Scholarship on shaming often focuses on its negative effects (Reference NorlockNorlock 2017), but some recent scholarship defends the social and political uses of shame in ways that would seem constructive for feminists. Jennifer Jacquet's recent argument that effective shaming stigmatizes bad practices and challenges inconsistent or hypocritical norms fits effectively with some feminist political analysis and advocacy (2015, 174). By Jacquet's logic, publicly identifying and shaming sexist statements and practices is an effective strategy, given that people and groups want to protect and enhance their reputations. However, the reality of feminist shaming as an effective tactic is more complicated than this. Shaming that asserts authority over a larger and heterogeneous group makes questions of credibility and authority visible and disputable, and depends for its success on a friendly reception. Feminist shaming, particularly when done by the relatively marginalized against those with greater social authority and capital, is thus more visible and potentially dangerous than its more conventional, top‐down counterpart. Two recent cases of attempted online feminist shaming illustrate this complication; in each case, women called out sexist jokes by male colleagues in STEM professions. Each shaming resulted in a shame backlash against the shamers, in which shame was redirected back on the person attempting to shame. In other words, a shame backlash is more than simply the refusal of a judgment of shame (I am not embarrassed or stigmatized by what I did), but an aggressive redirection of shame (you, the shamer, should feel shame for trying to diminish me. Your attempt at shaming reveals your lack of character). In these two cases, the women who identified sexist utterances and practices found their own authority, competence, and credibility publicly and aggressively undermined. Shame backlashes turn the power of the shame back onto itself. Shame backlashes, in essence, function to tell aspiring feminist shamers: who do you think you are, and how dare you speak for us? The two elements that drive shame backlashes are the relationship between status and credibility in shaming, and how reliant shaming is on a friendly audience. My analysis demonstrates that shaming is a risky feminist tactic.

Shame and Shaming

In this argument, I am following Martha Nussbaum's definition of shaming as a stigmatizing judgment, where an actor or group condemns another actor or group for failing to adhere to a shared ideal or norm. There must be an element of publicity to shaming; if it is not done before an audience (in person or simulated), there must be a sense of its ability to be publicized. There is an assumed third party: I cannot shame people simply because they do not live up to my expectations; my expectations of them must refer to a larger community (Reference NussbaumNussbaum 2004, 184–86). Shaming is not act‐based, but directed at the full self. It is totalizing, and functions by stigmatizing the object of shame; she is rendered less than a full or “normal” member of the community (221). This is why so many formal shaming punishments were bodily in nature: tattoos on disgraced Romans, iron masks worn by medieval German gossips, stocks or cages for public display of wrongdoers. Tactics of physical marking and separation are ways of temporarily or permanently assigning someone the status of other or less‐than. Although contemporary shaming is usually not physical, its effects can still be longstanding.

Nussbaum differentiates shame from guilt in part on their scale: shame is a totalizing judgment, in which a self is found unworthy, whereas guilt is applied or experienced merely in response and in proportion to a single act. One's identity is literally “spoiled” by shaming, which indicates a sense not just of totality but permanence (176). I experience guilt, not shame, when a librarian sends me a chastising email about my long‐overdue library book; the email is aimed simply at correcting my careless act, and returning the book eliminates the guilt. By contrast, a public list of those with lost books shames me. Seeing my name publicized as a miscreant reduces my reputation for integrity, and per Nussbaum, I would find it hard to recover my reputation in the eyes of the university community. My reputation as a reliable person, which is a relevant part of my professional identity, is compromised when I am publicized as a book truant. Thus shaming experiences are also alienating. In being shamed, someone identifies and recognizes the legitimacy of the community criticism, and recognizes oneself through the scornful eyes of the community (see Reference TaylorTaylor 1985, 60). This alienating judgment reveals shame as, in Nussbaum's assessment, “broadly moral” (204).

For Jacquet, the alienation of shame is not simply a moral act, but one that can be put to effective political use as a way of urging constructive public action from groups on behalf of less‐well‐publicized causes. Working from her background in environmental science (which is not dissimilar to feminism in the sense of having some but not widespread public support), and using an analysis of shame that largely corresponds with Nussbaum's model, Jacquet contends that there are circumstances in which shame can be politically effective. Whereas conventional shaming is done by those with more social power or status against those with less, Jacquet creatively defends shaming as a tool for political change particularly for the powerless against the powerful. Her defense is conditional rather than absolute: shame is effective when a shared and desirable norm has been violated, when it is socially contextualized or tied to bad shared practices, rather than merely directed at individuals qua individuals, and when no formal punishment is readily available (Reference JacquetJacquet 2015, 100, 106, 174). This defense of shaming as informal social correction would appear to be ripe for feminist uses. Her examples of successful political shamings (among many, lists of those who fail to pay taxes, consumer campaigns for companies to adopt environmentally friendly practices) all cohere around a common theme: pressuring someone or a group to recognize that their actions have wider social effect than they otherwise assume. Many past campaigns on feminist issues (rape awareness, sexual harassment, equal pay) have had precisely that effect, so it would seem, on the face of it, that there is substantial feminist potential in this strategy.Footnote ¹

The common gap I see in Jacquet's and Nussbaum's approaches is in their minimizing of the role of the audience in shame. Nussbaum focuses carefully and acutely on what it is to experience shame and to be shamed, but spends comparatively little time on the importance of the audience (whether real or imagined) and its reaction to the shame. To be specific, Nussbaum restricts the role of the audience in the shame experience: it is not “essential,” but intensifies or amplifies the experience of shame (Reference NussbaumNussbaum 2004, 191). The audience functions as a mirror‐reflection of how the shamed person sees herself, but more intensely so. To take my library example, the audience—whether literal or figurative—for me are people who equally value public reputation for reliability and responsibility. Their values, whether ostensive or practiced, are my values; their responses are identical to mine.

In other words, Nussbaum, as with many other theorists of shame, takes for granted the comprehensibility and acceptability of shame: that there is a shared body of values among shamer, shamed, and audience, and thus the shaming will be effective. This treatment of audience both minimizes its importance to a theory of shame, and also reduces its role to a passive one; the audience exists merely to amplify the judgment being passed. Perhaps because so many of the examples of shaming we see in the shame literature focus on top‐down status‐shaming, this treatment may be less controversial.Footnote ² But the role of the audience is particularly important when we consider bottom‐up or socially disruptive shaming, where a marginalized speaker or group challenges the norms of a much larger or homogeneous group (even and especially a group in which the shamer claims membership, as we will see). Theories of shame that focus primarily on its ability to maintain individual or group integrity (Reference TaylorTaylor 1985) minimize the power differences of shame experiences and the ways in which group integrity can perpetuate discriminatory or harmful norms. The presence of shame can be an early‐alert system for the fact that standards we hold may, in part or whole, be culturally determined in ways that are not always supportive of personal agency and autonomy (see Reference ManionManion 2003, 37). A woman's momentary experience of shame when a man tells her to smile more at work may not indicate embarrassment at failure to live up to an ideal of femininity, but rather she might be experiencing a subconscious cue of a discriminatory social norm that is worth relitigating. In these instances of shaming, our feminist shamers are attempting to use shame to call their audiences’ attention to the ways in which claimed norms of nonsexism are not being honored; this call can succeed only if it is recognized by the broader audience to whom it is directed.

For her part, Jacquet's primarily tactical discussion of shaming pays scant attention to the role of the audience; she acknowledges that the audience needs to be worried about the transgression, but her brief discussion of this (Reference JacquetJacquet 2015, 100–102) does not contend with the possibility that audiences can be meaningfully diverse, and that a stated concern for a norm violation could be selective or theoretical only. Although Jacquet does briefly acknowledge that the shamer must be seen as trustworthy by the intended audience (111–12), her discussion does not complicate trustworthiness in ways that are helpful to those shaming up (who, again, are those to whom her discussion is directed). She praises TV host Jon Stewart for his selective shaming, and newspapers like the New York Times and the Wall Street Journal for their national credibility. Given the dramatic gap between institutions like these papers and national media figures and those who would typically want to make use of upwards shaming, these examples are not helpful in thinking through the tactics for shaming up.

Audiences of shaming matter because shaming is a hybrid act. It is an epistemic as well as an ethical act (I judge this behavior as unworthy of this stated community norm), and to be effective relies upon a friendly social reception on both accounts.Footnote ³ In other words, the relationship between the act of shaming and the audience to whom the shaming is directed deserves more philosophical attention.Footnote ⁴ Given that shaming is both an epistemic and a social assertion of credibility and authority, in cases where someone from a marginalized position is asserting the right to speak on behalf of a wider or socially dominant community, a friendly reception—which would be required for the shaming to be effective—cannot reasonably be assumed. There are epistemic and ethical acts on all sides—the shamer, the shamed, and the audience of the shaming—and especially when the shaming is disruptive or critical of a dominant group, epistemic agreement should not be assumed.

The form of epistemic disagreement we see in the shame backlashes I will discuss is particularly troubling. In these two cases, the backlash consists not merely of the targets rejecting the shame, but the audience enacting its own explicit credibility‐shaming that grounds and leads to status‐shaming. In other words, so much of the shaming of the feminist shamers in these two incidents focuses on what they do or don't know—as professionals, about social norms and behavior—as opposed simply to judgments of their ostensive failed characters or absence of ethics. There is a disproportionate focus on their professional and social credibility and competence, compared to other instances of shaming we see discussed in the literature. This poses distinct risks to those who would adopt shaming tactics as a feminist stratagem, particularly if one hopes to “shame up,” or shame the more powerful.

Let me be clear about the way in which the credibility assertion of shaming is epistemic as much as social. In successful shaming, audiences recognize and legitimize shamers as reliable witnesses to shaming; they judge them as epistemically and socially credible. To be explicit, because it is so common for shaming to be done from a top‐down status position, the default assumption is that the shamer has social or epistemic credibility; in formal shaming, shamers are judges or people of power, and in informal shaming, shamers are typically at least socially powerful (the popular marginalizing the misfits). But for shaming to be a successful subaltern strategy, as Jacquet contends, both an assertion and an acceptance of credibility needs to occur; in other words, bottom‐up shamers are implicitly asserting their epistemic and/or social competence to shame (I am making a credible judgment), and for the shaming to be successful, this credibility must be acknowledged and recognized. It is this link that I see not only broken in these instances of shame backlashes, but explicitly rejected. Feminist shamers’ professional and epistemic credibility is denounced and denied. It is important to be mindful of this, because denouncing credibility in this way carries with it longer‐lasting professional and economic harms, as I will show.

Shaming Events

The two incidents of feminist shaming I will briefly present here adhere well to Jacquet's model. In the first, Adria Richards, a web developer and blogger attending a 2013 technology conference, overheard two men sitting behind her making juvenile and sexist jokes about a female presenter on stage. She quickly tweeted about it: “Not cool. Jokes about forking repo's in a sexual way and ‘big’ dongles. Right behind me,” and in a follow‐up tweet called attention to their violation of the conference's stated code of conduct (Reference RichardsRichards 2013). In the second, Connie St. Louis, a science journalist, tweeted about Nobel biochemist Tim Hunt's sexist commentary during his toast at a conference for women science journalists in 2015. Her initial tweet was a screenshot of her brief news story, saying in part: “[S]exist speaker Tim Hunt [Fellow of the Royal Society]… stood up on invitation and says he has a reputation as male chauvinist. He continued, ‘let me tell you about my trouble with girls 3 things happen when they are in the lab; you fall in love with them, they fall in love with you and when you criticize them, they cry’ not happy with the big hole he has already dug he continues digging. ‘I'm in favour of single‐sex labs’ BUT he ‘doesn't want to stand in the way of women.’ Oh yeah! Sounds like it?… Really does this Nobel Laureate think we are still in Victorian times????” (Reference St. LouisSt. Louis 2015). She made the connection between Hunt's behavior and the status of women in British science more explicit in the following day's tweet (June 9), in response to the Royal Society's asserting its commitment to a “diverse science workforce”: “Do you not take any action when one of your fellows and especially a Nobel laureate, is so embarrassingly off your diversity message and behaves in such a sexist manner? May I also ask that if you are so committed to a ‘diverse’ science workforce why there has never been a female President of the Royal Society… in its entire 350+ history?”

There are several relevant parallels between these two events. Both Richards and St. Louis were responding to sexism that manifested as jokes. Both responded relatively quickly via social media. Although, initially, both simply identified the bad behavior as individual events, both also then made explicit connections between the jokes and the underlying institutional and structural sexism in their respective fields. Lastly, as women of color, both Richards and St. Louis were doubly underrepresented, and doubly visible, in their fields. Thus, there are several concrete ways in which both events correspond well to Jacquet's stated model for productive advocacy shaming. Both cases represented a violation of a social norm, in neither case was there a clear mechanism for a formal punishment, and in both cases, the shaming represented a social call for a social response. One parallel whose significance is crucially relevant is that both sexist incidents occurred in locations explicitly premised upon their professions’ recognizing and trying to improve the position of women in their profession. The conference Adria Richards was attending was publicizing its women‐friendliness in many respects (attendance statistics, code of conduct, presence on panels), and Tim Hunt made his joke at an international conference for women in science journalism, which featured panels about how to address the problem of sexism in science.Footnote ⁵ Thus both groups would have had a clear, positive incentive to respond constructively to shaming. Nussbaum's assumption of an audience that functions to amplify or intensify the judgment of the shame should have been operative here. And finally, both Richards and St. Louis were speaking to professional norms, as professionals.

Both Richards and St. Louis were on solid epistemic ground in directing their shame not primarily at individuals but at professions. Richards noted that the organizers of the conference celebrated the fact that 20% of the attendees were women (Reference RichardsRichards 2013); given that this particular conference supports and promotes diversity initiatives, pride in an 80/20 gender ratio demonstrates just how underrepresented women are in the tech world. Similarly, St. Louis's initial column about Tim Hunt's remarks (Reference St. LouisSt. Louis 2015) criticized the systemic underrepresentation of women in British science leadership.Footnote ⁶ Available data on the representation of women in STEM validates these criticisms.Footnote ⁷ There are also persistent climate issues for women in STEM: sexual harassment has been documented as a widespread problem in several natural‐science disciplines, particularly for early‐career women (Reference Clancy, Nelson, Rutherford and HindeClancy et al. 2014; Reference Richey, Clancy, Lee and RogersRichey et al. 2015). Two new policy announcements threatening funding withdrawals from institutions that fail to deal effectively with sexual harassment suggests public support for and recognition of the value of a nonsexist climate in STEM (Reference BoldenBolden 2016; National Science Foundation 2016). Thus, Richards's and St. Louis's employment of shame to set an example that bad behavior should not be socially tolerated appears to be justified by the larger social context in which these events occurred. Their purposes were less about criticizing the behavior as such, but the behavior as a reflection of a norm of sexism that is ostensibly no longer openly tolerated in these professions. Neither shaming attempt asked for a formal punishment for transgressors. St. Louis redirected her focus quickly and exclusively onto the Royal Society and the larger sexist track record of British science. Richards focused more on the behavior of the individual jokers—her only request for sanction, in her third tweet, was a request for PyCon to enforce its code of conduct and have the men leave the conference—and her response stayed at the merely social, rather than the lastingly punitive. (She did not name the men or call out their employers.) Thus, Jacquet's indicators of effective shaming appear to have been followed in both of these cases.

In both cases, however, institutional or systematic discussions of norms and identity did not happen. Both sexist “jokers” faced immediate and material consequences: one of the PyCon programmers (“Hank”) was fired from his job, and Tim Hunt promptly resigned his honorary positions at University College London, The European Research Council, and the Royal Society.Footnote ⁸ But it was Richards and St. Louis who both became the focus of longer, serious, financially and even physically threatening responses. It is not simply the seriousness, but the nature of the backlash that Richards and St. Louis experienced that I think is telling. Both were shamed for asserting epistemic and professional credibility; their attempts to shame their professions for sexism were revalued as themselves shameful acts. Particularly given the studied moderation of their initial criticisms, I believe that the double standard between Richards's and St. Louis's initial criticisms and the ferocity they faced is in part due to their assertion of epistemic and social authority. Their critics find their assertion of authority over a large and homogeneous group much more than risible; they find it objectionable.

Shaming as Assertion of Epistemic and Social Authority

These acts of shaming were both an epistemic claiming of authority and an attempt to challenge social privilege. Amitai Etzioni praises shaming as “deeply democratic” because it reflects a community's values (Reference EtzioniEtzioni 2001, 42). But for a shamer to so articulate a value, she must see herself as epistemically and socially empowered to speak on behalf of that community. This empowerment is straightforward in formal or judicial shaming, as discussed by Nussbaum and Etzioni, but is disputable and presents risks in informal or social shaming (as defended by Jacquet, and enacted by Richards and St. Louis). In both cases, we see Richards and St. Louis taking explicit steps to justify their ability to speak on behalf of a broader professional community.

The language Richards and St. Louis use in their more considered arguments reveals their speaking‐for and on behalf of a larger community of science workers. Richards closes her blog post by saying “Let's unify the message to our daughters and to the women developers we work with… . we need to build bridges and be aware of our actions” (Reference RichardsRichards 2013; emphasis added). Her pointed and repeated use of the first‐person plural, clearly directed not simply at women programmers but at the programming community in full, hypothesizes a possible future, better version of the community; it implicitly frames “Hank” as someone who violates norms the community ought to more consistently honor.Footnote ⁹ Richards is calling for public witnessing of bad behavior as violating a shared norm, which may thus lead to a recommitment to a more pluralist coding community in the future. Although St. Louis's speaking‐for and ‐to is located primarily on the institution of the Royal Society rather than the full community of British science, her chagrin at the story's focus entirely on Hunt reveals this broader concern: “the narrative of the rogue scientist belies larger, institutional failings” (Reference St. LouisSt. Louis 2015). Each, in short, sees her audience not just as the individual sexist jokesters, but the larger communities of STEM colleagues.

Both shamers make a point of positioning themselves as credible representatives of these broader professional communities; they assert professional identities and membership in the groups on whose behalf they are speaking. St. Louis notes that she works both as an academic and a science journalist, and makes the norms of her work visible (corroboration of her observations with journalist colleagues) to demonstrate professional credibility (Reference St. LouisSt. Louis 2015). In doing so, she explicitly positions herself as speaking on behalf of an “us” of science journalists, about the community of science and its failure to live up to stated norms. Richards's blog post explicitly positions her in terms of her history with programming, her work as a developer evangelist, and the ways in which this work addresses both technical challenges and social challenges like sexism. Richards goes beyond simply asserting technical competence and identity when she acknowledges how her work often addresses issues of sexism in the programmer community. She asserts her dual authority as both a member of the tech community, and as someone conversant with feminism, who can recognize everyday sexism when it occurs and as something worthy of critical and professional response. She is thus both inside and outside the tech community: a participant in it, but also outside its usual demographic and power dynamics, and hence capable of observing and constructively responding to injustices within it. On her blog, Richards describes many incidents of teachable moments in her work, where she talks through issues of implicit and explicit bias in companies on whose behalf she's working. Both Richards and St. Louis are explicit about asserting professional, epistemic, and social credibility so that a larger community will take their testimony seriously.

Shame Backlash as Community Closure

In Nussbaum's theory of shaming, the intended audiences of STEM workers, particularly at these conferences espousing gender inclusion, should have responded constructively to these assertions of credibility, and taken up an invitation to examine collective bias. However, this taking‐up did not happen in either case; rather, there was a backlash: a striking reversal of shame and a reassertion of narrower community identity (a community closure) in the face of public challenge. In other words, it was not simply that Richards's and St. Louis's claims were ignored or rejected; Richards and St. Louis, as aspiring shamers, themselves became the objects of shame and rejection from their professional communities. They became the objects of sustained, public, and shaming scrutiny; they were condemned as professionals and thus diminished in both their epistemic and social authority. St. Louis is criticized for teaching at a second‐rate university, condemned as a practitioner of “sloppy” or “irresponsible” journalism, and described as “too intellectually limited to recognize irony” (Reference SaulSaul 2015; Reference ForemanForeman 2015, respectively). She is dismissed as a mere science journalist who does not have the standing to criticize a Nobel laureate (Reference AdamsAdams 2015). Her background and journalism are examined for inconsistencies; the one purported instance of resumé‐inflation is assessed as undermining her credibility (Reference AdamsAdams 2015).Footnote ¹⁰ Thus, she is judged as professionally inadequate (where she teaches, the quality of her work), and thus unreliable. Commenters on tech blogs describe Richards as “dumb,” “to [sic] dumb for logic,” “racist,” and a “cunt.” Her professional competence is minimized; her work is described as marketing only, with no technical content. She has no professional standing to comment on programmers, according to these judgments, because she is an illogical marketer. Her Twitter feed is scrutinized for instances where she referred to or made jokes about sexuality. Two jokes found—relevantly, jokes about sexism rather than sexist jokes—are cited as undermining her authority to speak authoritatively on gender issues. In other words, the criticism focuses persistently and explicitly on Richards's and St. Louis's worthiness of being considered legitimate members of the programming and scientific (or academic) communities. Both of them are judged as not really in their professions; as journalists or marketers, their work is emptied of any technical content.

Their authority is not challenged simply on technical grounds but on social as well: both Richards and St. Louis are criticized for their failure to read and respond appropriately to social situations. This tactic renders the feminist shame socially incomprehensible: the audience claims there is no sexism to see. “Hank” criticizes Richards online for failing to confront him directly at the time; by publicizing his remarks to conference organizers instead of turning around and challenging him, he claimed, she made too much out of what was simply an intended joke (Hacker News 2013). Similarly, Tim Hunt's defenders used a twelve‐second segment of tape from the conclusion of his toast as dispositive of the harmless nature of the sexist jokes. Defenders consider the mere presence of laughter in response to his comments as a definitive indicator that his comments were interpreted by his largely female audience as harmless humor, and evidence of St. Louis's overreaction. At minimum, the quality and pervasiveness of the laughter simply cannot be judged by a snippet of audiotape. Such responses ignore the fact that not all laughter is the same, let alone sincere; people can laugh uncomfortably in social situations or in response to sexist or racist jokes. Regardless, research documents the ways in which sexist jokes can still cause harm (Reference SueSue 2010, 28, 175–76). The fact that St. Louis's view on Hunt's toast was widely corroborated by other lunch attendees—no fewer than twelve additional attendees confirmed its accuracy (Reference St. LouisSt. Louis 2015; Reference Waddell and HigginsWaddell and Huggins 2015)—and the fact that the organizing committee of the conference formally requested (and received) an apology from Hunt suggests that the joke wasn't received as harmless humor (European Women in Mathematics 2015). The persistent refusal to believe Richards's and St. Louis's ability to report and respond accurately undermines them as socially untrustworthy; not only do they not speak for a broader technical audience, they cannot even respond appropriately in social situations.Footnote ¹¹ These are not simply “social” criticisms but epistemic as well. They are being judged for being incapable of reading and responding to social cues; they are shamed for being not like others in their professional worlds (everybody else, the implicit response is, knows how to take a joke).

Part of the reason Richards and St. Louis are dismissed as socially untrustworthy is that they are redescribed as overly ideological: critics caricature them as angry women with agendas attacking harmless men who just want to do their jobs. The implicit contrast between harmless men who merely want to work and agenda‐driving angry feminists who want to turn everything into a controversy is itself, of course, another way of reasserting professional authority. The targets of feminist shaming are professional and responsible, in contrast to the newly diminished authority of the angry ideologues doing the shaming. This criticism is often paradoxical; Hunt is simultaneously described in terms both venerating and condescending. His Nobel prize and pathbreaking research are mentioned alongside his advanced age; he is described as “bespectacled” or literally “defenceless” (Reference AdamsAdams 2015; Reference Whipple and MoodyWhipple and Moody 2015).Footnote ¹² By contrast, St. Louis is described in aggressive and intentional terms; she is the “architect” of a “witch‐hunt” (Reference AdamsAdams 2015). Richards is similarly contrasted with “Hank”; his status as a father of three is perennially mentioned, thus rendering him a safe and benign family man who is maliciously and unjustifiably attacked. In both cases, professional innocence is reasserted against ideological one‐sidedness.

Finally, the fact that Richards and St. Louis are also women of color working in overwhelmingly white professions becomes a vulnerability employed against them in these backlashes. Women of color represent a mere 10% of those employed in STEM disciplines in the United States (Reference Falkenheim and BurelliFalkenheim and Burelli 2012). Of the already small number of STEM workers of color, comparatively few of them are African‐American (as are both Richards and St. Louis). In her blog post, Richards's photo of the two programmers behind her, with its backdrop of the overwhelmingly white and male crowd, acts as a visual reminder of the homogeneity of Silicon Valley. Comments on one article criticizing St. Louis explicitly or implicitly refer to her race (claiming she doesn't look British, doesn't look like an academic, and doesn't understand spoken English [Reference AdamsAdams 2015]).Footnote ¹³ St. Louis notes that she initially thought about not publicizing Hunt's comments in part because of her status as a woman of color; she notes that black women are even more likely to be viciously trolled on Twitter than Anglo women are. Much of the online harassment Richards experienced was racist in character (Reference Raja and LiebelsonRaja and Liebelson 2013). The explicitly racist nature of the backlash is a move of exclusion, reasserting the existence of STEM communities that are not spoken for or on behalf of by Richards and St. Louis.

Consider these three components of the shame backlashes as a totality. Both feminist shamers have their professional credibility challenged and minimized; both shamers are redefined as ideologues who do not have social credibility; both shamers are persistently and aggressively reminded of their outsider status in overwhelmingly homogeneous professions. All three of these effects work singly and collectively to tell these women that they are not, in fact, legitimate members of the communities on whose behalf they claim to speak. Their attempted shaming is thus illegitimate, because they do not have the credibility or competence to speak on behalf of these communities, and they are in fact shameful for trying to shame their communities. Both of these backlashes, which began with women of color calling larger communities to try to live up to their claimed ideals of nonsexism, function as community closures, a protective move to reassert narrow boundaries of authority and identity. Sara Ahmed describes a similar dynamic and danger in her recounting of national responses to narratives of injury. When the subaltern recount narratives of injury to a privileged majority, which should function to open up questions of identity and concepts of the whole, they sometimes find narratives of injury turned against them. Because white males have more narrative privilege and access to resources, they can use these very tactics to portray themselves as the injured party, and close off debate on national injury and identity (Reference AhmedAhmed 2017, 33). This is the tactic we see here, where Richards and St. Louis recount sexist injuries to larger communities that claim to be striving for inclusion, to find that the real victims are apparently the white men and the institutions charged with sexism. The undermining of Richards's and St. Louis's authority works in part to redefine the boundaries of their professional worlds; unruly “others” do not fit. In these cases, there are two additional consequences: those who risk speaking up and challenging bad behavior are themselves the focus of sustained shaming, and those who might wish to speak up in the future are implicitly cautioned against doing so, because of the visible scapegoating and disciplining of rogue feminists.

Disciplinary Damages

Nussbaum describes shame as a stigmatizing judgment; shame marks us (whether literally or figuratively) as less‐than or other in a community. How, precisely, were Richards and St. Louis stigmatized or damaged by the shame backlash each experienced? Most basically, both suffered lasting economic damage.Footnote ¹⁴ Richards was fired from her job for her role in the shaming of “Hank,” and the program St. Louis ran at City University was eliminated in the following year, as was her position (Reference HoltHolt 2013; Reference St. LouisSt. Louis 2016).Footnote ¹⁵ Richards and St. Louis were not merely being criticized for the content of their views or positions; their abilities to function as professionals were consistently and aggressively undermined. Thus the online criticism directly resulted in sustained economic damages for both women; their professional identities were in fact permanently (at this writing) harmed and limited due to their attempts at feminist shaming. The stigma of shaming that Nussbaum describes (being marked as less‐than) is financial and professional; they are judged as less capable and competent as STEM professionals.

For both women, the stigma extended well beyond the financial realm. Richards and St. Louis were both the recipients of explicitly violent and sexually violent threats online. St. Louis reported several messages to the police (Reference HariaHaria 2015), and Richards received multiple death and rape threats, including publication of her home address (Reference HoltHolt 2013). This is more than the risk women face for participating in political activism (Reference FergusonFerguson 2010, 251–52). These threats are not simply psychic or tactical, and they go beyond rejection, ridicule, offense, or even losing the fight (251–52). Both experienced a pattern of threats of violence, sometimes sexual in nature. Threats of violence and sexual violence label those threatened as vulnerable and violable, and the diffuse and anonymous character of the threats renders them constant. There is a gender disparity in these threats; this was not merely an idiosyncratic response to these two incidents. Women are disproportionately the victims of online sexual threats now categorized as “technology‐facilitated sexual violence” (Reference Henry and PowellHenry and Powell 2015, 759). And although it may be true that many of these threats are idle, or arise from varied social motives that have little to do with actually enacting violence (Reference NorlockNorlock 2017, 6), someone on the receiving end of a torrent of threatening messages has no way of making such a discernment. This would be particularly true for Richards in the aftermath of her home address being publicized. It is understandable in cases like this to retreat from public engagement after a flood not just of criticism but of threat; this is not a retreat from public engagement but from the possibility of personal harm. Indeed, the fact that neither Richards nor St. Louis currently has a public profile reinforces this.Footnote ¹⁶

The Role of Online Anonymity

I contend that the material damages Richards and St. Louis experienced occurred in part because of their attempts to “shame up”: to critique communities’ failures to adhere to stated norms (even if aspirational), and to challenge them to improve. As women of color within overwhelmingly white, male communities, Richards and St. Louis risked more than most whistleblowers in challenging outdated norms: they were visible and visibly other in their professions. Thus, there are specific risks in feminist shaming: when women, especially women of color, use shame as a tactic against an overwhelmingly male community, they risk isolating and publicizing themselves as outliers in ways that are more memorable and lasting than for those who enjoy demographic privilege (and thus invisibility). Jacquet suggests that shaming the powerful is more acceptable and effective than shaming the marginalized (Reference JacquetJacquet 2015, 174), and for good reasons: shaming‐down, particularly if informal, can make someone look like a bully, taking advantage of her power to further punish the powerless. However, her minimal consideration of audience elides the risks that visibly marginalized knowers take by shaming up. The absence of widespread condemnation over the shame backlashes Richards and St. Louis experienced suggests that those with less social and political capital cannot count on shaming as an effective tactic. Indeed, at least one commentator appears tacitly to endorse, or at least minimize, the response Richards received against that of “Hank.” Jon Ronson expresses deep sympathy for “Hank”'s (momentary) job loss, but none for Richards's job loss. His response to Richards's description of feeling personally endangered by regularly hearing sexist jokes in her presence is that “some people” might think her reaction is “overblown”; when she responds by saying that “some people” would probably be white men, he considers it ad hominem (Reference RonsonRonson 2015, 118).Footnote ¹⁷ Although he describes the sustained and violent public campaign of harassment Richards endured, he expresses no sympathy for her position, as opposed to his sympathy for “Hank”'s temporary joblessness (120).

The fact that these shamings occurred online rather than in person may exacerbate, rather than mitigate, the disciplinary power and likelihood of shame backlashes. The democratic nature of shaming that Etzioni praises can be dangerous when unshackled to real identities and lived communities. In contrast to Richards's and St. Louis's double visibility as women of color, many of their attackers are not only demographically anonymous, but literally anonymous, working under avatars and commenting on blogs.Footnote ¹⁸ To be sure, St. Louis came in for criticism by well‐known public figures (among them fellow science writer Richard Dawkins and former MP Louise Mensch), but most of those attacking Richards and St. Louis did so behind a veil of anonymity; their credibility and credentials were not publicized for the kind of discussion and critique they so eagerly extended to Richards and St. Louis. Indeed, online shaming is often a largely anonymous venture, and what Jacquet clearly imagines as a kind of online citizens’ brigade can be a dangerously unpredictable tool because of this anonymity. Kathryn Norlock suggests that one of the dangers of this anonymity is that it renders the object of the shaming almost beside the point; shamers, she contends, are in “imaginal relationship” with one another and thus are seeking recognition and approval from their fellow shamers (Reference NorlockNorlock 2017, 4). This relationship, Norlock suggests, makes it easier for shamers to see themselves as a disempowered group taking down the unjustly powerful; thus, they shame more viciously than justified, with no regard for the effect of their words or actions (6–7). Shaming here is a mob activity; the shamers talk to each other, and credential and reinforce their judgments and their outrage. The target of their shame is just the occasion. Norlock's concern that this decontextualized and careless shaming is damaging is well framed and may well apply to the hordes of commenters issuing violent threats to Richards and St. Louis.

The power disparity in these two instances is only enhanced by the visibility and identifiability of Richards and St. Louis within their communities. The objects of feminist shaming, as white men, can safely be reclaimed by their white male crowds, in ways that feminist shamers often cannot. We see precisely this result in the Richards/“Hank” episode. When Ronson asks “Hank” in what ways being shamed has changed him, “Hank” asserts that he now avoids female programmers at work (Reference RonsonRonson 2015, 130). But in overwhelmingly male Silicon Valley, this is a hollow vow; indeed, Ronson confirms in a follow‐up question that there are no female programmers at “Hank”'s new company for him to avoid. Thus, “Hank” gets to claim the privilege of withdrawal and retreat into the homogeneous crowd, at no apparent social or professional cost to him whatsoever. Ronson's failure to comment on the irony of “Hank” claiming to avoid women when there are no women for him to avoid stands as an inadvertent endorsement of “Hank”'s sexist conclusion. Ronson's implicit position is that Richards's shaming accomplishes nothing to address systemic sexism; rather, she has driven an otherwise harmless man into woman avoidance, which can only exacerbate rather than ameliorate sexism. Only someone who has never been publicly identified, and whose visual markers are hard to differentiate from the vast majority of his Silicon Valley colleagues, would have the privilege of withdrawal. By contrast, Richards appears to have been permanently ejected from this community.Footnote ¹⁹

The Insufficiency of “Constructive Shaming”

Feminists who are still interested in shaming as a political tactic might be tempted to employ Nussbaum's “constructive shaming,” a less aggressive and less individually targeted shaming.Footnote ²⁰ Constructive shaming is pedagogical in orientation, and diffuse in direction; it focuses on all members of the community, who are invited to examine their vulnerabilities and interdependence (Reference NussbaumNussbaum 2004, 213). Nussbaum cites Barbara Ehrenreich's biting commentary in Nickel and Dimed (Reference EhrenreichEhrenreich 2001) about economic privilege at the expense of working‐class poverty as a paradigmatic example of constructive shaming that invites all Americans to examine the costs of middle‐class comfort (Reference NussbaumNussbaum 2004, 241).Footnote ²¹ But this tactic looks less promising upon reconsideration. First, the absence of major reform of economic inequality in the intervening seventeen years undercuts Nussbaum's confidence, and may indicate the weakness of such generic constructive shame appeals. Shaming is specific in its direction: it attacks particular reputations, whether individual or collective, as unearned or inconsistently practiced. The identity of Americanness, or middle‐classness, may be too diffuse and general for it to have any meaningful claim on ethical attention.Footnote ²² Second, shaming is pointed: it aims to reduce the social status, whether formally or informally, of the object of its shame. This gives shaming concrete bite, as well as presenting risks to its users.Footnote ²³ Although constructive shaming avoids the risk by being framed as pedagogical rather than punitive, it also eliminates any sense of formal or informal consequence or urgency to the examination. It invites anyone who receives the message to contemplate their own failures to adhere to ideals, but presents no consequence for failing to take up this invitation.

There is concrete evidence in light of these feminist shamings to undercut the effectiveness of constructive shaming as politically effective. The #distractinglysexy campaign some women scientists launched in response to Tim Hunt's sexism both fits the parameters of constructive shaming and simultaneously suggests the political limits of the approach. After some women scientists created the hashtag, they began posting parodic captioned pictures of themselves as working scientists. It quickly generated thousands of contributions and virtually universal praise. Most photos featured visual parodies of Hunt's comments: pictures of women vamping while holding test tubes, captions sarcastically praising themselves for managing to stave off sobs as they examine slides of tissue under microscopes or excavate archaeological sites, signs labeling “mixed gender labs” in which weeping and falling in love are prohibited. Taken as a whole, the photos function as a direct rebuttal to Hunt's glib characterization of “girl” scientists who cry in the face of criticism and function mostly as perennial love objects. But collectively, the photos also demonstrate the variety and range of the science women do across the disciplines. By accumulating an immediate, online, and ironic catalog of the array and diversity of women's contributions to science, the photos stand as an assertion of women's collective legitimacy as scientists. The campaign functioned as much as a roll call of women in science (we are here, doing legitimate work) than simply as a criticism of sexism in science. It is a pedagogical assertion of credibility as much as, if not more than, a condemnation of institutional sexism, and thus a paradigmatic example of Nussbaum's constructive shaming.

No doubt part of the reason that the campaign was well‐received was its collective and anonymous nature. No single woman was identified as the campaign's leader; no single woman's credibility could be the focus of criticism and undermining; the rapid accumulation of photos from many parts of the globe gave the campaign widespread credibility.Footnote ²⁴ In certain respects, the women got to enjoy anonymity while expressing hilarious solidarity with St. Louis. But it is the tenor, as much as the volume, of the campaign that makes it relevant. The women were only partially responding to and about Tim Hunt and his retrograde comments. Rather, they were really talking about themselves and their work, and (no doubt tediously) reiterating to the world that their work had independent value, and was in fact the opposite of sexy. What the women scientists did, in effect, was to respond, as if conversationally, to awful conversation. When someone tells a sexist joke, feminist listeners can speak up and refute their sexism, seethe silently, or walk away.Footnote ²⁵ All three tactics bring with them risks and costs, and both Richards and St. Louis chose the most direct tactic (speaking up) that potentially brings with it the most direct costs. But as Richards's and St. Louis's experiences demonstrate, to speak up and identify sexism as such, particularly when there is a status gap between the accuser and accused (as with St. Louis to Hunt), or when a demographically marginalized person criticizes a more homogeneous group (in both situations), one risks not simply being drawn into a debate about whether or not the comments are sexist, but the possibility of more serious and long‐lasting stigma. Both Richards and St. Louis were criticized for not choosing one of the other options (ignoring the joke, or just walking away), but neither of those would have been clear as political responses to sexism. Both jokes occurred at conferences, which are large and impersonal; attendees often leave and enter the room mid‐meal and mid‐speaker. Richards overheard a joke from audience members, so walking away would not even have registered as a response at all to anybody else in the room. By contrast, although the women of #distractinglysexy chose speaking up as their response, their speaking up is more like how one responds to a sexist joke at a party than a direct repudiation of a public statement. The women simply mocked the assumptions behind the joke; they made fun of the idea that this joke of weepy women scientists would still be told.Footnote ²⁶ The fact that #distractinglysexy, while legitimately hilarious, received such a uniformly positive response simultaneously with St. Louis's very public repudiation and stigmatization speaks not simply to the power of the voice of the crowd against the lone critic, but more worrisomely, to the gendering of acceptable feminist responses to sexism. Direct challenges of public sexism are still safer when expressed in a genial or harmless manner. Direct expressions of feminist anger are apparently still too subversive, to say nothing of feminist demands for political or institutional responses to sexism. Direct expressions of feminist shaming risk women being shamed back to their places.

The Limits of Shaming Up

To recapitulate, shaming, as a stigmatizing tactic relying on an assertion both of status and credibility, requires that its users be able to depend upon a receptive audience for their shaming. But when audiences are not sympathetic to the judgment of shame, there is potential for a shame backlash: not simply a rejection of a shaming judgment but a redirection of shame back onto the would‐be shamer. Despite attempts to defend shaming as a political tool for the marginalized, these two instances of shame backlash demonstrate the ways in which more marginalized members of professional communities cannot count on their epistemic and social credibility being recognized, and more worrisomely, may expose them to a shame backlash as an aggressive response. Does this recounting of online shame backlashes mean, then, that women should never use shame as a feminist tool, particularly when employed against a more powerful or homogeneous group? No, but it suggests the need for intentional strategy around shaming. At minimum, would‐be shamers should be aware of the risks of this strategy, and prepared for the possibility of a backlash. More productively, shamers may want to work in tandem where possible, and recruit allies to speak publicly and simultaneously. Although several allies spoke up to corroborate St. Louis's reporting of events (Reference BlumBlum 2015), the supporting statements came later, and thus could be ignored in the media narrative. Choosing a single spokesperson for a collectively experienced event meant that all the backlash ire could be narrowly focused.Footnote ²⁷ Their delay (even if only brief) in speaking up to support St. Louis meant that she became the single, and singly vulnerable, face of feminist shaming.

Aspiring feminist shamers should also think about potential responses in the event of a shame backlash. Whether shame is formal or informal, it functions psychologically as much as socially; it persuades us that we truly are undeserving or unworthy of attention. Shame backlashes, because they function to narrow the boundaries of who is worthy of an audience, can be productively challenged as hypocritical or discriminatory; their tactics can be publicly named and demystified. A version of this tactic was employed by LGBT activist groups like ACT/UP and Queer Nation in the 1980s and 1990s. In response not just to discriminatory policies (such as a refusal by the US government to acknowledge the existence of AIDS as a disease or to fund its research and treatment) but to the persistent homophobic beliefs that justified and perpetuated this policy discrimination, these groups chose deliberately outrageous public protests—“die‐ins” or disruptions of Catholic masses—as a way of deliberately rejecting and undermining the shaming of their lives that resulted in systematic discrimination. They refused to accept social shaming, and made visceral, dramatic claims that majority silencing was in fact the shameful act (Reference MuntMunt 2007, 55–77). Top‐down shaming shames rebels into their places by a threat of stigma, and bottom‐up shaming threatens to reveal institutional hypocrisy. Organizations that claim to work for the greater presence and flourishing of women in their male‐dominated professions, and then stand by and accept persistent, personal, and cruel shaming of those who speak up to ask for them to live their values, deserve to have that hypocrisy publicized. Shame backlashes are simply a second layer of hypocrisy; a response to a complaint about bad behavior that reifies and normalizes the bad behavior. Demystifying the hypocrisy behind this punitive shaming response publicizes the sort of disproportionate response experienced by the powerful when unjust uses of their power are exposed. Although no failsafe, it can serve to minimize the damage of a backlash.