be a majoritarian one.It is a well-known fact about many 'beauty standards' that most people violate them.
44 So, we should not think of McDonald's (2021) distinction between 'agential' and 'non-agential' shaming as involving mutually exclusive categories.There may, however, be some temptation to say that all shaming is really about standards.This stems for the thought that communicating that someone has violated a norm is not yet to shame them.Blaming also does this.So, we need to appeal to the idea that shaming expresses a negative evaluation not merely of the act performed by the target, but of the target herself to the effect that it would be appropriate for her to feel that (as we put it in an earlier discussion) her whole being is lessened or diminished.We might want to say that when we appeal to such 'ought-to-bes' we are dealing with standards rather than norms.McDonald comes close to making the claim about 'ought-to-bes'but appeals to other resources to draw the distinction between shaming and blaming (2021, pp.152-53).So even when shaming is occasioned by norm violations, we would say that it counts as shaming in virtue of drawing attention to standard violations.Thus, the somewhat mysterious distinction between evaluating the act of a person and evaluating the person herself gets unpacked in terms of the distinction between evaluating relative to a norm and and standards.We might still wonder how stigma, on my view, is related to the ordinary notion of a stereotype (as opposed to Goffman's technical notion).There could be many complicated connections between stereotypes and social norms/standards in any given case.For example, some stereotypes attaching to a group may give rise to social norms that require its members to conform with them (the expectation that Blacks should be especially deferential to whites when occupying customer service roles may be a relevant example).In other cases, a group may be stereotyped as liable to commit social norm violations (the stereotype of Black women as aggressive may fall into this category).I do not have a general account of these interrelations.

ACKNOWLEDGMENTSI
would like to thank Han van Wietmarschen for comments on several versions of this article.Thanks also to Lucy O'Brien and an anonymous reviewer at this journal for comments.I presented ideas from this article at the UCL Philosophy Graduate Work-in-Progress Group, UCL Philosophy Society, UCL-Leipzig Graduate Philosophy Conference, and UCL Philosophy Graduate Conference.I am grateful to the attendees for their feedback, especially Kenta Sekine and Victor Weber.Special thanks to Jonas Held, who acted as a respondent to my paper at the UCL-Leipzig Graduate Philosophy Conference.This work was supported financially by an AHRC London Arts and Humanities Partnership Research Studentship (Training Grant Reference Number: AH/R012679/1).