Critical injustice in academic philosophy

  1. Critical injustice

In his recent work, Allan Hazlett argues that “critical injustice” is a type of epistemic injustice, defined by “an unfair deficit of criticism” (in progress). Critical injustice perpetrates ‘primary’ and ‘secondary’ harms, including disrespecting or impugning the dialectical competence of the target, silencing the person’s criticism, and failing to create opportunities for the person to engage in criticism and develop their dialectical competence. Further, critical injustice targets people on the basis of their demographic attributes or social identity, consistent with Fricker’s (2007) definition of epistemic injustice as a type of ‘identity prejudice.’  Interestingly, both liberals and conservatives can commit critical injustice – liberals because they may go ‘too easy’ on visible minorities, failing to respect their dialectical competence, and conservatives because they may go ‘too hard’ on them, denying them the credibility they deserve. An epistemically virtuous critic gives the right amount of criticism, neither too much nor too little.

This position connects with my earlier posts on the value of criticism. Feminists have argued that the underrepresentation of women in the New York Times Book Reviews – which in 2016 published only 1/3 as many reviews of women’s literary fiction as men’s – signals a lack of respect for women’s work and undermines their standing in the profession. The gender gap in literary criticism can be seen as a kind of critical injustice.

Hazlett, however, is more concerned with critical injustice in educational settings, in which teachers must decide how to critically engage with students, some of whom are from underrepresented groups and are therefore susceptible to identity prejudice (including implicit bias, which is not directly introspectable). How should teachers engage with these students? The straightforward answer is that they should give them the right amount of criticism, taking into account the possible role of implicit bias in their unconscious cognitive architecture, and the possibility that students from underrepresented groups may have non-transparent epistemic privileges rooted in their lived experience. These epistemic complexities make critical justice harder than it may seem at first glance. Still, the goal is to express the right amount of criticism to each student.

I wouldn’t deny that quantity of criticism is a factor in critical justice, but I think that quality of criticism may be just as important. What Hazlett doesn’t explicitly address is, once you’ve epistemically evaluated a speech context and determined that criticism is warranted, how should you express that criticism? Most of the examples of epistemic injustice offered by Miranda Fricker (2007) involve someone expressing criticism in a demeaning way. For example, in The Talented Mr. Ripley, when Marge Sherley correctly discerns that Tom Ripley is the murderer, Herbert Greenleaf dismissively retorts, “Marge, there’s female intuition and then there are facts” (Fricker 2007: 9). The problem isn’t so much that Greenleaf criticizes Marge – if he had ignored her suspicion, this, too, would have been critically unjust. Both arrogant silence and dismissive sexism are instances of critical injustice, because both express contempt for the speaker’s epistemic standing as a woman. More specifically, both responses have the same “ethically noxious” content (Fricker 2007: 34) – implicit gender bias. This suggests that simply increasing the amount of criticism directed to minorities won’t necessarily reduce critical injustice – it might just open the floodgates to a deluge of epistemically corrupt criticism.

Similarly, if teachers take steps to fairly distribute the quantity of criticism in their classroom, this would be no guarantee that critical injustice would be reduced, because the content of the newly expressed criticism could be epistemically problematic. That is, the criticism expressed to minority students, albeit in the correct ratio, could contain epistemically corrupt content, such as condescending superiority, dismissive contempt, or hostility activated by implicit bias. Indeed, this is perhaps the predictable effect of increasing criticism to minority students (without taking any other measures), especially if the initial lack of critical engagement was caused by implicit bias. Simply increasing criticism toward minority students could allow the problematic content of the initial prejudicial disengagement to take a more overt form – namely, prejudicial criticism. This isn’t to say that we shouldn’t try to distribute criticism fairly – it’s simply to say that, in doing so, we need to think about the motives that underlie both prejudicial critical disengagement, and prejudicial critical engagement. These are two sides of the same coin.

To bring this into relief, consider a comment on the philosophy blog Daily Nous, cited in a paper on the diversification of philosophy by Luvell Anderson and Verena Erlenbusch:

“I am very ignorant of non-Western philosophical traditions and perhaps this post will only expose my ignorance. Perhaps someone who knows non-Western traditions better will comment on my post. However, I am under the impression that the best arguments are generally to be found in Western philosophy. When I last read Confucius, just to take an example, I don’t recall finding any arguments at all. That doesn’t make Confucius unworthy of study, but it does make him a lot less philosophically (as opposed to anthropologically) interesting to me” (Weinberg 2016, cited in Anderson & Erlenbusch 2017).


This is an example of someone who criticizes non-Western philosophical traditions (and, indirectly, people who specialize in these sub-disciplines), but only to dismiss them as epistemically inferior. This illustrates the point that suspending criticism toward someone whose standpoint you don’t respect may be better than critically engaging with the person, given your motivational architecture. (If you don’t have anything nice to say, don’t say anything at all!) The lesson for critical injustice is that robust critical justice probably requires remediation of implicit biases, not just a fairer distribution of criticism.

 In sum, we should try to criticize others both in the right proportion, and in the right spirit.

2. Hermeneutical injustice: Background conditions for testimonial injustice

This task is easier said than done because, even if a teacher has the intention to criticize fairly, the context might not allow the person to do so. Indeed, many teachers don’t have the contextual epistemic resources (or ‘hermeneutical resources’) required to criticize epistemically marginalized students fairly, precisely because their epistemological community deprives them of those resources. In psychology, for example, most Western psychology research uses disproportionally WEIRD (western, educated, industrialized, rich, democratic) subjects (up to 80% of samples) although these subjects comprise only 20% of the global population (APA 2010). Until recently, findings based on WEIRD samples were generalized as universal psychological truths – a norm that marginalizes non-WEIRD frames of references. On the standard psychological model, traditional societies are seen as outliers with ‘abnormal’ psychological traits. This is a type of epistemic injustice insofar as non-western frames of reference are construed as less credible.

One of the worst offenders is arguably my discipline – philosophy – which is one of the whitest and malest of all Humanities disciplines, even though everyone apparently truly values diversity. Notwithstanding everyone’s good intentions, most academic philosophers are white, male, and middle-class (Cruz forthcoming). Registered members of the American Philosophical Association (APA) are 76.4% white and 74.6% male (Trott 2018). Of published authors in prominent journals, POC comprise only 0.5% (Cherry & Schwitzgebel 2013), and women comprise only 14-16% (Wilhelm et al. 2017). Historically, virtually all philosophers were white and male, and these philosophers are still overrepresented on college syllabi. Interestingly, representation varies by sub-discipline;  60% of women in philosophy have a research interest with a value theory keyword (which includes feminist philosophy), but only 40% have a research interest with a LEMMING keyword (Jennings, Hypatia Conference presentation: 2015); and women specialize in philosophy at seven times the rate of men according to placement data (ibid.). For Black philosophers, the top areas of specialization are Africana, race, social and political philosophy, ethics, and continental philosophy (Botts et al. 2014); yet ‘specialist’ journals (e.g., in feminist philosophy, Africana) are considered less reputable than ‘generalist’ journals, receive less attention as measured by citations, and are un-ranked on ranking websites like Scimago (2018). As Helen de Cruz points out, top-ranked philosophy “journals avow themselves generalist, but in practice they tend to publish a narrow range of specializations, with a heavy focus on analytic philosophy of language, epistemology, metaphysics, and philosophy of mind, the so-called ‘Lemming’ subdisciplines. This leaves many areas of philosophy underrepresented in these journals” (forthcoming), including feminist and Africana theory, which are also underrepresented on college syllabi (Baron et al. 2015; Figdor & Drabek 2015; Shepherd 2014).

There are a multitude of explanations for the diversity problem in philosophy, including the course content hypothesis, the role model hypothesis, the aggressive argumentation hypothesis, the gender schema hypothesis, the field-specific ability belief hypothesis (viz., Baron et al. 2015), the perfect storm hypothesis (Antony 2012), and the prestige bias hypothesis (Cruz forthcoming). Regardless of the cause of the diversity problem, there’s a case to be made that the diversity problem constitutes a type epistemic injustice because it is a situation in which underrepresented groups are epistemically marginalized, not because they lack philosophical qualities, but because of epistemically corrupt features of the community. This is consistent with Louise Antony’s explanation, on which the underrepresentation of women is caused by a “perfect storm” of implicit bias, stereotype threat, disciplinary conventions, and schematic assumptions about women’s capacities (stereotypically irrationally) and philosophical values (rationality, critical thinking) (2012). These factors interact in such a way as to produce, reinforce, and intensify epistemic injustice against women (and, by parity of reasoning, racialized minorities) in the field.

This explanation is closely related to an empirically-supported hypothesis – the gender schema hypothesis. On this hypothesis, women are underrepresented because of internalized schemas (Valian 1998), “which directly code philosophy as a male discipline (Haslanger 2008; Calhoun 2009),[3] or indirectly code it as male, e.g., through the combination of a ‘field-specific ability belief’ that philosophy requires natural brilliance and a societal stereotype of women as lacking this brilliance (Leslie, Cimpian, Meyer, & Freeland 2015)” (Baron et al. 2015). Because philosophy is coded as male, women are less likely to choose philosophy as a major. Notably, Baron and colleagues find that pre-university effects explain women’s failure to continue in philosophy, as opposed to classroom effects, the reason being that men’s and women’s attitudes toward philosophy are not differentially affected by their classroom experience (in one semester). However, this doesn’t mean that gender schemas don’t exist in academic philosophy; gender bias might be operating in academic settings in such a way as to confirm students’ pre-existing gender schemas. Classroom experiences certainly don’t disconfirm these preconceived notions according to the evidence, which shows that gender differences in intention to major, perception of ability in philosophy, ability to imagine becoming a philosopher, and comfort in philosophy class are relatively stable across the semester, while interest in philosophy decreases in women but not men (Baron et al 2015). If philosophy disconfirmed pre-existing schemas, we would expect women’s attitudes to change. Perhaps gender schemas in philosophy are equivalent to extra-philosophical gender schemas (which are sexist!); perhaps in-classroom factors have small, cumulative effects over time, producing “micro-inequalities” that gradually discourage women from majoring in philosophy (Brennan 2013); perhaps gender schemas in philosophy are stronger or more salient than in other disciplines – it’s impossible to know for sure because the study was limited to 13 weeks and there was no control group. But what we do know is that gender schemas that exist outside of academic philosophy persist in undergraduate philosophy classes, as opposed to being extinguished by classroom experience.

Antony proposes the ‘prefect storm’ hypothesis as an alternative to the ‘different voice’ hypothesis (Code 1987, Scheman 1987, Moulton 1993), on which women are underrepresented in philosophy because their gender-specific standpoint is marginalized by the dominant male frame of reference. Antony is right to reject some different-voice theories – for example, the theory that women have gender-specific philosophical intuitions, a hypothesis that failed in replication (Sayedsayamdost 2014). However, I think that there is evidence for a certain type of different-voice theory, which goes like this. Seeing that women specialize and publish disproportionately in value theory (including feminist philosophy), and Black philosophers specialize disproportionately in Africana and race theory, there is a dominant sub-disciplinary focus specific to these demographic groups, and this sub-disciplinary focus is less publishable in top-tier generalist journals, less cited and discussed, and less visible on college syllabi – that is, less epistemically respected. Because women and Black philosophers disproportionately specialize in these sub-disciplines (for whatever reason – that cause is irrelevant), their speech in academia (comments, analyses, critical insights) is epistemically marginalized. And their speech is marginalized not because these sub-disciplines are inherently less credible, but, again, because of internalized gender schemas.

This situation can easily result in critical injustice. Let’s say you’re a woman presenting a paper on trans feminism at a generalist philosophy conference, and the participants reflect the APA registration demographics – they’re mostly men. These philosophers, suppose, have male-dominant specializations – LEMMING subjects. They earnestly want to give your talk the epistemic respect it deserves, so they listen attentively and in the spirit of epistemic humility. At the end of the talk, having absolutely no background in trans feminism, the audience members either don’t ask any questions, or they ask naive questions reflecting their total ignorance of your area of specialization. In this (not-too-hard-to-imagine) scenario, the philosophical audience wants to treat you with critical justice, but can’t. Meanwhile, you have the standard basic knowledge of LEMMING subjects conferred by most American departments, so you’re in a position to ask them interesting and engaging questions, but they can’t reciprocate your level of critical engagement – not because they don’t want to, but because they don’t have the epistemic means.

Now imagine that you’re a college professor at a top-tier department and you have a standard philosophical background, meaning that you know more about LEMMING subjects than non-standard sub-disciplines. A student asks you a question about consciousness and you describe higher-order, representationalist, narrative interpretive, and cognitive theories of consciousness, and then field questions from the class. Then another student asks you about Eastern heart-mind and things-event theories of consciousness, and you make a few naive comments before admitting that you’re not conversant in Eastern philosophy, sorry. Then you go back to Wester theories of consciousness. You want to fairly critically engage with your student… but how can you?

This is just to say that good intentions won’t fix critical injustice in philosophy, or any other discipline. The hermeneutical climate has to be one that supports critical justice. A corrupt hermeneutical climate virtually guarantees the commission of testimonial injustice, particularly by privileged individuals (like professors) against disadvantaged individuals (like minority students), regardless of the privileged person’s intentions and beliefs.

Miranda Fricker addresses specifically disciplinary types of epistemic injustice, which she describes as “a very weak sense of injustice” (2007: 22). She gives the example of a philosopher who doubts a colleagues’ authority on her project because she (falsely) assumes that the colleague doesn’t share her specialization – ethics. So maybe the disciplinary imbalances I’m describing are merely instances of these ‘weak’ epistemic injustices, not ‘strong’ epistemic injustice, which is based on identity prejudice per se. I would argue, on the contrary, that sub-disciplinary biases in philosophy are rooted in strong identity prejudice, for several reasons. First, there is good reason to think that feminist philosophy, Africana, and other non-standard sub-disciplines are less respected, cited, discussed, and taught (basically, marginalized) because they are favoured by minorities, and minorities are less respected in the profession. That is, our epistemic practices as a profession reflect our collective implicit gender biases and implicit racial biases (as per the ‘perfect storm’ theory). There’s no reason to assume that LEMMING subjects are inherently more epistemically respectable than other sub-disciplines – and, indeed, no diversity researcher ever says they are – but because men prefer them (and men historically respected them exclusively, seeing other frames of reference as inferior), they receive more uptake. A similar phenomenon affects the U.S. economy: predominantly-female jobs pay less on average than predominantly-male jobs (IWPR 2016), and Black and Hispanic Americans are overrepresented in low-paying service and sales jobs compared to white and Asian Americans (BLS 2015). When women migrate into male-dominant jobs, the average pay drops (Miller 2016). This suggests that male-dominant jobs pay more because men are more respected (and a similar logic probably explains low pay in racialized economic sectors). The same white-male bias may explain topic bias in philosophy: LEMMING subjects are more respected in philosophy because they’re male-dominant, and (correspondingly) associated with emblematic white masculinity. This isn’t to say that we explicitly value white-male preferences more than others, but the profession is structured around implicit biases that favour white-male preferences, and give these preferences (and associated norms, perspectives, and conceptual frameworks) pride of place in journals, discussions, and class materials. Because our preferences are constructed in a corrupt hermeneutical climate, our preferences are likely to reflect the tacit gender and racial schemas of the field, especially if we specialize in dominant sub-disciplines that reinforce these preconceived notions.

If the reason we epistemically marginalize female-dominant and Black-dominant sub-disciplines is implicit bias, then these marginalizations are arguably strong epistemic injustices, because they aren’t just biases against an individual’s subjective (a-contextual) disciplinary preferences – they’re biases against a person’s disciplinary preferences because those preferences are coded as ‘female,’ ‘Black,’ or ‘other,’ and thus as ‘inferior,’ according to our shared gender schemas. This results in critical injustice, insofar as our gendered and racialized topic biases prohibit us from fairly critically engaging with minorities in their dominant areas of specialization – and other sub-specializations, too. Minorities who specialize in dominant sub-disciplines are epistemically disadvantaged by the gender and racial coding of their sub-discipline. This points to an epistemic double-bind in which, as a minority in philosophy, you can either choose to be underrepresented in a sub-discipline that’s respected because it’s coded as white and male, or your can choose to be overrepresented in a sub-discipline that’s disvalued because it’s coded as ‘other.’ Either way, you’re facing structural epistemic disadvantages.

Philosophy’s hermeneutical background conditions, then, practically guarantee that critical injustice will take place, because philosophers are inculcated into a culture that values sub-dsciplines emblematic of white masculinity, and disproportionally preferred  by white men, above other subjects. Institutional changes are needed to create hospitable ecological conditions for the virtuous deployment of criticism in the field.

3. Prestige bias

Another compelling explanation of the diversity problem in philosophy is the prestige bias hypothesis, recently advanced by Helen de Cruz, who holds that prestige bias ramifies institutional inequalities (forthcoming).* De Cruz argues that prestige bias is “an unchallenged phenomenon in philosophy,” as well as a neglected factor in diversity discussions (3). She shows that “there is a structural lack of upward social mobility in hiring practices: someone from a prestigious school may end up in a lower-ranked institution, but the reverse is unusual,” and “this advantage of prestigious institutions also appears in graduate school admissions, citations, and research assessments” (4). Although pedigree is often taken as a heuristic for competency, there is no substantive evidence for this association: teaching excellence and publication record play little role in hiring decisions; candidates selected for tenure-track positions have a median of two publications – less than the number of publications by job candidates from unranked departments. In fact, “the lower the rank of the PhD granting institution, the more papers in peer-reviewed journals a candidate was likely to have” (De Cruz: 11). Cruz links the pedigree heuristic to the field-specific ability belief (FSAB) on which philosophy is “a special aptitude that can’t be taught” (ibid.) – the same attitude that Baron and colleagues describe as a type of internalized gender schema (because it stereotypes men as innately talented). Cruz points out that this attitude is more salient in philosophy than any other discipline (Leslie et al. 2015). Philosophers also believe (correctly) that this attitude is prevalent in their discipline, but that doesn’t mitigate their predilection for FSAB. In fact, philosophers tend to openly admit that they hire on the basis of prestige bias (though they would call it a ‘reliable heuristic’ as opposed to a bias). As one hiring committee member said without reservation, “How did we prune our field from 637 to 27? An important selection criterion was holding a Ph.D. from a good university” (De Cruz: 14).

Prestige bias, says Cruz, leads to the underrepresentation of ethnic minorities and people of low socioeconomic status in philosophy, and to the testimonial “smothering and silencing of philosophical work outside of the dominant traditions” (21). Asians outperform all other ethnic groups on SATS and all other measures of academic excellence, yet their admission to prestigious U.S. universities is 30% lower than that of white students (Espenshade, Chung, & Walling 2004). African American students are also underrepresented at prestigious universities because of the racial wealth gap, educational inequality, and lack of cultural capital, which harm their admission chances. As a result of these inequalities, white people – who are overrepresented at prestigious universities – benefit from pedigree bias. Because racial injustice gives white people educational advantages in elementary school, high school, and college, they’re more likely to get tenure-track jobs, even if they’re not the best candidates by objective measures.

De Cruz says that the prestige of top-tier journals, which publish mainly on LEMMING sub-disciplines, results in a testimonial “smothering” and “silencing” of non-standard philosophical positions, because these positions appear less often in prestigious journals (‘silencing’), and because people who specialize in non-LEMMING subjects submit only half as often to top-5 journals (‘smothering’) (24). When philosophers don’t meet their publishing goals, they may “try to compensate for this by changing their topics of work to fit more prestigious fields” (22) (more ‘smothering’). Women in philosophy are also epistemically marginalized because they earn less than men, are underrepresented at prestigious universities, and, as a result, “may have less access to institutional funding” (26).

De Cruz doesn’t connect pedigree bias to the gender/racial schema hypothesis, but I’ve already indicated how they might be related. If pedigree bias is connected with field-specific ability belief about men’s philosophical talents, then implicit bias is a factor in pedigree bias. This is actually the dominant explanation of FSAB, viz., that FSAB interacts with gender bias to produce overt discrimination against women. As Myer, Cinpian, and Leslie explain the phenomenon, FSAB, “in combination with the stereotypes against women’s intellectual abilities… lead a variety of individuals (parents, teachers, peers, etc.) to see women as somewhat unsuited for ‘brilliance-required’ domains” (2015). If FSAB were gender-neutral, it wouldn’t motivate people to favour men. Cinpiani and Leslie say that “the same logic extends to race: our country has a long history of portraying African Americans as intellectually inferior, which is particularly likely to affect their participation in a field that focuses so single-mindedly on the quality of one’s intellect” (2017: 63). One can’t ignore that FSAB has gender and racial dimensions. Thus, it has what Fricker calls “ethically noxious” content – biased attitudes toward minorities. Pedigree bias is typically justified on the basis that the selected candidate is the better choice, regardless of any supporting evidence, and often in spite of evidence to the contrary – in other words, on the basis of a discipline-specific ability belief about the candidate. This is a species of FSAB, which is activated or partly constituted by implicit bias.

These considerations imply that hiring decisions in philosophy tend to be instances of testimonial injustice – interactions that epistemically marginalize women and racialized minorities. Implicit biases are so robust that even people who know they are pervasive in the field still hold them (albeit under a false representation of them as reliable heuristics). These biases have a number of downstream epistemic consequences, such as epistemically smothering and silencing minorities who work in philosophy, who must choose between specializing in a marginalized field, or being marginalized in a valued one.

De Cruz offers a number of solutions for minimizing pedigree bias, including citing more philosophers, especially philosophers from underrepresented fields; citing more publications from less prestigious journals; and ranking job candidates on the basis of non-pedigree-related criteria. These strategies could help to establish the hermeneutical background that I have described as essential to the reduction of critical injustice in philosophy. We need to take steps at all levels of the profession (and, ideally, outside of the profession as well) to limit the role of epistemic injustice in philosophy. More than individual tokens of good will, we need institutional change to promote critical justice in the field.


*All subsequent citations of De Cruz refer to this forthcoming paper.


Why women’s blame means less than men’s in conditions of epistemic injustice

  1. Introduction

Women’s blame means less than men’s in conditions of epistemic injustice because women’s speech means less, and blame is a type of speech (McKenna 2011). ‘Epistemic injustice’ is a species of injustice that harms someone in her capacity as a knower. It functions by recruiting ‘identity prejudice,’ or cultural stereotypes and biases, to discredit the testimony of socially marginalized groups. Women, as a socially marginalized group, are vulnerable to epistemic injustice on the basis of their gender identity – that is, they are susceptible to gender-based epistemic injustice, or (what I shall call) “epistemic gender bias.” When I say that women’s blame ‘means’ less than men’s, I mean that it is less articulable, intelligible, and authoritative than men’s. Because women are affected by epistemic injustice, they can ‘do less’ with blame than men can (Austin 1962).

There are many sociological factors involved in the epistemic marginalization of women’s speech (and thus women’s blame). Rebecca Solnit addresses several intersecting causal vectors in her recent book, “The Mother of All Questions: Further Reports from the Feminist Revolutions” (2017). I use her work as the starting point for my analysis of epistemic gender bias and its effects on women’s blame. By women’s blame, I simply mean blame expressed, or expressible, by women. Women’s blame doesn’t necessarily have distinct (‘feminine’) features, though it might; ‘women’s blame’ encompasses any criticism, censure, resentment, disapprobation, and related negative responses that women express, or would express if the epistemic resources required for uptake were available. Some instances of women’s blame are unspeakable due to epistemic injustice, and some are speakable but ‘unhearable,’ or unheard, because they lack a receptive audience.

In this post, I will explain, following Solnit, how (i) mainstream pornography, (ii) canonic literary fiction, and (iii) mainstream humour, all function to silence and suppress women’s speech. I will argue that (i)-(iii) all produce, reproduce, and naturalize “the male gaze” (Laura Mulvey 1975), by depicting the world from a cisgender male, or masculine, perspective. (By ‘the male gaze’ I mean, more precisely, the masculine point of view, into which many birth-assigned males are acculturated; but the ‘male gaze’ as I understand it is associated with gender, not sex. ‘Male’ from now on should be taken to denote ‘cisgender male’ or ‘masculine.’ Not all males, or men, take the perspective of the male gaze).

2. Sexual objectification

One factor in epistemic gender bias is the sexual objectification of women, perpetrated largely by mainstream pornography, which is a man-made cultural artifact that depicts women through the ‘male gaze’ – that is, as objects of male sexual pleasure, not agents in their own right (Laura Mulvey 1995; Andrea Dworkin 1981; Catherine MacKinnon 1989). Solnit notes that, while pornography has become more diverse in recent years, mainstream pornography has only gotten more violent, suggesting that porn may have evolved into “a compensatory parallel universe where male privilege has been augmented and revenge on female power is incessantly exacted” (Solnit 2017).

There is evidence that mainstream pornography impairs recognition of sexual assault: for example, “Male and female college students who report recent porn use have been repeatedly found to be more likely than others to believe ‘rape myths,’ [such as] that only strangers commit sexual assault or that the victim ‘asked for it’ by drinking too much or wearing ‘slutty’ clothing or by going to a club alone” (Peggy Ornstein 2016, cited in Solnit 2017). It would be naive, however, to think that the male gaze only impairs women’s ability to rebuke rapists and exercise sexual autonomy; the sexual objectification of women is so pervasive that it impairs women’s ability to say anything with the same credibility as a man. That is, it impairs women’s general epistemic standing.

In 2016, the Guardian reported that “eight of its most attacked columnists were women, two were men of color, and the most attacked of all was feminist Jessica Valenti” (Solnit 2017). Valenti is a vocal critic of rape culture and toxic masculinity, and her blame, as expressed in her feminist critiques, is rejected, dismissed, and attacked by her detractors, some of whom have threatened to murder and rape her. Rape threats are a common misogynistic threat, and even if they are not sincere, these threats are part of a broader system of male aggression that does provoke “women to change their behaviour to minimize harassment and worse” (Paul Crider 2017). Men who deny that rape threats can be consequential simply lack the ability to empathize with women (ibid.). Valenti is a particularly public example of a woman whose blame is epistemically marginalized, but her experience should be seen as a symptom of a deeper problem, not an anomaly. Harassment helps to explain why, while the Guardian‘s publication rate has increased over the last 15 years, the gender gap has remained the same (Gardiner et al. 2016).

Mainstream pornography doesn’t just normalize misogynistic violence against women – it depicts women as stereotypically infantile, naive, inexperienced, subordinate, and complacent. This is evident in Hugh Hefner’s casting of a 10-year-old Brooke Shields, along with at least nine other minors, in Playboy Magazine – not to mention countless depictions of childlike models posed as “cheerleaders, students, babysitters and sorority girls,” for the sexual gratification of a heterosexual male audience (Michelle Smith 2017). The popularity of  “barely legal” and “teen” pornography over the last few decades also speaks to the infantilization of women through porn (Griffith 2017). Hefner, who was touted by many as a champion of freedom of speech, was in reality a sexist jerk who belittled and harassed his female employees, plied them with drugs, refused to use condoms or be tested for STDs, and “require[d] depressing group sex at regularly scheduled times” (Robinson 2017). Though he portrayed himself as a cheerful hedonist, by all appearances, “he didn’t even care about pleasure: he cared about the taming and conquest of women” (Robinson 2017). Playboy illustrates Foucault’s argument that sexual discourse doesn’t just break taboos; it creates an “authorized vocabulary” (1976: 17) that codifies how one can discuss, think about, and perform sex. Mainstream pornography didn’t simply liberate us from conservative sexual mores of the past: it also constructed male sexual desire in Hefner’s image, as demeaning, infantilizing, and hateful to women.

The objectification of women is also evidenced in the fetishization of Asian women (who are the most ‘desirable’ group according to white men), which can be traced back to racist cultural artifacts like ‘Madame Butterfly,’ which depicts ‘the ideal’ Asian woman as “pretty and not much bigger than a doll,” “dainty,” and “little”, and the Korean, Vietnamese, and Second World wars, during which American soldiers had sex with, and raped, Asian sex workers and sex slaves, many of whom were underage (Juliana Wang 2014; Essertier 2018). Some American men ‘had sex’ (by their lights, but in many cases it was actually rape) for the first time with an Asian sex worker or sex slave, and there is evidence that first sexual experiences strongly influence future mating preferences (Quintana et al. 2015) – in this case, producing a lifelong ‘Asian fetish.’ The male sexual preference for (stereotypically childish) Asian women thus represents a culturally constructed desire for a thoroughly objectified, infantilized, and possibly enslaved, woman.

How can you respect blame from a woman if you see her as a child, if not a sex slave? The intergenerational sexual objectification of women partly explains why women’s speech products – women’s literary fiction, humour, and conversation – are less valued than men’s. Since these produces contain blame, women’s blame is epistemically marginalized.

Before exploring these vectors of epistemic injustice, I should clarify that I am not saying that all men contribute to epistemic gender bias. Solnit identifies three types of men: (i) the “raging misogynists and haters,” who actively perpetrate epistemic gender bias, (ii) the allies, who support gender equality, and (iii) the well-meaning but ignorant men who discredit women by accident (2014). The third category includes, for example, mansplainers, who condescendingly explain complex concepts to women as if they were speaking to a child. This infantilizing attitude is another aspect of the male gaze: it is an objectifying stance that treats women as epistemic recipients as opposed to epistemic producers. Even if not all men contribute to epistemic gender bias, we still live in what Miranda Fricker (2007) calls a climate of ‘hermeneutical injustice,’ in which shared interpretive resources for accurately framing (and adequately responding to) women’s epistemic marginalization is lacking. The ‘hermeneutical gaps’ in our epistemic climate, caused in large part by the pervasive ‘pornographication’ of women, discredit all women’s speech, even if allies are hard at work refuting the slut-shaming, rape-threatening, infantilizing, manterrupting, mansplaining speech of other men.

A second necessarily qualification is that women are not discredited across every domain; women in general are perceived as more credible in traditionally female domains, and less credible in traditionally male domains. For example, male experts are rated as more credible in murder trials, whereas female experts are rated as more credible in custody hearings (Larson & Brodsky 2010; Swendon, Nash, & Roos 1984; Walters 1994;  Neal 2014; Mill 2018). This shows that gender bias is context-sensitive. That said, because stereotypically female roles are poorly respected (and poorly compensated), women’s perceived credibility in these roles does little to enhance their overall epistemic standing (or socioeconomic status). Women are seen as experts on ‘trivial’ subjects, and non-experts on ‘serious’ subjects. This creates a double-bind: should women gain expertise on ‘frivolous’ subjects, or face adversity in ‘serious’ discursive spaces?

3. Women’s literature

Solnit notes that Esquire‘s 2015 list of “80 Books Every Man Should Read” included only one book by a woman (Flannery O’Connor). This isn’t a rare case of literary sexism, but a symptom of a deeper problem. Amy Hungerford, an English Professor at John Hopkins, argues that we should read, cite, and promote fewer canonized authors because these authors are predominantly male. In a journal called Modernism/modernity, created specifically to diversify literary scholarship, “the top 11 (mostly male) authors cited as subjects [still] claim 41 percent of the articles. Most authors not already canonical appear only once or twice each, never achieving the critical mass of scholarship that motivates further study and writing within the context of scholarly careers, let alone further reading by the general public” (Hungerford 2016). Thus, canonized men still dominate modern literary criticism. Students at Yale recently protested a two-semester course requirement on canonic English literature, in which 100% of the assigned authors were male (Flood 2016). In a press statement, the university argued that it had a responsibility “to provide all students with a generous introduction to the abiding formal and thematic concerns of the English literary tradition,” no matter how white and male those concerns happen to be. The problem is not just that junior scholars are forced to cite and analyze predominantly white, cisgender, male writers if they want to be taken seriously, which might harm the career prospects of junior women. The problem is that immersing students in a swamp of dead-white-male perspectives is harmful to their emotional development.

Departments that prioritize the literary canon are churning out graduates who have been inculcated into a historical-white-male worldview, which is essentially misogynistic (and racist). It’s not just mainstream pornography that teaches men how to objectify women; it’s the classic literary canon. The male gaze was constructed by the literary canon long before Hugh Hefner and his fellow pornographers recruited those masculinist schemas, images, and ideals into the pornographic medium. The mix of pornography and literary fiction in Playboy‘s pages, which may prima facie seem paradoxical, is actually a perfect fit, insofar as both mediums strive to promote and naturalize the male gaze. On pornography, Sandra Bartky writes, “it powerfully reinforces male dominance and female subordination because, by linking these phenomena to our deepest sexual desires – desires defined by an ideologically tainted psychology as instinctual – it makes them appear natural” (2012: 48). What better way to naturalize misogyny than by pairing mainstream (male) literature with mainstream (male) pornography, intertwining the two in the male psyche? The convergence of male sexual appetites and male intellectual curiosity, which Plato would describe as fundamentally antagonistic, are merged harmoniously in the motivational psychology of the average Playboy reader. While anti-pornography feminists like MacKinnon and Dworkin see pornography as the original cause of rape culture (viz., Papadaki 2017), pornographers didn’t invent the male gaze – they simply up-cycled the misogynistic scripts of the male literary canon into a more modern and simplified visual format.

Solnit provides several examples of emotionally harmful canonic literature, from Ernest Hemingway to Normal Mailer to William Burroughs (2015). Other masculine literary heroes, like Jack Kerouac, are less toxic, but still harmful if read in large quantities and through a passively masculine (as opposed to critical) lens. Anyone who fails to empathize with the Latina farmworker from Kerouac’s ‘On the Road,’ who is abandoned by Kerouac’s protagonist (a “free-spirited maverick”) to raise their child on her own (no child support!), is reading the story through the lens of the male gaze – the gaze through which Kerouac, in fact, tells it. Sharing in the mirth of Charles Bukowski’s protagonist from ‘Factotum,’ (‘Chakowski,’ based on himself), while he rapes a house-bound woman on his mail route, reveals more than a smidgen of toxic masculinity. Empathizing with Lolita’s rapist HH, the protagonist of Nabokov’s iconic novel, is hardly any better. (Note that Nabokov didn’t intend, or anticipate, this popular, revisionary reading of his work). Submersing oneself in the male canon, with no counter-narratives to offset the male ego, can make these acts of objectification, erasure, and rape seem natural, if not romantic.


‘Lolita,’ in fact, has been advertised by publishers as a novel about a “teenage seductress” as opposed to its true subject matter: the sexual abuse of a 12-year-old girl by her step-father (Noelle Talmon 2018). The cover of my own issue from high school (pictured above) was labeled with a blurb from Vanity Fair describing the novel (about a pedophile who literally ruins his step-daughter’s life) as “the only convincing love story of our century.” The author of the blurb, Gregor von Rezzori, parsed ‘Lolita’ as hovering between “a delightfully frivolous story on the verge of pornography,” and “a literary masterpiece, the only convincing love story of our century” (Shapiro 2009). While Rezzori admitted that his reading didn’t necessarily coincide with Nabokov’s authorial intentions, he defended it as “one of the novel’s many dimensions” (ibid.). Nabokov’s wife Vera, meanwhile, lamented that nobody seemed to “notice the tender description of the child, her pathetic dependence on monstrous HH, and her heartrending courage all along…”; and Azar Nifisi, an Iranian author, empathizes with Lolita as “a double victim—not only her life but also her life story is taken from her” (in Solnit 2015). Nabokov himself requested a cover with “no girls,” perhaps only a “white jacket,” but subsequent publishers increasingly eroticized the eponymous rape victim, responding to a market that already traded in the hyper-sexualization of young girls (Siobhans Lyons 2015).

The worry about the sexualization of young girls and the infantilization of women isn’t so much that these cultural scripts make men into rapists (although Peggy Orenstein notes that there is a correlation between frequent porn use and rape proclivity [2016]), but that these scripts prevent men – indeed, everyone – from empathizing with, and respecting, women. Solnit remarks that “there’s a currently popular argument that books help us feel empathy, but if they do so they do it by helping us imagine that we are people we are not” (2015). Indeed, research has shown that reading literary fiction improves various empathic measures, such as role-taking ability, motivation for prosocial behaviour and altruism, and theory of mind (Koopman & Hakemulder 2015). Theorists posit a cluster of psychological explanations for these effects, including role-playing, or mental modeling of the protagonist’s narrative world, and perspective-taking, or mentally simulating the narrator’s thoughts and emotions (Koopman & Hakemulder 2015).

This is consistent with Solnit’s theory that reading literary fiction increases empathy by enticing the reader to imaginatively take the perspective of the protagonist. But, in spite of empathy’s positive connotations in the popular imagination, empathy isn’t always a good thing, and this, too, is empirically demonstrated. For example, research shows that empathizing with in-group members in a competitive environment increases “intergroup empathy bias: the tendency not only to empathize less with out-group relative to in-group members, but also feel pleasure in response to their pain (and pain in response to their pleasure)” (Cikara et al. 2014). In an environment in which pornography, classic literary fiction, and other cultural artifacts depict women as infantile, unknowable, ‘mysterious’ (Friedan 1963), and ‘other’ (Beauvoir 1949), one would expect intergroup empathy bias to be part of the cultural fabric, and to operate in such a way as to marginalize women. This is because, while women are part of the group, they are ‘othered’ by cultural stereotypes, giving rise to within-culture out-group bias against them – that is, masculinist empathy bias. Even though men and women share common interests, the male gaze depicts women as devious, threatening, and – as male “dating gurus” see it – intent on withholding sexual resources from men, which is allegedly part of a ‘battle of the sexes’ (Berit Brogaard 2015: 9). This competitive framework forces one group into a position of subordination, and precludes any fair social contract. Masculinist empathy bias may be part of the explanation for why violence against women is a staple of mainstream pornography: it is both the cause, and the effect, of excessive empathy with the male gaze.

Solnit correctly observes that empathy is not enhanced in a robust way when we only encounter narratives that present “versions of our self rendered awesome and eternally justified and always right, living in a world in which other people only exist to help reinforce our magnificence, though those kinds of books and comic books and movies exist in abundance and cater to the male imagination” (2015). By fetishizing the male canon, we imaginatively project ourselves into a historical male perspective – a perspective that feminist sociologists have identified as misogynistic. It’s one thing to read about misogyny in a history book or a newspaper, and quite another to experience it through the simulated perspective of the perpetrator in a piece of literary fiction. Research confirms that reading literary fiction enhances empathy more than reading expository non-fiction or popular fiction (Kidd & Castano 2013). While theorists have had difficulty explaining the empathy gap between literary fiction and popular fiction, one possible explanation is that masculinist empathy bias makes it easier for us to empathize with the male gaze as presented in male-dominant literary fiction, and harder to empathy with the protagonists of popular fiction, a genre dominated by women.

In sum, reading literary fiction might be a healthy exercise if you read the works of many different authors, but if you read mainly books by men (or any privileged social group for that matter), or you read books by many authors, but disproportionally respect those written by men, this can impair your capacity for ‘robust empathy,’ i.e., the capacity to empathize with a diversity of perspectives. Intergroup empathy bias implies that empathizing too much with one perspective can make it harder to empathize with many.

Sonit’s work illuminates how the dominance of the male gaze silences women’s voices. By extension, the male gaze also silence’s women’s blame, insofar as women’s blame is expressed in women’s voices. The silencing of women’s blame can take several forms. For one, the loss or epistemic marginalization of women’s literary fiction in which blame plays a role entails an erasure of women’s blame; women’s blaming speech is literally lost, ignored, or misunderstood. Second, the loss or epistemic marginalization of women’s literary fiction makes women’s blame less intelligible in every context – not just literary fiction – because it creates hermeneutical gaps that discredit women as blamers, or as people who know things about moral norms, blame, and blame-relevant concepts. Third, the loss and epistemic marginalization of women’s blame is the flip-side of a cultural apparatus that exalts, promotes, and normalizes empathy with men. Men’s blame is therefore more likely to be taken seriously. This might allow men to ‘manblame,’ or blame with more perceived credibility than women. Stiegg Larsson (2005) sold millions of copies of a story about a rape survivor who seeks to punish her rapist, and though this is a refreshing departure from the classic male gaze, the fact that he outsold many women writing on the same subject, and that he arguably still replicated several masculinist ideals – such as the idea that “in order for a woman to be considered beautiful, she must be unhealthy” (Serdar 2014: 14; Robles 2015) – suggests that his critique of rape culture is not completely epistemically virtuous. ‘Lolita’ was not only written by a man, but re-interpreted by critics as a grand romance. Overall, women’s blame has been subjected to the same epistemic marginalizations as women’s writing in general.

An important caveat is in order here. The male gaze is currently losing traction because of a collective effort to diversity the literary canon, and all academic canons. Still, more men’s books are reviewed than women’s. Although women publish and read more than men, they are less cited and less respected as ‘serious authors.’ Research shows that “women are less likely to be published in top tier literary outlets, or to have their work reviewedespecially by men”; additionally, “women are less likely to receive reviews when writing about topics that aren’t deemed ‘feminine'” (Jane C. Hu 2017). Author Jennifer Weiner characterizes the credibility gap as “a very old and deep-seated double standard that holds that, when a man writes about family and feelings, it’s literature with a capital L, but when a woman considers the same topics, it’s romance, or a beach book—in short, it’s something unworthy of a serious critic’s attention” (2010). In the New York Times Book Review in 2016, two thirds of reviewed authors were men, and reviews tended to reflect gender stereotypes, revealing “topic bias” (Hu 2017). This means that women are taken less serious when writing about ‘serious subjects,’ and taken more seriously when writing about ‘frivolous subjects,’ resurrecting the classic double-bind: should I write frivolously about serious subjects, or write seriously about frivolous subjects?

Pornography is also becoming much more diverse. “Porn for women” surged in popularity by 1400% between 2015 and 2016 (Pearson 2018). This trend, however, doesn’t erase decades of misogynistic pornography, which has permeated mainstream culture, from television to magazines to advertisements. ‘Teen’ was still the second-most popular category on in 2016 (PH 2017). The most popular category was ‘lesbian,’ and explanations include that more women are watching porn and that men are more interested in porn for women. But neuroscientist Ogi Ogas offers a darker explanation: “To the extent that lesbian erotica is popular, it can be explained by the fact that men are most aroused by visual cues that emphasize youth and downplay drama and emotional complexity. Lesbian porn, therefore, works for straight men by ‘doubling up’ those visual stimuli… The only thing better than one nubile, personality-free woman is two of them” (Khazan 2018; viz., Ogas 2012). Ogas adds, “very few men visit websites containing erotica featuring actual lesbians that is targeted at actual lesbians” (ibid.), implying that the lesbian pornography that men watch still has elements of the male gaze.

It is worth mentioning that philosophy, my discipline, shares with literary fiction the dubious distinction of having a gender gap. Schwitzgebel and Jennings have found that “gender disparity remains large in mainstream Anglophone philosophy,” and “by most measures, women’s involvement and visibility in mainstream Anglophone philosophy has increased only slowly [since the 1970s]; and by some measures there has been virtually no gain since the 1990s” (2016). This means that academic philosophy is not immune from epistemic gender bias. Women’s blame (including criticism, censure, disapprobation) in philosophy, then, is subject to epistemic marginalization. Feminist blame – produced mostly by women – might be especially susceptible to epistemic marginalization, as it may be harder to publish feminist content in generalist journals, and ‘disciplinary’ journals are ranked fairly low on ranking lists (Jender 2007; Haslanger 2008).

Another aspect of women’s speech that is subject to epistemic marginalization is women’s humour, and women’s humour, like women’s literary fiction, contains blame. Thus, if women’s humour is silenced, then women’s blame, as conveyed in women’s humour, is silenced.

4. Women’s humour

Solnit notes that it is only recently that people have started to see rape jokes as not very funny, and this coincides with growing recognition of ‘rapist jokes’ (about rapists) as an intelligible type of humour. In 2013, comedian Sam Morril ‘joked’ that, “my ex-girlfriend never made me wear a condom. That’s huge. She was on the pill.” [Pause]. “Ambien.” (in Solnit 2015). This is just one example of a dominant theme in Western humour: making fun of rape survivors. (Surely we’ve all heard the rape joke about prison inmates being careful not to ‘drop the soap,’ which trivializes the human rights violations perpetrated in, and by, the U.S. ‘justice system’). Lately, many feminist comedians have turned the tables by making fun of rapists. Amy Pohler and Tina Fey made fun of Bill Cosby at the Golden Globes in 2015, and Amy Schumer wrote a parody of ‘Friday Night Lights’ that featured a football coach trying, unsuccessfully, to teach his football players that they’re not allowed to rape women (Solnit 2015). The fact that rape humour has been a staple of Western comedy for centuries, while rapist humour is a recent invention (stoked by the #MeToo movement), reveals that western humour is steeped in epistemic gender bias.

Not only have women been unable to make fun of rapists – or, more specifically, their jokes about rapists were not recognized as humour – women have generally been considered ‘unfunny.’ This is exemplified in Netflix’s comedy lineup. From Netflix’s inception in 1997 until 2014, 57% of comedy specials featured men, 36% featured men and women, and only 7% featured women exclusively. In 2014, 82% of Netflix comedy specials were male-only, while a paltry 14% were female-only (Muller 2016). The top male comedians on Netflix this year were paid much more than the top female comedians (Lynch 2018). These numbers reflect broader trends in the cultural perception of humour. Women appreciate humour in men more than vice versa (Barelds & Dikkstra 2010). Teaching evaluations show that male professors are rated as funnier than female professors across every discipline (Khazan 2017). These are all examples of epistemic gender bias in Western humour, which depicts men as funnier than women.

Feminist humour is especially vilified. In 2012, after comedian Daniel Tosh was criticized for saying that it would be funny if a woman in his audience got raped, Louis C.K. commented that this is part of the “fight between comedians and feminists, which are natural enemies. Because stereotypically speaking, feminists can’t take a joke” (Solnit 2015). Hugh Hefner expressed similar views about the relationship between feminism and pornography, saying that “these chicks [feminists] are our natural enemy. It is time to do battle with them” (Altman 2008). The feminist response to the male gaze is a visceral threat to both male-dominated pornography and male-dominated comedy, which is why pornographers and comedians have gone to such lengths to bash feminists, depicting them as unfunny and unsexy. Mainstream pornography and mainstream comedy, with their naturalization of misogynistic violence through rape fantasies and rape jokes, are perfect bedfellows. They work together to solidify male entitlement.

Feminist comedian Michelle Wolf was the most recent target of epistemic gender bias. At the 2018 White House Correspondents Association Dinner, she used humour to attack Trump, his supporters, and other high-profile misogynists. She opened her set with, “like a porn star says when she’s about to have sex with a Trump, let’s get this over with.” Then she proceeded to criticize Sarah Huckabee Sanders and Kellyanne Conway for lying on Trump’s behalf and sanitizing his sexist, racist, xenophobic behaviour (“Maybe she’s born with it. Maybe it’s lies. It’s probably lies”). Through insinuation, she called Roy Moore a child rapist (“I’m 32 years old, which is an odd age: 10 years too young to host this event and 20 years too old for Roy Moore”); she described Roger Ailes and Bill O’Rielly as sexual harassers (“Fox News is here. So, you know what that means, ladies: Cover your drinks”); and she ridiculed Mike Pence’s anti-choice views (“He thinks abortion is murder, which, first of all, don’t knock it til you try it. And when you do try it, really knock it. You’ve got to get that baby out of there”). What topics would you expect a feminist comedian to address in her comedy? Sexism, rape, sexual harassment, and abortion are high on the list. Wolf’s comedy relentless roasts public figures who don’t support women’s rights, and this is one of the reasons many critics have denounced her as “mean-spirited, vulgar, and unfunny” (Smith 2018).

Feminist humour is Wolf’s comedic brand. In her HBO special, Wolf warns the audience, “I should just let everyone know: I am a feminist” (Cunnigham 2018), and then proceeds with the type of feminist humour that she displayed at the WHCA dinner. Wolf seems to recognize that gender bias is a factor in the comedy world, or at least, she realizes that being recognized as having a sense of humour is connected with being seen as a person. In an interview with Marlow Stern, she says, “to do stand-up, you have to have a strong point of view, and doing stand-up gave me that strong point of view… It made me a person” (Stern 2018). Because of epistemic gender bias in comedy, women have been prevented from having a strong comedic voice: they have been stereotyped as, to paraphrase the title of Wolf’s comedy special, “nice ladies.” If women are funny, they aren’t nice ladies: they are un-ladylike, mean, unattractive, unsexy, and corrupt. Women face a double bind in comedy, as elsewhere: be funny, or be a nice lady, but not both. Wolf is one of the comedians deconstructing this binary logic with unapologetic feminist humour.

Because women’s humor is susceptible to epistemic gender bias, women’s blame, as expressed in humour, is susceptible to epistemic gender bias. If rape jokes are funny but rapist jokes aren’t even recognized as a type of humour, women can’t use humour to blame rapists (or to defend themselves from rape jokes). More broadly, women can’t use humour as a vehicle for blame at all, or at least, they can’t use it as effectively as men. This is a tragedy, because humour has always been deeply political. Wolf has been compared to George Carlin, a comedian “whose goofy delight in the sounds and cadences of American speech never obscured the genuine irritation, bordering on rage, that fuelled his jokes” (Cunningham 2018). Humour by its very nature violates cultural norms (McGraw & Warren 2010). If women can’t make fun of misogynistic cultural norms, they can’t reveal those norms as, not just wrong, but absurd. While some theorists worry that humour is apolitical because it trivializes norm violations (see Benatar 2014), what current trends show is that humour is a potent political tool that can be used to resist social injustice, and hold its protagonists and apologists accountable (e.g., Weinstein, Cosby). Feminist humour is such a tool, because it punches up rather than down.

Feminist comedy, then, is an important vehicle for blame, but it is still not as respected, on average, as male humour. This produces gender bias in our shared blaming practices.

The epistemic gender biases reflected in (i) pornography, (ii) the literary canon, and (iii) the world of comedy are brought into stark relief in the recent New Yorker short story, ‘Cat Person,’ by Kristen Roupenian (2017). In the next section, I analyze how this story illuminates the intersection of epistemic gender bias, pornography, the male canon, and humour.

5. Cat Person

‘Cat Person; is the first piece of literary fiction ever to go viral. It has been widely characterized as a story about ‘bad sex,’ where ‘bad’ has multiple connotations (Bennett 2017). The story describes the protagonist Margot’s sexual encounter with a virtual stranger, Robert, who initially appears to be sympathetic (albeit enigmatic), but is exposed as a misogynist when he harasses Margot through text messages and calls her a ‘whore.’

Although ‘Cat Person’ is one of the most read ‘New Yorker’ stories of all time, it has not been treated with the seriousness accorded to most literary fiction. As Constance Grady observes, “The trivializing of women’s stories also plays into one of the persistent oddities surrounding ‘Cat Person’; namely, the frequency with which readers have called it an ‘article’ or an ‘essay’ or generally treated it as a piece of nonfiction rather than as a short story” (2017). While ‘Cat Person’ has few of the hallmarks of autobiographical writing (e.g., it was written in the third person, and in an elevated style), the fact that it features the narrative perspective of a female protagonist marks it as a ‘confessional tell-all’ to many readers. “Women’s subjectivity,” after all, “is not for serious literary fiction…; it’s for unserious, uninteresting, unpaid-for online writing” (Grady 2017). Responses to ‘Cat Person’ therefore illustrate ‘topic bias’ and the trivialization of women’s literary voices.

In spite of epistemic gender bias, how did ‘Cat Person’ attract so much attention? One reason is that it gave a voice to an experience shared, in silence, by so many women: the experience of ‘bad sex’ – not just in the sense of unpleasant or disappointing or awkward sex, but more specifically, objectifying, infantilizing, and alienating sex – the kind of sex represented in Hugh Hefner’s depressingly mechanical orgies, in mainstream pornography, and in the masculinist perspective of classic literary heroes. When Margot has sex with Robert – or, more accurately, when Robert fucks her – Robert is described as “mov[ing] her through a series of positions with brusque efficiency, flipping her over, pushing her around, and she felt like a doll again, as she had outside the 7-Eleven, though not a precious one now—a doll made of rubber, flexible and resilient, a prop for the movie that was playing in his head.” Robert is, in effect, assuming Hefner’s role as pornographer, posing Margot in postures that one would find in mainstream pornography, and treating her as a ‘doll,’ reminiscent of the girls depicted in ‘Madame Butterfly’ and ‘Lolita.’ Robert isn’t the least bit worried about Margot’s being 14 years younger than him (20 versus 34), that she is presumably a virgin, or that she is very intoxicated (having been illegally served alcohol by him); in fact, her youth and presumed inexperience seem to turn him on. (He’s irritated when she denies being a virgin, and later slut-shames her for being sexually active). Robert never thinks to ask Margot what she wants from him; instead, “he slapped her thigh and said, ‘Yeah, yeah, you like that,’ with an intonation that made it impossible to tell whether he meant it as a question, an observation, or an order.'” Feminists like Rae Langon have noted that “pornography carries authority as instruction as well as entertainment, and [Langton] cites evidence to suggest that a high percent of boys and young men regard men’s satisfaction as a right and women’s rights as an irrelevancy” (in Solnit 2015). Pornography, it seems, instructed Robert to direct Margot in a glib pornographic film that completely erased her agency, in an encounter that fell precipitously short of the ideal of affirmative consent. (How can you consent if you’re never asked a question?) The only time Margot experiences any pleasure is when she imagines herself through Robert’s eyes: “Look at this beautiful girl, she imagined him thinking. She’s so perfect, her body is perfect, everything about her is perfect, she’s only twenty years old, her skin is flawless, I want her so badly, I want her more than I’ve ever wanted anyone else, I want her so bad I might die.” While this might feel empowering to Margot, it is actually an empathic simulation of Robert’s point of view as he objectifies Margot through the lens of the male gaze. Margot, too, has been instructed by mainstream pornography: she has been directed to empathize with the male gaze to the extent that she loses sight of her own agency.

This sexual encounter involves epistemic gender bias, then, because it erases Margot’s agency. This loss of agency is reflected in Margot’s inability to say ‘no’ to Robert. She cannot say ‘no’ because (1) she can’t fully recognize that what Robert is doing is wrong, (2) she is ‘too nice’ to hurt his ‘feelings,’ (3) she worries that Robert might be a rapist, in which case, he wouldn’t listen anyways, and (4) if Robert is a rapist, it might be easier to go along with getting fucked by him than to resist him and get raped while struggling.

Margot wonders multiple times throughout the story whether Robert might be a rapist (“And, as though fear weren’t quite ready to release its hold on her, she had the brief wild idea that maybe this [his house] was not a room at all but a trap meant to lure her into the false belief that Robert was a normal person, a person like her, when in fact all the other rooms in the house were empty, or full of horrors: corpses or kidnap victims or chains”). She continually dismisses this worry as absurd; but is Robert a rapist? We don’t know. Although Robert doesn’t rape Margot, what would he have done if she had resisted? At first, Margot thinks that Robert is a sensitive person, and this seems confirmed when he starts “talking about his feelings for her” after fucking her, admitting that during her reading week, “an entire secret drama had played out in his head, one in which she’d left campus committed to him, to Robert, but at home had been drawn back to the high-school guy, who, in Robert’s mind, was some kind of brutish, handsome jock, not worthy of her but nonetheless seductive by virtue of his position at the top of the hierarchy back home in Saline.” This might (superficially) suggest that Robert is a sensitive person who cares about Margot, but this illusion is dashed when Robert texts Margot after seeing her at a bar with a friend, disclosing his toxic masculinity: “Is that guy you were with tonight your boyfriend… Or is he just some guy you are fucking… Sorry… When u laguehd when I asked if you were a virgin was it because youd fucked so many guys?… Are you fucking that guy right now… Are you… Are you… Are you… Answer me… Whore.” Robert doesn’t have “feelings” for Margot – he has toxic masculinity. He doesn’t care about Margot’s feelings, or autonomy, or sexual pleasure – he only cares about being ’emasculated’ by another man.

Margot is – like the title of Michelle Wolf’s comedy special – a ‘nice lady’: too nice to say ‘no’ to Robert’s degrading fuck fest, which she describes as “the worst life decision I have ever made!” Margot worries repeatedly about the prospect of hurting Robert’s feeling, on the false assumption that Robert is a sensitive person and a ‘nice guy.’ But Robert isn’t sensitive, unless you count fragile narcissism as a kind of sensitivity, and he isn’t a ‘nice guy,’ except in the stereotypical sense of a mediocre white dude who feels entitled to sex with women. As Erin Tatum writes, “the most stereotypical incarnation of the Nice Guy is a fedora-clad dudebro who spends too much time on Reddit and would probably push a six-year-old girl out of the way to get his hands on My Little Pony merchandise, [but] the more garden-variety Nice Guy can be more difficult to spot” (2015). That nice guy is Robert. Roupenian intentionally wrote Robert as an ambiguous character, and this makes it easier to project positive qualities onto him (at least, until he calls Margot a ‘whore’). Margot can’t say ‘no’ to Robert because, aside from fear of rape and empathizing with the male gaze, she can’t discern whether Robert is a nice guy or a Nice Guy™. 

While Margot is disgusted with Robert, she also finds much to laugh at in her encounter with him. ‘Cat Person,’ in fact, reads much like a comedy of manners, a genre of comedy epitomized by Oscar Wilde, which satirizes the manners and customs of contemporary society, and represents social conventions in stock characters (e.g., the foppish rake). If ‘Cat Person’ achieves anything, it is the satirizing of toxic masculinity, as embodied in Robert. Robert is depicted unflatteringly as “awkwardly bent, his belly thick and soft and covered with hair,” and hovering above Margot with his “fat old man’s finger inside her.” Robert is, in effect, a clown, and not just any clownish figure, but the incarnation of toxic masculinity as ridiculous and absurd – yet also inescapable and irrepressible, penetrating Margot (symbolically) with its fat, hairy, probing finger. Robert, as a clown, is a narcissist full of sexual confidence, but completely stupid when it comes to sexually pleasing a woman. He stumbles through sex like a clown stumbling across a stage, making a fool of himself, all the while full of preening confidence. ‘Cat Person,’ on scrutiny, is an excellent, albeit subtle, example of feminist comedy. If it is not recognized as comedy, this is because feminist comedy lacks cultural uptake. Roupenian seems aware of the lack of uptake for this type of humour. At one point, Margot imagines herself relating her sexual fiasco to a future boyfriend: “‘And then he [Robert] said, ‘You make my dick so hard,’ and the [imagined] boy would shriek in agony and grab her leg, saying, ‘Oh, my God, stop, please, no, I can’t take it anymore,’ and the two of them would collapse into each other’s arms and laugh and laugh—but of course there was no such future, because no such boy existed, and never would.Why can’t such a boy exist? Perhaps because our culture lacks the hermeneutical resources to frame Robert as a clown, to see toxic masculinity as ridiculous, or to take Margot’s perspective, not Robert’s, as the dominant frame of reference.

‘Cat Person,’ then, nicely illustrates how epistemic gender bias operates through the male gaze to sexually objectify women, treat women as pornographic objects, discredit women’s literature and women’s literary perspectives, and render women’s humour virtually unintelligible. All of these threads come together in this short story, and perhaps this is one of the reasons ‘Cat Person’ is one of the most-read stories of the year.

What does all of this imply about blame? It is notable, and disturbing, that Margot never blames Robert for his demeaning, objectifying fucking-of-her, for asking if she’s fucking her friend, or for calling her a whore. One has to wonder: has anyone ever blamed Robert, or is Robert another Donald Trump, non-consensually grabbing women’s pussies to his heart’s content? If Robert has no idea that he’s terrible at sex, that he treats women like garbage, maybe this is because no one – not Margot, nor anyone else he’s fucked – has held him responsible. Maybe Robert lives in a world, as Solnit describes it, “in which other people only exist to help reinforce [his] magnificence.” When women can’t say ‘no,’ can’t tell men that they’re bad at sex, can’t make fun of male egos, can’t tell men that it’s wrong to harass, stalk, and slut-sham them, then women’s blame is silenced. ‘Cat Person’ is a story about how women’s blame is suppressed, and sometimes can’t even be spoken. How many other women does Robert ruthlessly fuck after Margot ghosts him? Who knows? Robert might not have “a house full of horrors,” but he has a closet full of skeletons.

6. Conclusion

Mainstream pornography, the male literary canon, and mainstream humour all contain strong overcurrents of the male gaze, although the masculine point of view is being overturned by feminists and their allies. Yet we still live in conditions of epistemic injustice in which women’s voices are marginalized, and this includes women’s blame. With enough political will, women’s speech can receive equal credit with men’s.


Gaslighting & its effects on the moral community


This post is about gaslighting as a distinctively morally ‘dark’ type of epistemic injustice, which impairs our moral relationships. In section 1, I describe gaslighting as a type of epistemic injusticewith distinct moral features that render it particularly morally pernicious, following Kate Abramson’s illuminating discussion (2014). In section 2, I argue that, because there are two types of epistemic injustice – testimonial and hermeneutical – there should also be two corresponding types of gaslighting. I say that, while paradigmatic cases of gaslighting are instances of testimonial injustice, some cases of gaslighting are examples of what Alyssa Cirne calls “willful hermeneutical marginalization” (2012: 46), a type of epistemic injustice involving the manipulation of hermeneutical resources to harm a vulnerable person or group. In section 3, I argue that the suppression of pertinent knowledge, as described by Charles Mills (2007), is an undertheorized type of “hermeneutical gaslighting,” though this type of gaslighting is increasingly visible in public discourse. In section 4, I give a prominent example of hermeneutical gaslighting – namely, Sam Harris’ interview with Charles Murray (“Forbidden Knowledge”)on the putative connections between IQ, “race,” and genes, and I explain why this interview is a case of hermeneutical gaslighting. In section 5, I argue that cases of hermeneutical gaslighting via the suppression of knowledge tend to involve or invoke just world bias, and I give some examples from professional philosophy. Finally, in section 6, I argue that gaslighting perpetrates distinct moral and epistemic harms; specifically, it undermines the epistemic and moral standing of its victims and creates epistemic and moral inequality.

  1. Gaslighting as a distinct moral violation

Gaslighting is commonly understood as the act of psychologically manipulating someone into questioning her own sanity. Kate Abramson gives this definition more substance by characterizing it as “a form of emotional manipulation in which the gaslighter tries (consciously or not) to induce in someone the sense that her reactions, perceptions, memories, and/or beliefs are not just mistaken, but utterly without grounds – paradigmatically, so unfounded as to quality as crazy” (2014: 2). Gaslighting, then, has a pathologizing effect. The gaslighter can have any number of subjective motives; not all gaslighters have the same motivational profile, or even substantively overlapping motivational profiles, except that they all wish, on some level, to have their worldview validated and “placed beyond dispute” (Abramson 2014: 10). The act of gaslighting nonetheless has certain paradigmatic features, including that (1) it undermines the victim’s standing to make moral claims, and thereby excludes the victim from the moral community (viz., Darwall 2002, Strawson 1963); (2) it undermines the victim’s self-conception as a person, i.e., someone capable of making intelligible moral and epistemic claims; and (3) it involves manipulation. Gaslighting, seen in this light, is a type of epistemic injustice, or an act of epistemic marginalization rooted in identity prejudice (Fricker 2007). But gaslighting is not just an ordinary type of epistemic injustice; it is especially morally heinous, or ‘dark,’ because it involves a number of overlapping violations, including: (i) it is manipulative, (ii) it involves a lack of recognition respect, (iii) it silences and pathologizes the victim, and (iv) it is a type or torture.

Gaslighting is particularly heinous because it doesn’t just involve a depreciatedcredibility rating of the victim’s speech; it involves a credibility rating of zero. The victim is depicted as “crazy,” “irrational,” and incapable of understanding the world and her place in it. Galighting is also a type of torturebecause it deceives the victim into thinking that there is something she could do to gain credibility in the eyes of the gaslighter, which is a false hope; and it destroys the victim’s sense of self – it persuades her that she is not even a minimally rational agent. False hope and loss of agency are characteristic features of classic torture, and also of gaslighting. Gaslighting involves a lack of recognition respectbecause it depicts the victim as an outsider to the moral community, a ‘moral refugee,’ as it were. From this epistemic position, the agent cannot make intelligible claims to be treated with respect, and she cannot negotiate effectively with members of the community. Because gaslighting has all of these features, it is a “horrifying puzzle” of injustice, a distinctively ‘dark’ moral violation (Abramson 2014: 18). Interestingly, gaslighting cannot simply be seen as a straightforward violation of the categorical imperative, because it’s worse than ordinary cases of objectification, too; it doesn’t justundermine a person’s agency; it usesthe person’s agency to undermine her own sense of self, and this is an especially egregious form of abuse (Abramson 2014). The agent herself is co-opted into doubting her own epistemic authority as well as contributing to the maintenance of the gaslighter’s narcissistic worldview.

Abramson adds that gaslighting has a distinctively genderedcharacter, given that paradigmatic cases of gaslighting involve a man-identified perpetrator and a woman-identified victim. Thus, gaslighting typically serves to entrench misogynistic cultural norms, on top of silencing individual women. This was the relational dynamic depicted in the original play by Patrick Hamilton(1938), in which the protagonist Paula was relentlessly gaslighted by her husband Gregory. This abusive relationship provides the frame through which we tend to interpret gaslighting. However, Abramson notes that gaslighting is not always so explicit, and can encompass a variety of motives and relational dynamics. Nonetheless, Abramson focuses on dyadicrelationships involving a gaslighter and a victim. To illustrate the paradigm case, she cites familiar examples of quid pro quo sexual harassment and racial discrimination, and includes Beauvoir’s (in)famous relationship with Sartre, in which Sartre manipulated her over many years into believing that her “opinions were based only on prejudice, bad faith, or thoughtlessness,” leaving her wondering whether she was capable of “think[ing] at all” (Beauvoir 2007, cited in Abramson 2014: 4). Beauvoir was essentially gaslighted into doubting her epistemic agency.

While gaslighting is stereotypically gendered, it is a type of epistemic injustice that can target members of anyepistemically marginalized group (or the group as a whole, I shall argue). Thus, gaslighting can be used against not only women, but People of Color, members of the LGBTQIA community, and virtually any other marginalized social group. As epistemic injustice, it relies on identity prejudice to gain purchase. Gaslighting exploits cultural stereotypes to silence members of marginalized groups, to withhold knowledge from members of marginalized groups, and to suppress pertinent knowledge about the lived experiences of marginalized groups – or so I shall argue.

  1. Gaslighting as epistemic injustice

Abramson focuses on dyadic, intimate cases of gaslighting, and these are indeed the paradigms supported by Hamilton’s play and popular culture. Construed as types of epistemic injustice, these are tokens of what Miranda Fricker calls “testimonial injustice,” in which a hearer gives a speaker a “deflated level of credibility” due to identity prejudice, i.e., bias triggered by the perception of a person’s demographic attributes (Fricker 2007: 1). While these cases of testimonial injustice are especially morally pernicious because of the distinctive features of gaslighting (e.g., pathologizing effects, torture), they are still essentially instances of epistemic injustice. Abramson mentions that testimonial injustice can contribute to a second type of epistemic injustice: “hermeneutical injustice,” which Miranda Fricker describes as a “gap in collective interpretive resources that puts someone at un unfair disadvantage when it comes to making sense of their social experience” (2007: 1); but, beyond this, Abramson doesn’t elaborate on the relationship between the two types of injustice. Fricker herself says that hermeneutical injustice occurs at a “prior stage” to testimonial injustice (ibid.), meaning that it creates fertile ground for testimonial injustice, normalizing and naturalizing prejudiced credibility assessments. In fact, it might be more accurate to see the two types of injustice as mutually implicated in a positive feedback loop, wherein testimonial injustice produces deficits in hermeneutical resources and hermeneutical deficits cause testimonial injustice, which again produces deficits in hermeneutical injustice, and so on. This process can be seen in Sartre’s treatment of Beauvoir: he exploited gaps in the hermeneutical resources pertaining to women’s credibility to perpetrate testimonial injustice on Beauvoir, which in turn reinforced the very hermeneutical gaps he was using against her – cultural ignorance about women’s epistemic authority. As a result, Beauvoir quit philosophy and we lost invaluable epistemic resources within the field and popular culture (Fricker 2007). Hermeneutical deficits are not free-standing cultural facts; rather, they are instantiated in people’s cognitive architecture, in the form of implicit biases, heuristics, and other psychological states (Scorberg 2007). The mental states that give rise to a backdrop of hermeneutical injustice also cause distinct acts of testimonial injustice perpetrated by individual agents.

That said, the distinction between the two types of epistemic injustice is not insignificant. One of the key asymmetries between the two, according to Fricker, is that people perpetratetestimonial injustice whereas “no agent perpetrates hermeneutical injustice – it is a purely structural notion” (2007: 159). “Moments of heremeneutical injustice,” she says, are caused by the epistemic conditions of the time (ibid.). Therefore, perpetuators and victims of testimonial injustice are both victimsof the same climate of hermeneutical injustice that they mutually inhabit; people who epistemically marginalize members of oppressed groups are also often victims of “epistemic bad luck” (2007: 151). Alyssa Cirne argues that this picture is flawed because it omits an important type of hermeneutical injustice, “willful hermeneutical marginalization,” in which an agent “obstructs or withholds hermeneutical tools from those agents who need those tools the most,” for purposes of self-aggrandizement and self-conceit (2012: 45). This subspecies of hermeneutical injustice allows us to realize the “agency and culpability” in such scenarios, and to identifyperpetratorsof hermeneutical injustice (ibid.). Hermeneutical injustice, then, is not simply a set of background conditions, but the cumulative effect of people’s willful choices.

Notably, although Cirne describes acts of hermeneutical injustice as “willful,” this doesn’t entail that they are intentional harms. Social epistemologists generally accept that epistemically marignalization can be unintentional, as they can operate through implicit states such as implicit biases and heuristics, as noted above (see also Peels & Blaauw 2016). Acts of epistemic marginalization are nonetheless ‘willful’ in that they are motivated by a vested interest in unfair asymmetries of power, and, as such, they express the agent’s ‘will’ or value system or deep self. ‘White ignorance,’ for instance, is not caused by intentional racism – if it were, it would not be a case ofignorance; but white ignorance is always motivated by a personal stake in relations of power that favour white people (Mills 2015). White ignorance reflects the ignorant person’s deep self.

While Cirne focuses on the willful withholdingof valuable epistemic resources from a victim of epistemic injustice, I think that willful epistemic marginalization could extend further, to the willful suppressionof epistemic resources, which doesn’t necessarily involve the epistemic marginalization of any individual directly, but rather involves the marginalization of an entire social group. Mills provides a good example; he observes that white ignorance is sometimes manifested in the pervasive cultural myth that, “after the abolition of slavery in the United States, blacks generally had opportunities equal to whites,” and he attributes this myth to the “suppression of pertinent knowledge.” (2007: 21). Following Cirne’s analysis, the “suppression of pertinent knowledge,” like the withholding of epistemic resources, can’t be seen as an inert historical fact, but must be recognized as a result of human agency– something perpetrated by people. This suggests that speech acts that suppress pertinent knowledge could potentially count as instances of willful epistemic marginalization, even if the speaker doesn’t withholdknowledge from anyparticular victim. Instead, they suppress cultural knowledge about a social group’s shared experiences, thereby rupturing, or “poking holes” in, the heremeneutical resources of the community, making it difficult for the target group to accurately frame and express their shared experience of oppression. Speech acts that suppress knowledge in this way may be instances of willful hermeneutical marginalization, acts that epistemically marginalize an entire social group.

Testimonial injustice and willful hermeneutical marginalization tend to go together, but they are nonetheless conceptually distinct. Someone who denies the harmful effects of slavery on the life prospects of African Americans is likely toalsodiscredit the speech of African Americans to their faces (because this person harbors racial bias), but the person commits twodistinct actsof epistemic injustice, one hermeneutical and one testimonial. These types can be differentiated in practice.

  1. Testimonial and hermeneutical gaslighting

Now, if gaslighting is a type of (especially pernicious) epistemic injustice, and there are two types of epistemic injustice, then there should be two corresponding types of gaslighting. Gregory’s dismissal of Paula’s speech as “crazy” is an example of the first type, “testimonial gaslighting”; he gives her speech zero credit. But Gregory also commits willful hermeneutical marginalization by barring Paula’s family from visiting, knowing that her family could debunk her false beliefs about her nil epistemic standing. Thus, Gregory also commits “hermeneutical gaslighting,” the willful withholding of hermeneutical resources to manipulate Paula into subordination and false consciousness about her situation. Gregory thus commits two epistemically distinct types of gaslighting. Both instances of epistemic injustice are also examples of gaslightingbecause they involve manipulation, undermine the victim’s moral standing, assault the victim’s self-conception, and they pathologize, disrespect, and torture the victim.

Is the willful suppression of knowledgealso a type of gaslighting? Paradigm cases of gaslighting involve an intimate relationship between a gaslighter and a gaslighting victim, as opposed to a statement of false belief about a social group (with which the gaslighter may have no contact at all, particularly in a segregated society). Abramsom says that in paradigm cases of gaslighting, the gaslighter uses the promise of love and intimacy to manipulate the victim; he appeals to the victim’s empathy; and he threatens the victim with reprisals for non-compliance (2014: 20). When someone suppresses pertinent knowledge, that person doesn’t necessary rely on intimacy, affection, empathy, or reprisals to manipulate the target group, but the person’s act of epistemic marginalization still has central characteristics of gaslighting, i.e., manipulation, moral disrespect, pathologizing, torture (albeit of a group). The ‘manipulation’ in question here is not the physical manipulation of an individual, but rather, the manipulation of epistemic resourcesso as to render the shared experiences of a marginalized group unintelligible and ostensibly “pathological.”

Another salient difference between the two cases of gaslighting is that testimonial gasighting targets a specific victim, whereas knowledge-suppression targets an entire social group, without directly harming any individual. Hermeneutical gaslighting, in effect, creates fertile epistemic groundfor tokens of testimonial injustice, by infusing the social imaginary with pernicious stereotypes – stereotypes that ‘testimonial gaslighters’ can exploit to effectively discredit and pathologize members of the target group. People who commit willful knowledge-suppression give others ammunitionfor committing testimonial injustice behind a cloak of plausible deniability.

One way of framing this difference is to think of testimonial injustice as akin to quid pro quo sexual harassment, which affects an individual victim, whereas hermeneutical injustice is akin to hostile workplace sexual harassment, which creates a hostile environment for all members of a protected group. If an employer posts sexualized pictures of women in his office, this is an example of hostile workplace sexual harassment because it affects all of the women who enter the office. Similarly, hermeneutical scapegoating creates an epistemically hostile environmentfor a marginalized group. Even if only one person complains, everymember of the group is affected.

This type of scapegoating may not fit the classic paradigm, but it fits with an emerging popular conception of gaslighting, which we see in, for example, claims that Donald Trump is “gaslighting” the American public by lying, distorting the facts, and disparaging the press (Leve 2017).

We also see this type of gaslighting in Angelique M. Davis and Rose Ernst’s theory of “racial gaslighting,” defined as “the political, social, economic and cultural process that perpetuates and normalizes a white supremacist reality through pathologizing those who resist” (2017: 1). This type of gaslighting is not part of an intimate relationship, but is instead part of a complex network of discursive relationships and power structures. Racial gaslighting relies on “racial spectacles,” or “narratives that obfuscate the existence of a white supremacist state power structure” (ibid.), so as to normalize and perpetuate racism. Racial spectacles, in effect, suppress knowledgeabout the shared experiences of racialized minorities. While Davis and Ernst describe racial gaslighting as a sociopolitical process, and while they focus on the historical and sociological foundations of racial gaslighting (similar to Fricker’s analysis of hermeneutical injustice), they nonetheless affirm that people (and groups of people) commit gaslighting – for example, “The U.S. government usedracial spectacles at the macro level to publicly justify its use of concrete state action against those of Japanese ancestry during World War II” (2017: 6). Thus, they do not deny the agency implicated in racial gaslighting; they admit that it can be perpetrated.

We can see Mills’ example of a white person (or people) denying the reality of racial inequality in America as a type of “racial spectacle,” a performance of white ignorance. And I think that we can apply my analysis of that case to racial gaslighting in general. That is, I think it could be useful to frame racial gaslighting as an especially morally pernicious type of willful hermeneutical injustice, i.e., hermeneutical gaslighting. These spectacles gaslight Black people by suppressing knowledge about their lived reality. While these spectacles do not, as Davis and Ernst note, effectively deceiveall, or even most, People of Color, they create an epistemic climatethat pathologizes, stigmatizes, and epistemically marginalizes them, placing them at an epistemic and moral disadvantage. These acts of injustice have the characteristic features of gaslighting: moral disregard, assault to the self, and torture. Racial gaslighting undermines the standing of racialized minorities to make intelligible moral and epistemic claims; forces them to choose between dominant cultural narratives and their authentic experiences; and represents them as fundamentally irrational, “mad,” and unworthy of having their knowledge inscribed in public discourse.

  1. Suppression of knowledge as gaslighting: Example and analysis

There are many, many topical examples of hermeneutical gaslighting via the suppression of knowledge. One that got a lot of publicity recently was Sam Harris’ interview with Charles Murray on his book on race and IQ, “The Bell Curve” (1994). In short, Murray holds that there are genetic differences between “races” that explain average differences in IQ. The critical flaw in this theory, as noted by Eric Turkheimer, Kathryn Paige, Richard E. Nisbett (2017), David Reich (2018), Ezra Klein (2018), and many other people from many disciplines (including genetics), is that Murray’s view falsely explains average differences in IQ by reference to genes shared by racial groups. As Klein says, “There is currently no reason at all to think that any significant portion of the IQ differences among socially defined racial groups is genetic in origin” (2018).

To make matter worse, Murray neverdiscusses environmental causes of IQ in his conversation with Harris, even though this relationship is extensively researched and well documented, as Harris and Murray are well aware. To give a few examples: we know that the black-white IQ gap is closing; there has been an 18-point gain in average IQ in the U.S. from 1948 to 2002 – more than twice the current racial IQ gap, which is only 10 points (misquoted as 15 points by Murray, whose data set is outdated); adoption from a poor family to a wealthy family is the strongest predictor of IQ gains, and so on. Harris and Murray never mention these clear relationships between environment and IQ. Their critical error, then, is not only to explain average IQ differences by reference to racial genetics, but also to willfully suppressknowledge about environmental causes of IQ differences.

But this is just the most basic error – there are many other errors that follow from this flawed axiom. One is that Murray commitspreciselywhat Mills identifies as a paradigmatic act of white ignorance – he says that Black people and white people have approximately equal social opportunities, which means that average IQ differences must be genetic. (In fact, Harris says this, and Murray confirms it: they both make the same mistake). Here is that part of the interview:


HARRIS: I have here a quote from Flynn — I don’t know when he wrote this or said this — but he says, “An environmental explanation of the racial IQ gap need only posit this: that the average environment for blacks in 1995 matches the quality of the average environment for whites in 1945.I do not find that implausible.” So what you just said seems to close the door to that [environmental] interpretation of the black-white gap.

MURRAY: Yes, it does, and this is a case where I am citing someone who has done analyses that are at a level of complexity that I am not independently competent to pronounce.


Second, Murray says that because IQ differences are genetically-based as opposed to cultural, social policies that address putative social inequalities between white populations and Black populations (which Murray doesn’t believe in) are misguided and should be eliminated:


HARRIS: I guess one thing that must be occurring to listeners now — and this is my misgiving about having this conversation and going into this area at all — the question is why talk about any of this? Why seek data on racial difference at all? What is the purpose of doing this?

MURRAY: Because we now have social policy embedded in employment policy, in academic policy, which is based on the premise that everybody’s equal above the neck, all groups are equal above the neck, whether it’s men and women or whether it’s ethnicities. And when you have that embedded into law [i.e., in the form of employment equity policies], you have a variety of bad things happen.


Third, Murray and Harris both believe that Murray is the victim of left-wing fascists and other ‘enemies of science.’ In their view, racialized minorities are decidedly not victims of Murray’s “junk science,” as described by his peers(2017); Murray is in fact the victim. This is an example of what I described in an earlier post as politically motivated scapegoating: denying responsibility by shifting blame onto a socially marginalized group, which is already stigmatized by pervasive cultural stereotypes, making them an easy target. Harris is clearly in on the scapegoating and epistemic marginalizing. He calls his talk with Murray “Forbidden Knowledge,” depicting it as the dissemination of repressed truths, instead of what it really is: yet another iteration of familiar racial stereotypes. The idea that Black people are unintelligent due to shared genetic deficits isn’t a radical new perspective, it’s an old racist myth! Indeed, it’s part of the very fabric of American society, which was founded on a principle of equality for white men, slavery for Black people, and domestic servitude for women. The original American social contract was, as others have pointed out, a racial (Mills 1997) and a sexual (Pateman 1988) contract, which legally codified white male supremacy as the de facto natural order. The legacy of this inegalitarian contract can still be seen in every major social institution. Every American with an elementary school education surely knows about America’s colonialist roots and racist cultural narratives; therefore, literally no one was enlightened by Harris’ interview with Murray. But many probably enjoyed it because it justified their privileges to them, mitigating any white guilt they may otherwise have felt.

What Harris and Murray did was, on my description, hermeneutical gaslighting. They didn’t discredit any individual person, which is the paradigmatic case of testimonial injustice. They also didn’t exploit an intimate relationship or emotional interdependence, which are characteristic features of classic gaslighting. But they did discredit Black people by saying that Black people as a group are genetically unintelligent. They didn’t exactly withholdhermeneutical tools from the target group; members of this group still have access to the same hermeneutical resources as before, though the Internet, libraries, and so on. But Harris and Murray suppressed knowledgeabout the shared cultural experiences of Black people by disseminating junk science about dubious connections between “race” and IQ, and willfully omitting pertinent information about historical and contemporary racial inequality. Their talk made salienta false narrative of racial inferiority, which serves to overshadowand obfuscate objective knowledge about racial injustice.

To be perfectly clear, I’m not saying that Harris and Murray are trying to gaslight their audience; I’m saying that they are gaslighting their audience. Their intentions are irrelevant to the question of whether their speech counts as a case of joint gaslighting, since gaslighitng is often unintentional.

How can we identify their speech acts as instances of hermeneutical gaslighting per se? Because these acts serve to: (1) undermine the standing of the target group, (2) undermine the self-conception of the target group (i.e., their self-conception as full persons), (3) involve manipulation. The manipulation, again, isn’t the manipulation of another person’s actions and choices, as represented in the movie “Gaslight,” but rather, the manipulation of epistemic resourcesrelevant to the target group’s standing. Harris and Murray are distorting the facts in a way that undermines the moral and epistemic standing of Black people – specifically, their standing to make intelligible claims to be seen as members of the community and to be respected as credible witnesses to their own lives. These violations are especially morally ‘dark’ because they discredit, silence, and pathologize their target; they misrepresent Murray’s opponents as left-wing fascists and enemies of science; they attempt to convince Black people that they don’t deserve better, and if they were to succeed in this aim, their target audience would be divorced from their authentic selves as people worthy of respect, and forced to “introject” white ignorance.

I doubt most people would be taken in by Murray’s flawed scientific perspective, but his speech still counts as a case of functionalgaslighting, as it has all the core features of hermeneutical gaslighting. If it succeeds in suppressing knowledge, then it is both functional and effective gaslighting.

  1. Just world bias as a paradigmatic type of gaslighting

Let’s revisit Charles Mills’ example of white ignorance, which I have classified as a typical example of hermeneutical gaslighting. In that example, the gaslighting consists in the speaker’s claim that abolition led to racial equality, i.e., that society is now fair. We see this myth echoed in Murray’s claim that racial equality emerged in the mid-20th Century. By denying the reality of racial injustice, the speaker functionally gaslightsvictims of social injustice. This typical example of hermeneutical gaslighting is also an example of “just world bias,” the tendency to see the world as if it were just (Hazlitt 2017; Burkeman 2015). Privileged people are susceptible to just world bias because they have a vested interest in denying the reality of their privileged status, as admitting to that unearned status would induce guilt and other negative emotions, and perhaps create a moral incentive to act differently (See my post on white guilt as a fitting moral emotion. To avoid feeling bad, privileged people invoke a false image of a fair world. Many cases of hermeneutical gaslighting, it seems, are caused or constituted by just world bias.

It’s best that we recognize the pervasiveness of this bias to avoid perpetuating it in our own expressive practices. As a rule, don’t say that any social space is equal. Just world bias isn’t necessarily about the state of the world; indeed, it’s usually expressed in a domain-specific way, regarding a specific social space, which is framed as being better than the ‘tawdry mainstream.’

Let’s say, hypothetically speaking, that someone were to proclaim that philosophy is ‘flat,’ meaning an even playing field for all. Although the world is unfairly hierarchical, philosophy is immune from the injustices of mainstream culture. This is an example of domain-specific just world bias, and therefore a paradigmatic case of hermeneutical gaslighting on my view.

The claim is patently false for too many reasons to list, but if we just focus on gender inequalityin the profession, we can see many counterexamples to the just-world hypothesis. As my colleague Jill Delston observed in a recent St. Louis ‘Faculty Forward’ talk (2018), men make up the majority of high-ranking and full-time faculty in philosophy in the U.S., whereas women are overrepresented in contingent faculty positions; women’s work is less likely to be accepted for publication, less likely to be read, and is less cited than men’s; co-authored papers help men’s careers but hurt women’s; having children helps men but hurts women in academic philosophy, etc. These are just a few quantifiable examples of gender bias in the field, and they don’t speak to the qualitative experiences of women who face these injustices in their daily work lives.

Denying the reality of women’s experiences as philosophers is hermeneutical gaslighting. The claim of disciplinary gender equality rests on a false, cisgender, white, male perspective. It implicitly denies women’s claim to be treated fairly and with epistemic respect, since the demand for equality, within a field construed as already-equal, can only possibly be irrational, selfish, and “crazy.” If women accept this false narrative, they’re deceived into thinking that they don’t deserve more credit and respect from their colleagues. The just-world claim also implies that women can succeedif they just try as hard as men, which is false, since women already try harder than men (on average) and get less recognition (on average) due to gender bias and systemic inequality. The deceptive framing of philosophy as equitable is, in fact, torturous, in the sense that it tells women that they are capable of achieving equality with men if they just try harder, which is false; and if we were to accept this myth, we would be alienated from our authentic experience of oppression in the field. We would also be forced to accept or “introject” the male gaze, which isthe typical phenomenological effect of being gaslighted by a man(Abramson 2015).

I bring this up because this is a philosophy journal, and I believe that gaslighting happens in philosophy, perhaps quite a lot. One way to avoid gaslighting your colleagues is to reject just world theory. Don’t pretend that philosophy is fair. But by the same token, don’t pretend that the demographic composition of the profession is fine as it is. Don’t say, for instance, that demographic diversity has no significant pedagogical value(Leiter 2018) when it patently does (Harding 2015; Ciurria 2016; Pammarota 2011; McLaren 2018). In sum, don’t pretend that the profession is or was or soon will be an even playing field, but don’t pretend that it’s no big deal that full professors are overwhelmingly white and male, either (Cherry & Schwitzgebel 2016).

To be fair, I’m not saying that everyone who gaslights is some kind of monster. In conditions of epistemic injustice, it’s hard to know when you’re committing epistemic injustice, including gaslighting and scapegoating, and these transgressions come in degrees. I’ve probably committed acts of “micro-gaslighting” – the gaslighting equivalent of a “microagressions” – given that I have epistemic blindspots rooted in my white privilege. Admitting that you’re susceptible to implicit gaslighting is part of the solution.

  1. The epistemic and moral harms of gaslighting 

The harms of testimonial and hermeneutical gaslighting are both epistemic and moral. Miranda Fricker describes the “primary harms” of epistemic injustice as the silencing and epistemic marginalizing of the victim, and the resultant loss of valuable epistemic resources to the community (2007). These epistemic harms are both individual and collective – they harm the victim’s epistemic standing and the community’s shared knowledge. They are ‘primary’ in the sense that they are epistemic in nature. The “secondary” harms of epistemic injustice are political, socioeconomic, and moral. Epistemic injustice inflicts socioeconomic and political hardships on victims of identity prejudice, who are seen as less eligible for jobs, housing, political representation, education, and other social goods. But victims of epistemic injustice also suffer distinct moral harms– harms that Fricker does not discuss in any detail. (Fricker says little about the ‘secondary harms’ of epistemic injustice in general, leaving it to others to examine them). Abramson points out that the moral harms of gaslighting, a type of epistemic injustice, include injuries to the victim’s (i) moral standing, (ii) moral self-conception (as a moral agent worthy of basic moral regard), and (iii) moral autonomy (specifically, freedom to live according to her un-coerced moral values, or to be herself). This account suggests that the moral harmsof epistemic injustice mirror the epistemic harmsof this type of act: just as epistemic injustice harms the victim’s epistemic standing (or ability to make intelligible claims about what is true), epistemic injustice harms the victim’s moral standing (or ability to make intelligible claims about what is just, fair, or morally permissible); just as epistemic injustice harms the victim’s self-conception as a knower, someone with functional truth-tracking capacities, epistemic injustice harms the victim’s self-conception as a moral agent, someone with functional moral-reasons-tracking capacities; just as epistemic injustice harms the agent’s epistemic autonomy by manipulating her perception of what is true or credible, epistemic injustice harms the agent’s moral autonomy by manipulating her perception of what is just or fair. Furthermore, although Abramson never says this, the moral harms of epistemic injustice are not only individual, but also collective; epistemic injustice derives the communityof valuable moral resources –namely, knowledge about what is right, fair, decent, and morally upstanding. Thus, just as epistemic injustice harms the community’s epistemic resources, it harms the community’s moral resources, too.

Notably, Fricker nowhere says that the ‘primary’ harms of epistemic injustice are worse than the ‘secondary’ harms of epistemic injustice. Epistemic harms are ‘primary’ only because they belong to the same category as the ‘primary’ offense (epistemic injustice) – though it should be noted that epistemic injustice as also a moral transgressioninsofar as it involves morally problematic content, i.e., “ethically noxious motiv[es]” (Fricker 2012; 34), and it produces morally problematic effects, e.g., the loss of moral autonomy. But there is a case to be made that the loss of moral standing is more injurious, or ‘darker,’than the loss of epistemic standing, because our personhood just isour moral agency, and that the loss of moral knowledge is worse than the loss of factual knowledge, because moral wrongs are worse than factual errors.

On the first point, influential theorists like Strawson (1963) and Frankfurt (1971) have argued that personhood just is moral agency, seeing that moral agency is the critical factor that distinguishes persons from both moral incompetents and simple organisms. This view of personhood has recently been corroborated by research showing that people tend to rate their loved one as “the same person” if they lose their memories, distinctiveness, emotional qualities, and global mental function, but not if they lose their moral capacities (Strohminger & Nichols 2015).When subjects are asked to judge the persistence of moral personality in Alzheimer’s patients, “the extent to which [the] patient seems different is predicted almost entirely by the extent to which their moral traits changed and not at all by their memory loss” (Chituc 2015), contraJohn Locke, but consistent with Strawson and Frankfurt. On the second point (on collective harms), the loss of moral knowledge is arguably more harmful to the community than the loss of factual knowledge because we need moral knowledge more than factual knowledge to live well. The application of factual knowledge to evil purposes gives rise to dystopian scenarios like the Holocaust; a society without cell phones is less horrifying than a society riven by genocide. There is reason to believe, then, that the secondary harms of epistemic injustice might be moral injurious on balance than the primary harms, both individually and collectively. In any case, the primary and secondary harms of epistemic injustice are commensurate, in that they involve a simultaneous loss of epistemic, moral, and siociopolitical standing, respect, and autonomy in victims, and a loss of valuable factual, moral, and sociopolitical resources in the community.

It is worthwhile for moral theorists to note the severe moral harms perpetrated by epistemic injustice. In fact, in neglecting the role of epistemic injustice in the moral community, we fail to adequately frame this system of interpersonal relationships. Strawson (1963) made this mistake when he described ‘the moral community’ as a collective of moral agents who deploy ‘the reactive attitudes’ (e.g., blame, praise) so as to consolidate the community around shared moral values, while deploying ‘the objective attitude’ to exclude moral incompetents. Of course, this is not how society works. Instead, we systematically deploy the objective attitude toward socially marginalized groups on the basis of systemic identity prejudice, in such a way as to ‘other’ and exclude those groups. This increases the perceived moral standing of privileged in-group members, making it easier for them to gaslight, scapegoat, and discredit disadvantaged groups who threaten their position of control. Whereas Strawson cited children and morally deranged people as targets of the objective attitude, in conditions of epistemic injustice this attitude is (illicitly) extended to women, People of Color, and other classic gaslighting victims, who are framed as infantile, incompetent, ‘crazy,’ and corrupt. Strawson’s theoretical resources are useful for understanding idealdeployments of blame and praise, but he does not delve into the distorting role of identity prejudice in our moral economy. This is the new horizon for responsibility scholars.

This topic is explored, amongst other places, in the new edited volume,“Social Dimensions of Moral Responsibilty.”










Refusing to read more in order to read better: on moral and epistemic injustice in academia

Identity prejudice in literature 

In my last post, I wrote about trait criticism as a virtue, and I argued that criticizing well is more important than criticizing more. I also argued that the vice of judgmentalism reflects a distorted critical compass and a fixation on trivial norm violations to the exclusion of more significant ones. I called this disposition ‘critical fetishism.’ The critical fetishist practices criticism to excess on a narrow range of critical targets, becoming too critical relative to those targets, and insufficiently critical outside of that domain. I then argued that critical fetishism is a problem in academic philosophy because it marginalizes non-standard perspectives in  the field, and this translates into discrimination, content bias, and a lack of strong objectivity in academic research (Harding 2015).

Recently, Amy Hungerford (2018) has noted a similar issue in English. She argues that literary scholars should refuse to read more, in order to read better. By declining to read every new publication recommended by the literary press, literary scholars are choosing to be, in a sense, academically irresponsible. They are conscientiously embracing a “distinctively nonscholarly form of reasoning,” deciding whether a book is worth reading prior to reading it. They are, in a sense, judging a book by its cover. This method of literary curation is necessitated by the ever-increasing speed of publishing, which makes it impossible to read every new piece of writing; but it is also a moral imperative: by ignoring the literary press, the canon, and the conventions of the discipline, the conscientious critic makes room in her life (and her discipline) for non-standard content and marginalized voices – content and voices that she would otherwise not have read.

Hungerford explains how literary conventions favour conservativism and white-male-bias. In ‘Modernism/modernity,’ a journal created to promote diversity in literary scholarship,

“A handful of major canonical authors — Joyce, Woolf, Eliot, Stein, Beckett, etc. — continues to preoccupy the journal’s attention while subjects outside that canon fail to create a similarly shared body of criticism. The top 11 authors cited as subjects claim 41 percent of the articles. Most authors not already canonical appear only once or twice each, never achieving the critical mass of scholarship that motivates further study and writing within the context of scholarly careers, let alone further reading by the general public. Such poorly known and rarely taught works are not reissued as their canonical cousins are — in cute new formats, anniversary editions, or as the object of some fresh backlist marketing effort.”

Rather than letting literary traditions and market forces determine which books are worth reading, critics ought to prioritize talented but overlooked and under-appreciated authors and subject matters. Hungerford doesn’t explain precisely how to decide what to read, but one small step in the right direction would be to retain core disciplinary standards while venturing outside of the normal canon, exploring writing by women, people of color, and other underrepresented authors and critics. This is a fairly conservative, but effective, way of diversifying the field of literature and of distributing epistemic respect more fairly. (More radical revisions would be beneficial, but less accepted by the majority; I’m identifying the least controversy method, not the ideal one). Literary conventions essentially create a climate of epistemic and moral irresponsibility in the literary community, which spills over into mainstream society.

Critical fetishism

This connects with my last post, where I argued that “critical fetishism,” or an obsessive focus on a limited range of critical targets, is a vice, whereas critical curiosity is a virtue (inspired by Kate Norlock 2017 and Mariana Alessandri 2018). The world of literary publishing is marred by critical fetishisim, an obsession with a limited range of (predominantly cisgender, white, male) authors and critics and their academic products. Critical fetishism creates a climate of epistemic injustice in which the literary talents and insights of marginalized groups are discredited, and it creates a climate of moral injustice in which historically disadvantaged groups are silenced and oppressed due to systemic identity prejudice. Critical fetishism reproduces mainstream social inequalities in knowledge-producing systems, giving cisgender white males privileged access to the means of production. Thus, they work less hard for more acknowledgment and respect than they deserve.

In my last post, I suggested that critical fetishism is a vice because (1) it sanctifies certain disciplinary subspecializations and discredits others, and it channels attention, money, and respect into canonized subspecializations; (2) it promotes extreme specialization, which may cut off connections between disciplines, and between academic research and community issues; and (2) it can consume a person’s whole life, resulting in ‘alienating saintliness’ (Wolf 1982). This isn’t to say that we shouldn’t specialize, but rather, that specializing to the exclusion of important connections across disciplines and extracurricular boundaries carries epistemic and moral costs. Hungerford shows how the sanctification of canonized authors produces epistemic and moral injustice within the literary community. This is an illustration of the kind of critical fetishisim that I’m describing as a vice – the fetishization of privileged voices to the exclusion of the marginalized.

Hungerford gives a concrete example from her own discipline: David Foster Wallace (the author of the abstruse, 1000+-page tome ‘Infinite Jest’) has been effectively deified within literary criticism, elevated to the status of a literary saint. Hungerford speculates that the marketers of ‘Infinite Jest’ intentionally catered to the intellectual hubris of their target clients, designing a “marketing campaign that appealed to a Jurassic vision of literary genius” (a vision that was a social construct). Rebecca Solnit, too, has called Wallace’s book an example of literary hubris, a conventionally male trait (2017). One thing that men like to mansplain about is esoteric literature, which gives them intellectual cred in the eyes of their academic peers. Women, by contrast, are punished for appearing ‘too smart,’ which is why the job market favours modestly academically successful women but highly academically successful men (Quadlin 2018). In the time it takes to read, analyze, or review ‘Infinite Jest,’ a critic could read several books written by underrepresented authors. This would be ‘academically irresponsible’ in a narrow sense, but it robustly academically responsible, as it would increase epistemic, moral, and social responsibility in academia. There may, then, be an all-things-considered imperative to read less canonized scholarship in favour of exploring more uncharted territory.

Because conventional academic standards are so pervasive within the Ivory Tower – and so imperative for publishing and getting a job – academics tend to feel guilty for not reading the core canon. Hungerford explains how a famous Americanist was reluctant to admit that he had never read Moby Dick; he felt shame over a what he took to be a disciplinary transgression. I don’t think that this is a strange case; rather, it reflects the normal process of academic enculturation, whereby academics internalize the value of the canon and related disciplinary norms so deeply that these norms become part of their personal identity. Questioning disciplinary norms, then, can feel like blaspheme, or a rejection of the self. Implicit bias against non-standard perspectives is probably, in part, a result of this process: the value of the canon is tied up with the value of the Academic Self. This may also partly explain the casualization of academic labour: academics’ over-identification with their discipline makes them highly vulnerable to market exploitation. Being deeply (as opposed to superficially) academically responsible, then, requires not only changes in academic culture, but changes in the inner lives of academics. It requires that we commit what we may perceive as acts of academic irresponsibility: refusing to prioritize canonized scholarship above other types of writing.

Is there identity prejudice in philosophy?

That title is a rhetorical question – of course there’s identity prejudice in academic philosophy. In response to Aleander Rosenberg’s “philosophy is flat” comment, I responded:

“Philosophy is notoriously plagued by pedigree bias (de Cruz 2014), gender bias (Schwitzgebel & Jennings 2016), and racial bias (Botts et al. 2014). Moreover, 35% of philosophy professors are neither tenured nor tenure-track (AAAS Report 2013), and many of these instructors are living in poverty or very close to it. Philosophy is one of the least diverse disciplines in the Humanities (Schwitzgebel & Jennings 2016). Proposed reasons for the lack of diversity in the profession include naivety, conservativism, pride, and hostility to the interests and insights of underrepresented groups amongst the privileged (Kidd 2017).”

Helen de Cruz then wrote a more thorough response to the ‘philosophy is flat’ (‘just world’) theory in the Disability and Disadvantage Blog (2018), with a focus on pedigree bias.

Some of the inequalities in philosophy may be caused by contingent factors, but many are caused by implicit and explicit bias and stereotype threat (Jenkins & Hutchison 2013), which are psychological vehicles for what Miranda Fricker calls identity prejudice (2007), i.e., bias activated by the perception of a person’s demographic attributes, which produces epistemic injustice – a climate that systematically discredits the target’s speech.

 In my last post, I wrote that philosophy is beset by critical fetishism, or a disciplinary obsession with specific objects of philosophical scrutiny. Critical fetishism sanctifies certain philosophers to the exclusion of others, and it has distinctive gendered and racial characteristics. Philosophers have a remarkable fetish for David Lewis, whose work comprises 6 of the top 20 most-cited publications in philosophy (Healy 2013). Of the top-20 most-cited publications, 6.3% are by David Lewis, and 6.3% are by women. That’s right: David Lewis is more cited than all of the most-cited women combined. Saul Kripke is the second-most cited philosopher, and John Searl is high on the list, both of whom have been accused of sexual harassment. According to Jennifer Saul, maybe we shouldn’t be citing sexual harassers at all, let alone citing them more than all of the women in the field (2017). Most philosophers in the profession are also white, so there are racial exclusions as well. I’m not saying that we shouldn’t read and cite canonized philosophers, but that the amount of attention we give them has epistemic and moral costs.

The sanctification of these scholars is an example of critical fetishism, which is, in effect, a kind of epistemic injustice. Critical fetishism is a problem not because it elevates some philosophers above others, but because it elevates philosophers with privileged socioeconomic status above members of historically disenfranchised groups, creating epistemic injustice in the field and tainting the production of philosophical knowledge. This is a moral and an epistemic problem because it discriminates against minorities (a moral transgression) and it undermines epistemic equality and the production of objective knowledge (an epistemic transgression) (viz., Harding 2015). Therefore, critical fetishism is something that we should care about and try to remediate, as scholars and as moral agents. Obviously this is not an easy feat, since it requires a reconsideration of our very identities as academics and a restructuring of the means of academic production, but effectively practicing any virtue is hard. The virtue of critical responsibility is no different.


A colleague pointed out that there is a difference between literary criticism and philosophical criticism, so we can’t treat criticism in both fields as if it were the same. I am using the term ‘criticism’ loosely to denote critical scrutiny of any piece of writing, so any differences across disciplines are immaterial. The main point is that in both fields, analysis is focused on scholarship with disciplinary credibility, and that credibility is partly constructed by conventions within the discipline rather than objective (unbiased) criteria. We should work to ‘un-bias’ the distribution of credibility in all academic fields.

Virtuous blaming/criticizing: Is there such a thing?

  1. Is criticizing a virtue? (Is blaming a virtue?)

In my last post, I commented on Kate Norlock’s paper on criticism (2016), and remarked that blame can take the form of criticism (and that unintelligible blame can be morally valuable). In a recent response to Norlock’s paper, Mariana Alessandri commends Norlock’s insight that “complaining, if rightly understood and employed, is a virtue” (Pea Soup 2018). Alessandri continues, “If we can win this one and convince people that complaining is a virtue, then there would exist, at least theoretically, a phronimos of complaining: the one who knows how and when and in what way to complain. Very few complain virtuously; most fall to the extreme of deficiency or excess” (ibid.). Rather than cautioning people to complain less, Alessandri says that we should complain more, since, “in [her] opinion, we as a society fall into the deficiency category of complaining…; shouldn’t we [therefore] shoot for excess and see where we end up?”

This is an interesting subject for responsibility theorists because blame can take the form of criticism. More specifically, blame, according to Michael McKenna (forthcoming), can involve a criticism about the quality of someone’s action (a deontic judgment) or a criticism about the quality of someone’s character (an aretaic judgment). If blame can contain, or simply be, a criticism, then perhaps blaming, too, can be virtuous. Virtuous blaming, like virtuous criticizing, would be the mean between extremes of deficiency and excess.

In what follows, I will dispute the idea that we should advise people to criticize more, full stop. Instead, I think that we should advise people to criticize better. If criticism is to be virtuous, it must satisfy Aristotle’s conditions of correct (1) frequency, (2) degree, (3) duration, (4) target, and (5) activating circumstances. These conditions produce epistemically responsible criticism. Arguably, epistemically irresponsible criticism is a bigger problem than insufficient quantity of criticism. I note that the term ‘critical’ is ambiguous, and can denote either the capacity for criticism or the tendency to criticize. Following Michael Brownstein’s treatment of self-control [2018]), I argue that the capacity for criticism cannot be in excess, but criticalness (or ‘trait criticism’) can be – that is, one can be too disposed to exercise one’s critical capacities. A ‘viciously judgmental’ person, as described by Gary Watson (2013), is an example of a vicious critic. But is this person too critical in general, or simply too critical about certain things? Although the vice of jugdmentalism may seem to reflect excess trait-criticism, I argue that it may actually represent a deficit of critical priorities and a corresponding fetish for trivial norm violations. An example of a critical fetishist is someone who complains excessively about white-person-problems and doesn’t care about racism – racism isn’t even on the person’s critical radar. The fetishist doesn’t complain too little; he complains too much, but about the wrong things. He has a domain-specific excess of trait criticism, and, at the same time, a domain-general deficiency, because his critical field of vision is too narrow. Even if his domain-specific criticisms are valid, they’re the wrong things to care about. What the fetishist should do is not increase his output of criticism, but rather expand his critical horizons. Once the fetishist has achieved a more balanced life, he will be in a position to criticize in the right way and to the right extent. Lastly, I argue that if someone can be too trait-critical, then she can also be too trait-morally-responsible (meaning too disposed to express vicious moral criticisms). And I make the case that academic philosophy may be suffering from a case of domain-specific trait criticism.

Here’s a break-down of the entire argument. In section 2, I’ll argue that the main problem with our society is not insufficient criticalness, but epistemically irresponsible criticism, especially critical fetishism; in section 3, I’ll argue that trait-criticism, but not capacity-criticism, can be in excess; in section 4, I’ll outline Brownstein’s account of self-control for illustrative purposes; in section 5, I’ll apply this account to criticism; and in section 6, I’ll argue that academia’s obsession with effortful, educated criticism is pernicious because it downplays the epistemic value of experiential (effortless) criticism.

2. To criticize or not to criticize – is that the question?

Alessandri advises us (tentatively) to criticize more often, because criticism is ‘deficient’ in our society. This argument (I think) treats deficiency as a quantitative state: as too little of something. Deficiency, however, can also be qualitative, such as when we criticize irrationally or without grounds. Imagine someone named Wohn Jayne who hardly ever complains. When WJ does complain, though, his complaints tend to be sexist. WJ thinks of himself as a “man of few words,” someone who doesn’t like to “whine” because “whining is effeminate” (in his mind). As a result, WJ expresses criticism sparingly, but when he does complain, his complaints tend to be sexist. WJ doesn’t strike me as someone who complains too much, although, given his sexist disposition, one could argue that any complaint from him is one complaint too many. But the real problem with WJ is not so much the quantity of his criticisms as their quality. If a lot of people in our society are like WJ (i.e., biased), then it’s not clear that we should encourage them to complain more, seeing that their complaints are, on balance, epistemically worthless and socially harmful. Biased speech, in effect, undermines the ideals of deliberative democracy by disseminating misinformation, undermining discursive parity, and reducing receptiveness to diverse contributions (Fishkin 2009). Similar considerations apply to other types of biased criticism (racist, xenophobic, etc.). So, increasing the quantity of epistemically irresponsible criticism is not desirable.


A person can complain too much even if her complaints are generally well-founded. Imagine someone named Academic who only criticizes people for valid reasons, but criticizes people constantly, even for the most trivial infractions. Her criticisms are annoying and intrusive and they alienate other people. Academic also turns her critical lens on herself and develops a depressive mindset, which undermines her wellbeing. Even if Academic’s criticisms are valid, they don’t improve the moral ecology.

It’s difficult to show in the abstract that we should promote more criticism, quantitatively speaking. It’s less controversial to say that we should promote better criticism. That said, one might think that practicing criticism refines one’s sensitivity to criticism-relevant factors (or salient norms), in which case, more criticism leads to better criticism. (Practice makes perfect, as they say). On the other hand, practicing hateful criticism can lead to more effective antagonism, so practicing criticism can also cultivate viciousness. Online trolls practice vicious criticism all the time, and in doing so, hone their skill for antagonism and needless provocation. Some people become addicted to (or at least enjoy) trolling because it triggers their reward system, making it hard to stop. This suggests that the practice of criticism is not necessarily something to be promoted.

Alessandri is right, however, that most people are complacent in the face of severe moral infractions. But these same people most likely complain about petty annoyances, like monthly subscription costs, property taxes, and other people’s grammatical errors (McInnes 2012). That is, they don’t complain too little; they complain about the wrong things. As a retired lieutenant colonel for the U.S. Airforce lamented in a recent article, most Americans don’t know how many wars the U.S. is waging in foreign countries at this very moment (Astore 2018), but they can probably tell you when the price of a latte goes up at Starbucks. The U.S. government is invested in waging a “forever war” in the Middle East while increasing GDP by promoting consumer excess and political ignorance at home, and most Americans either don’t notice or don’t care (ibid). People who are insensitive to moral infractions but hypersensitive to minor annoyances shouldn’t complain more; they should refine their critical faculties so that they can complain better.

3. Can criticizing be a virtue? 

Ultimately, the question of whether criticizing is a virtue depends on how we conceive of critical faculties. (Likewise for blaming and blaming faculties). There are two dominant models of human faculties: the capacity model and the trait model. Is criticizing a capacity or a trait? The difference between the two models if significant. If criticizing is a capacity, like literacy, then one can never be too critical (just as one can never be too literate). Capacities are typically described as a sensitivity to relevant factors: The capacity to play the violin, for example, is described as Neil Levy as a patterned sensitivity “to the contours of strings, the pressure [the violinist’s] other hand is exerting, the volume of the orchestra (if relevant), and so on.” Levy uses violin-playing to illustrate the capacity for moral responsibility. Analogously, the capacity for criticism would be a patterned sensitivity to criticism-relevant factors or critical norms (and norm violations). This capacity can be perfected, but can never be in excess, just as a violinist can play too much, but never too well. To give another example, no one has ever complained that Gauguin painted too well, but people have complained that he painted too much, to the neglect of his familial responsibilities (Williams 1981). His obsession with art was a vice.

So, if criticizing is a trait, like agreeableness, then a person can have an excess of criticalness (or ‘trait criticism’), just as one can have an excess of agreeableness. High agreeableness, according to research, correlates with susceptibility to groupthink, exploitation, and alienation (Wong 2017) Virtuous agreeableness, then, is the sweet spot between naivety (groupthink, gullibility) and curmudgeonliness (antisociality). On Aristotle’s view, excesses of traits are vices: for example, an excess of courage is rashness; an excess of temperance is insensibility; an excess of friendliness (similar to agreeableness) is obsequiousness (see this chart by Thompson 1955 for a complete list). An excessively courageous person rushes into battle without forethought; an excessively temperate person fails to enjoy the good things in life, etc. In this way, vices tend to be bad for the person and the community; they impair individual and group flourishing. If criticalness is a trait, then it, too, can be in excess, in which case it would presumably be something like captiousness, pettiness, or ‘jugdmentalism.’ In a paper on blame, Gary Watson cautions us to avoid the “vice of judgmentalism,” which is the tendency to judge others uncharitably and without generosity (2013). The vice of judgmentalism is, in effect, an excess of trait-criticism. The judgmental person is disposed to judge too often, too harshly, too severely, without cause, in the wrong circumstances, or all of the above.

Even if someone has a well-honed capacity for criticism, she can exercise that capacity to excess, becoming too critical. Academic, for example, has a refined critical sense, but exercises that capacity gratuitously. WJ has a defective critical capacity, and exercises that capacity poorly. A virtuous critic has a refined capacity for crticism and exercises that capacity well, to the right extent, and at the right times. Aristotle says that a virtue satisfies five criteria that lie on a continuum. They are: (1) frequency, (2) degree, (3) duration, (4) target, and (5) activating circumstances (Losin 1987). Prima facie, Academic criticizes (1) too often, (2) too severely, and (3) for too long, whereas WJ criticizes (4) the wrong targets in (5) the wrong conditions (i.e., women, all the time). On scrutiny, however, Academic and WJ are both deficient across all of (1)-(5), since both are fixated on a narrow set of norms. Academic is too responsive to (4) academically-salient targets and (5) activating conditions for academic criticism, which is why she criticizes excessively vis-a-vis these targets. WJ, on the other hand, criticizes stereotypically feminine targets (1) too often, (2) too severely, (3) for too long, and for no reason. While both critics are too fixated on certain critical norms, WJ is the more vicious critic because his criticism are utterly baseless and misogynistic, whereas Academic at least gets academic criticism right. What they share in common is a narrow range of critical interests.

Usually, when someone is ‘too critical,’ the person has skewed priorities. Although Angela Davis is very critical of the U.S. military-industrial complex, the target of her criticism is evil, so it’s hard to imagine how her criticism of it could be excessive. Davis also frames this criticism as part of a broader analysis involving global vectors of injustice. In her view, a systems analysis that encompasses a network of intertwined political injustices is the correct method of critical analysis, so a fixation on any particular political injustice in isolation is methodologically flawed and criticizable (2016). Davis specifically criticizes our tendency to focus narrowly on Western politics, and to ignore links between local and international political systems. In other words, she argues that a narrow (local) critical focus can be a critical failing, or a critical vice. A virtuous critic has a broad range of interests.

Interestingly, research shows that criticism can be harmful in certain cases. For example, one study finds that self-criticism, in contrast to perfectionistic striving, correlates negatively with self-efficacy and aspiration level (Stoeber et al. 2008). Another study finds that online partisan criticism increases affective polarization between groups (Suhay et al. 2017). These studies don’t show that criticism is inherently negative, but that certain types of criticism (e.g., partisan), in certain contexts (online), can be harmful. Notably, the research highlights generalizations, not universal truths. In some cases, affective polarization could be beneficial, in which cases partisan criticism is virtuous. Virtues are context-specific, meaning that their normative valence depends on the context. Rushing into battle is rash, unless it is part of a rational plan of action, in which case it is courageous. The “thick” description of the action (as a virtue) is partially specified by the context (Burton 1992). For this reason, acting virtuously requires practical wisdom, including sensitivity to the demands of the situation; there is no rulebook for virtuous conduct that would eliminate the need for lived experience redundant (McDowell 1979). Criticizing virtuously, then, requires a range of experiences.

In sum, to avoid the vice of critical fetishism, we should expose ourselves to a variety of norms, as opposed to fixating on any insular set of norms (e.g., Western political systems, analytic epistemology, norms pertaining to scholarship on William Coleridge). Exposure to various domains increases our sensitivity to a range of domain-specific norms, which cultivates domain-general trait criticism (within limits). There will likely be a trade-off between domain-specific competency and domain-general competency, and we should seek the right balance between the two. Being too specialized is just as bad as being too unfocused.

4. Self-control

Michael Brownstein discusses the capacity-trait distinction in his recent paper on self-control, in which he disputes the accepted wisdom that there is no such thing as too much self-control (2018).* I think that there are interesting parallels between trait self-control and trait criticism, which I will examine here and in the next sub-section.

Brownstein says that, while it may be impossible to have an excess of self-control as a capacity, this is not true of self-control as a trait. Researchers nonetheless tend to describe trait self-control as unequivocally positive, neglecting this distinction. For example, Tagney et al. (2013) say that trait self-control correlates positively with achievement and task performance, impulse control, psychological adjustment, moral emotions (specifically, propensity to feel guilt and shame), productivity, and other putatively “positive” outcomes. Brownstein objects that these are not necessarily positive outcomes, and I have raised similar concerns elsewhere. Empathy, for example, can increase out-group bias and shadenfreude toward perceived out-group members’ pain, potentially escalating intergroup conflict (Cikara et al. 2012, cited in Ciurria 2017); shame and guilt-related distress are associated with clinical PTSD (Beck et al. 2011); people high in productivity experience excessive self-evaluation and high self-control regret, i.e., regret over lost opportunities for enjoyment (Brownstein citing Keinan and Kivetz 2006). Excess self-control, as described by Brownstein, sounds a lot like Aristotle’s description of “insensibility” as an excess of temperance, characterized by an inability to enjoy the good things in life. Aristotle, in a sense, prefigures Brownstein’s critique of the scientific view of self-control as unequivocally positive. Any trait, on Aristotle’s view, can be a vice if expressed too strongly or too mildly, in the wrong contexts, or in response to the wrong targets.

More interesting still, Brownstein notes that self-control can have mixed results for people with low socioeconomic status (SES): for example, “for low-SES Black teenagers, high trait self-control predicts academic success and psychosocial health, but at the expense of epigenetic aging (i.e., a biomarker for disparities between biological and chronological aging)” (citing Gregory Miler et al. 2015). To interpret academic success and psychosocial health as a net positive outcome would be to impose a privileged frame of reference on members of socially disadvantaged groups, which is a type of epistemic injustice. Accordingly, Brownstein rejects the view that self-control is unequivocally positive, which is (in effect) an erasure of the lived reality of socially disadvantaged people, who do not derive unqualified benefits from the exercise of self-control. Promoting the exercise of self-control without qualification stigmatizes members of socially disadvantaged groups who (rationally) choose not to exercise self-control because they value their health more than their grades. The accepted scientific wisdom falsely presents this as an irrational choice and a failure of self-control, instead of a rational exercise of self-control, which accurately weighs the pros against the cons.

Brownstein gives a good example of when ‘uncontrol’ (or the failure of self-control) could be admirable: if his children become so enraged about slavery during history class that they cannot concentrate, no matter how hard they try, this would make him proud. (I share this sentiment entirely). Brownstein also remarks that spontaneity, which only occurs in the absence of self-control, can be admirable and conducive to flourishing. Thus, self-control may be good in some cases, but it is not an unequivocal good.

5. Criticism again

The capacity-trait distinction can be applied to criticism as well. A virtuous critic (call her VC) is reliably sensitive to criticism-relevant factors (norms), i.e., she has the capacity to detect norm violations. But she also exercises that capacity virtuously. VC, that is, has trait criticism: she is disposed to exercise her critical faculties in the right way, to the right extent, and at the right times. Although VC is sensitive to a broad range of criticism-relevant norms, she doesn’t express criticism in response to every perceived norm violation, as doing so would be captious, petty, ungenerous, and perhaps epistemically arrogant. If VC’s friend is distraught about a break-up and texts her, “I’m devistated right now,” VC doesn’t text back, “I’m sorry; p.s., it’s ‘devastated,’ not ‘devistated’ (sic).” If VC fails her dissertation defense, she doesn’t ruminate over everything she could have done differently and berate herself for months one end, as this would serve no purpose. On the contrary, she might try to forget about what she did wrong by relaxing her critical faculties vis-a-vis this particular disappointment. It is reasonable to think that a temporary or partial suspension of criticalness may sometimes be the best course of action.

VC also prioritizes severe infractions over petty infractions. Rather than hyper-focusing on trivial annoyances, she turns her attention to severe moral infractions. She doesn’t complain if the barista at Starbucks gets her order wrong, but she is outraged about racist incarceration norms in the U.S. prison system. She is sensitive, in other words, to the priority ranking of various criticizable offenses, and focuses on the more severe violations.

Practicing virtuous behaviour, according to Aristotle, cultivates virtuous traits. This may seem to suggest that we can’t practice criticism too much. But this isn’t true. One can practice any skill to excess. In sports research, there’s evidence that over-training could harm one’s career prospects by increasing the risk of injury, and training above a certain number of total hours (10,000) may not increase performance (Reider 2017). Athletes, moreover, can become obsessed with their sport to the exclusion of worthier pursuits, like connecting with one’s community. After many years of dedication to tennis, Andre Agassi became so disenchanted with his life as a tennis player that he started self-medicating with crystal meth (AP 2009). Although practicing criticism doesn’t cause sports injuries and probably won’t lead to crystal-meth addiction, it could foster a hypersensitivity to perceived norm violations, and thus a propensity for jugdmentalism. A critic could also practice criticism redundantly if there is an upper limit on net gains, in which case the person’s criticalness could crowd out worthier pursuits, such as spending time with friends. (Notably, criticizing friends too much could undermine one’s friendships, and reveal a lack of the normal optimistic bias that extends to loved ones; thus, some degree of un-criticalness, or generosity, towards friends may be a virtue).

On the other hand, what appears to be an excess of trait criticism may actually be a narrow range of critical interests. Take a hypothetical graduate student, Jack, who is obsessed with critically analyzing the poems of Samuel Taylor Coleridge. Jack, say, spends years cultivating Coleridge-relevant critical sensitivities to the exclusion of more important critical sensitivities, such as responsiveness to moral violations. In the end, Jack is too critical of Coleridge’s work (and meta-analysis), and he also (partly for that very reason) is not sufficiently critical of non-Coleridge-related norms, which don’t interest him. In the non-academic world, Jack might be a jerk or socially incompetent due to a lack of critical sensitivity to social cues and moral norms, and basically any other norms unrelated to 18th Century literary criticism. A domain-specific critical fetish, in other words, may correspond to a lack of domain-general critical competency in a person’s cognitive architecture. Hence, what may at first appear to be a general excess of trait criticism (e.g., a judgmental personality) may be a deficiency of domain-general trait criticism (indifference to a plurality of norms), due to an obsession with a certain topic. We should not advise Jack to be more critical (without further specification), as he would likely take this as a cue to probe even deeper into Coleridge scholarship. What we should advise Jack to do is try to be less trait critical of Coleridge, and more trait critical of virtually any other subject matter, especially those that affect his capacity for democratic citizenship. If Jack is an excellent scholar but bad citizen, this is a sign of a fetish.

Another problem for Jack is that, if for some reason he cannot pursue a life of Coleridge scholarship (say, because he doesn’t get tenure), he will end up, as Susan Wolf puts it, “bankrupt,” or incapable of living a meaningful life by his own lights (2007). Wolf doesn’t say that the bankrupt person was living a meaningless life from the start, but she does say that an obsession with a domain-specific set of norms, to the exclusion of other values, is indicative of an impoverished life (1982). A “moral saint,” obsessed with cultivating saintliness, is not living a flourishing life in her view. If the moral saint isn’t living a flourishing life, then surely the Coleridge fanatic isn’t. But the Coleridge fanatic also isn’t living an ethical life, insofar as he is neglecting citizenship norms for the sake of a self-interested critical project. Jack is too trait critical of Coleridge scholarship partly because he lacks the disposition to cultivate other critical interests, which is a vice. (Essentially, critical fetishism and domain-general critical incompetency are two sides of the same vicious coin). The solution to this problem is to cultivate the disposition to explore other critical domains, even at the expense of one’s integrity or psychic integration (Arpaly 2000). Jack might transcend his limitations purposefully, or do so against his better judgment, in a moment of akrasia. Either way, he’s better off.

6. Moral responsibility

Moral responsibility is often described as a sensitivity to moral criticism. Can a person be too morally responsible, then? Not if responsibility is a capacity, but perhaps it’s possible if responsibility is a trait, i.e., a disposition to respond to moral criticism. If you internalize criticism to the point of pathological guilt, or criticize others to the point of psychologically injuring them, then perhaps you are too trait morally-responisble, and you would would do well to relax your critical faculties. On the other hand, your disposition to criticize people to excess may reflect an insensitivity to more important norms. If so, then you are not too trait-responsible, full stop, but too trait responsible relative to a particular normative domain – specifically, one with trivial inclusive norms. Jack is too trait critical when it comes to analyzing Coleridge because he is obsessed and doesn’t value other things. Similarly, a moral fetishist could be obsessed with criticizing people’s trivial faults (e.g., unpunctuality), but disinterested in more significant moral violations (e.g., racism), because the person’s moral compass is broken. It is, of course, easier to criticize trivial and illusory norm violations in the conditions of epistemic injustice that we live in. The upshot is that a person can be too trait morally-responsible relative to a particular domain, but that narrow critical specialization most likely reflects a deeper trait-irresponsibility – an indifference to other normative considerations. A robustly trait-responsible person, it seems, lives a well-rounded life (Wolf 1987), one that exposes her to a broad range of norms. A moral fetishist lacks curiosity about other people’s normative concerns, and focuses too intently on norms that affect her own life.

If a person is too trait-responibsle in a particular domain, then akrasia, or weakness of will, may be a virtue in that person. For example, if Jack were to quit studying early and go out with friends, then what he perceives as a lapse in responsibility (akrasia) would actually be virtuous, and would facilitate a more domain-general responsibility, viz., a sensitivity to non-Coleridge-related norms. Akrasia (much like ‘uncontrol’) can ostensibly be positive, in spite of its negative connotations. Nomy Arpaly persuasively argues that “inverse akrasia” (2014) in which someone does the right thing against her better judgment, is a virtue, and an expression of responsible agency, as it reflects the person’s deeper values. This confirms that what may at first blush appear to be a lapse of responsibility –  and is a lapse of domain-specific trait responsibility – may also be an expression of a deeper kind of responsibility: domain-general responsibility. The inverse akratic may be spontaneously sensitive to her (possibly subconscious) reasons to stop being so domain-specifically responsible in a moment of akrasia. Acting ‘irresponsibly’ in one domain, then, may represent a a deeper, or more domain-general, responsibility – a (possibly repressed) disposition to explore alternative normative spaces. Being curious about other people’s lives and concerns may be the best defense against ‘shallow responsibility,’ the fetishistic exercise of the capacity for responsibility on a narrow set of problems.


Screen Shot 2018-03-16 at 5.33.57 PM.png


7. The politics of criticism

Does the unqualified promotion of critical thinking have negative political consequences, similar to the unqualified promotion of self-control? I think that it does. Consider the over-intellectualization of knowledge as represented in Enlightenment philosophy, most notably, Descartes’ “Meditations.” Descartes believed that everyone has equal access to knowledge by virtue of reflective deliberation. This is just false. As standpoint epistemologists have pointed out, knowledge is situated (viz. Harding 2015). Therefore, no one has unfettered access to knowledge; knowledge is a function of the synthesis of our situated experiences, under conditions of epistemic justice in which uptake (or epistemic respect) is equally distributed. Knowledge, then, is a hypothetical ideal, given that members of historically disadvantaged groups are marginalized by asymmetrical power relations. Critical thinking, in the Cartesian (non-situated) sense, is still lauded by universities, and while I wouldn’t deny that it is a valuable skill, the inclusion of marginalized voices in epistemic communities is equally valued, but under-appreciated. Universities, in effect, overvalue critical thinking skills (which they can market), while downplaying the epistemic vale of diversity. Both are necessary for knowledge, but they are not equally touted as academic values. Philosophy is a notable example: cisgender white males occupy a majority of tenure and tenure-track positions. This climate doesn’t maximize the production of objective knowledge; in fact, it fosters critical fetishism about historically male perspectives. This is a critical vice within the profession.

Trait criticism, in sum, is not simply, or even primarily, the product of deliberative effort; it is also the natural byproduct of effortless life experiences. People with low socioeconomic status, for example, are more sensitive to the value of low trait self-control than are scientists, who experience self-control as an unqualified good, and inscribe this experience on their research. Although trait criticism can be enhanced by practice, it is also conferred effortlessly by lived experience. There are, in effect, two types of trait criticism: ‘practiced trait criticism’ and ‘experiential trait criticism.’ It is dangerous to promote critical thinking as a practical exercise (i.e., ‘practiced trait criticism’), without promoting, at least to the same extent, experiential trait criticism. In other words, we should not advise universities to produce more critical work, full stop, without also advising them to promote diversity at all ranks of every department. Thus, as with individuals, universities should not simply produce more criticism – they are already producing too much domain-specific criticism, arguably; but they should produce better (viz., more epistemically responsible) criticism. Increasing the quantity of criticism without increasing its quality will only normalize and reinforce existing biases.


*You can read Brownstein’s whole paper here. 

Scapegoating: Blame gone wrong

  1. Scapegoating: false and excessive blame and punishment for perceived norm violations

Scapegoating is the practice of blaming and punishing innocent victims for purposes of expediency and/or political gain. Blame is a contested term, but I take it to denote negative and avoidant interpersonal practices, such as resenting, sanctioning, distrusting, excluding, and marginalizing people. Punishment also includes negative and exclusionary (albeit institutionalized) practices, such as incarcerating, disciplining, and isolating people. Depending on your theoretical framework, punishment may or may not fall under the rubric of blame; punishment may be an institutionalized version of blame, though with additional constraints determined by relevant institutional considerations, or it may be an entirely separate practice. I agree with Wallace (1996) that blame includes a range of criticizing and sanctioning responses, including, “at the limit,” punishment (1996: 54). Some reasons for thinking that punishment is part of the blaming system is that it is part of a system of relations in which we assign responsibility, and it is amenable to the same cognitive distortions, including implicit biases. Scapegoating as a type of epistemically irresponsible blame, then, can occur in carceral and extra-carceral systems.

America is one of the most socioeconomically stratified and demographically unequal countries in the developed world. As a result, historically disenfranchised groups, including African American, Hispanic, and LatinX people, women, and people with mental illness, are systemically scapegoated. That is, members of these groups are blamed and punished for norm violations they did not commit, or excessively blamed and punished for relatively insignificant norm violations, on the basis of oppressive cultural stereotypes and social scripts. Scapegoating reinforces existing asymmetries of power and protects the positive self-conception and socioeconomic status of privileged groups.

In this post, I will explain how America’s extreme socioeconomic inequality harms Black and LatinX Americans, women, and people with mental illness*, making them vulnerable to scapegoating (section 2). In section 3, I will explain more specifically how scapegoating practices harm and vilify these groups across a range of social institutions. In section 4, I will define scapegoating as not just a type of misdirected blame, but a type of epistemic injustice with the characteristic feature of vilifying marginalized groups by blaming them for society’s real or imagined problems, and I will outline some of the harms perpetrated by this type of injustice. In sections 5 and 6, I will explain who is responsible for scapegoating, on two different theories of responsibility (the indirect-control view and functionalism). I hold that cognitively functional Americans are generally responsible for scapegoating on both views, though there may be some variation in degrees of blameworthiness depending on the circumstance; and I say that, on both views, scapegoaters can be blamed for both the harms and the contents of overt scapegoating actions.

2. Socioeconomic Inequality

Scapegoating is a common practice in modern society, in part because modern society is characterized by conditions of severe socioeconomic inequality and demographic stratification. The United States has the fifth highest level of income inequality (measured by household disposable income per year) of all OECD countries, behind only Turkey, Chile, Costa Rica, and South Africa  (OECD 2017). The top 1% of Americans control 38.6% of the nation’s wealth – almost twice as much as the bottom 90% combined (Egan 2017). Poverty does not affect every social group equally; it disproportionally affects historically disenfranchised groups. The real median income of non-Hispanic white households is $65,041, compared to only $47,675 for Hispanic-origin households and $39,490 for Black households (Semega et al. 2017). Income inequality, moreover, pales in comparison to wealth inequality: ‘white households in the middle-income quintile (those earning $37,201-$61,328 annually) own nearly eight times as much wealth ($86,100) as middle-income Black earners ($11,000) and ten times as much wealth as middle-income Latino earners ($8,600)” (Asante-Muhammad et al. 2017). That is, within the same income bracket, Black and LatinX earners own much less wealth than white earners.

Income also varies by gender: Women as a group earn just 80% as much as men, but Hispanic/Latina women and African American women, respectively, earn only 54% and 63% on the dollar, compared to white women’s 79% and Asian women’s 87% earnings (AAUW 2017).

These are some of the major populations studied by economists, but it leaves out many disadvantaged groups, including people with mental illness. Higher economic inequality is linked with higher national rates of mental illness (Pickett et al. 2006), and mental illness is highest amongst low-income families (McSilver Institute 2014). Research suggests that poverty is not just the result of disability due to mental illness, but a significant cause of mental illness (ibid). The United States has the third-highest disease burden due to mental illness of all WHO member states (after only China and India) (McPhillips 2014, WHO 2017), and ranks 37th for access to healthcare services – higher than Slovenia, but below Costa Rica (TPF 2018). 56% of Americans currently lack access to mental health treatment, with worse access in states that chose not to expand Medicaid (MHA 2017).

This isn’t an exhaustive list of socioeconomically disadvantaged demographic groups, of course, but it is a suggestive compilation of readily-available economic data. (I don’t have time to address further intersections of oppression here). The specified disadvantaged groups – Black, LatinX, Hispanic Americans, women, and those with mental illness – are victims of systemic scapegoating. That is, they are systemically blamed for illusory and trivial norm-violations because they are easy targets (due to low socioeconomic status, intergenerational trauma, etc.), and because blaming these groups reinforces the existing colonialist, patriarchal, ableist social order. Blaming members of disadvantaged groups for norm violations that they didn’t commit maintains the status quo, reinforces oppressive social narratives, and protects the socioeconomic interests of  the historically privileged.

3. The cultural scapegoating of socioeconomically disadvantaged groups

Here are some examples of systemic scapegoating.

The United States, which contains 21% of the world’s prison population – more than China (APA 2014) – incarcerates African Americans at almost five times the rate of white Americans, incarcerates twice as many Black women as white women, disproportionally arrests Black children, and has a majority Black and Hispanic prison population (56%), even though these groups make up only 32% of the U.S. population (NAACP 2018). Implicit racial bias, structural disadvantages, and racial profiling lead to high levels of incarceration amongst African Americans (TSP 2016). In terms of postsecondary education, African Americans have a 47.1% graduation rate and Hispanic/LatinX Americans have a 56.5% graduation rate at public 4-year colleges, compared to a 64.4% graduation rate for White students (Imagine 2008). This is in part because African American and Hispanic/LatinX students receive disproportionate discipline referrals (controlling for socioeconomic status), resulting in more suspensions and expulsions (ibid., Wallace et al. 2005). This is due in part to implicit racial bias and stereotype threat. Of students classified as aggressive, African Americans are more likely to be disciplined than any other student group, especially by white teachers (Horner, Fireman, & Wang, 2010; KITSRE 2018). (Same-race teachers judge Black students’ classroom behaviour more favourably than do white teachers). Students with a ‘black walking style’ are perceived by teachers as lower in academic achievement, highly aggressive, and likely to be in need of special education services (Neal et al. 2003). Teachers have lower expectations of Black students than other student groups, resulting in expectancy effects and stereotype threat that harm their academic performance (McKown & Weinstein 2002). Black girls are seen as more adult-like and less innocent than their peers, resulting in harsher and more frequent punishments by educators and school resource officers (Epstein et al. 2016). These effects converge in scholastic system of relations in which “less praise” and “more disciplinary action” is taken against Black students (KITSRE 2018).

Women don’t suffer incarceration rates equal to Black and Hispanic/Latino men, but they suffer higher rates of sexual violence: 90% of adult and 87% juvenile rape victims are female (RAINN 2018) – and women are often blamed for being raped, especially by people high in rape-myth acceptance and implicit gender bias (Grubb & Turner 2012). This helps to explain why only 6 out of every 1000 rapists go to prison, and most rapists are never reported (RAINN 2018). In the criminal justice system, female expert witnesses tend to be seen as more credible in civil cases than criminal cases, possibly because criminal litigation is stereotypically male (Larson & Brodsky 2010; Couch & Sigler, 2002; Jones et al. 2014). This implies that credibility in court depends on salient gender stereotypes. In postsecondary education, female teachers receive lower scores on Students Evaluations of Teaching than male teachers across almost all disciplines, controlling for student learning outcomes (Flaherty 2016; Flaherty 2017), which suggests that women are subject to harsher criticism and resentment from students than male teachers on average. In the workplace, women who exhibit leadership skills are seen as ‘bossy’ and ‘less effective’ than men (Kramer 2016). Women can’t just transfer into a more lucrative (historically male) profession on mass, because when the share of women in an occupation increases, the occupation is devalued and pays lower wages (Levanon et al. 2009). In short, women are blamed and punished more often than men when gender stereotypes are salient, including in criminal court, higher education, and corporate America. Women are seen as less praiseworthy, and more blameworthy, in their capacity as court witnesses, university professors, and workers in historically male fields. (I should note that I used statistics about ‘female’ vs. ‘male’ expert witnesses, teachers, and workers, because of the availability of the data; it is a reasonable conjecture that trans women face the same discrimination, plus transphobia, in historically male workplaces).

Next, people with mental illness tend to be incarcerated rather than being provided with mental health services, but a majority of mentally ill prisoners are not violent offenders (NAMI 2018). Nonetheless, people with mental illness are systemically scapegoated for America’s culture of mass shootings. After the 1999 Columbine shooting, psychiatrist Peter Breggin blamed mentally ill people; after the 2012 Newtown shooting, psychiatrist E. Fuller Torrey blamed ‘mentally ill subgroups’; and in 2008, the U.S. Supreme Court endorsed prohibitions on gun ownership for people with mental illness, on the assumption that there is a correlation between mental illness and homicide (AJPH 2014). There isn’t. Research shows that “fewer than 5% of the 120 000 gun-related killings in the United States between 2001 and 2010 were perpetrated by people diagnosed with mental illness” (ibid), while about 20% of American adults have a diagnosable mental illness (Insel 2015). In other words, having a mental illness predicts not committing a mass shooting. People with mental illness are more likely to be assaulted than to commit assault (AJPH 2014), but they are stereotyped as ‘dangerous,’ ‘violent,’ and ‘criminal.’

Factors that do predict gun violence include gun availability and social relations: up to 85% of shootings occur within social networks (Papachristos 2012). There is also a positive correlation between mass shootings and domestic violence: while “perpetrators of domestic violence account for only about 10 percent of all gun violence, they accounted for 54 percent of mass shootings between 2009 and 2016” (NPR 2017; citing Everytown 2017). Psychiatrists are not effective barriers to mass shootings, as they are no better than laypeople at predicting whether a patient will commit a violent crime (Steadman 1978). As Jonathan Metzl clarifies, there is no psychiatric diagnosis that includes gun violence as a symptom; hence, when it comes to mass shootings, there is no “predictive value to psychiatric diagnosis” (Metzl on NPR, February 18, 2018).

Mental illness is diagnosed three to four times more often in Black and Hispanic/LatinX service users than in white service users, possibly due to clinical racial bias, differential access to healthcare, and different attitudes toward mental healthcare (Schwartz et al. 2014). This suggests that mental illness may not have robust construct validity, let alone predictive value. In any case, the majority of mass shooters are white males (54%) (Foleman et al. 2017), and most do not have a diagnosed mental illness. Salient predictors of gun violence, then, do not include having a mental illness; they having access to guns, knowing the victim, having a record of domestic violence, and being a white male. If there is a predictively valid stereotype of a mass shooter, then, it is a normal white male.


In sum, racialized minorities, women, and people with mental illness tend to be falsely and excessively blamed and punished due to the saliency of cultural stereotypes.  These patterns of blame and punishment reinforce patriarchal, colonial, and ableist social scripts.

4. Scapegoating as epistemic injustice 

Scapegoating reinforces oppressive social scripts rooted in America’s colonial-patriarchal history. Black and Hispanic/LatinX Americans are disproportionally blamed and punished for perceived norm violations and perceived suberogatory performance in judicial and educational contexts; women are unfairly blamed for perceived subordinate performance in stereotypically male judicial, educational, and corporate contexts; and people with mental illness are unfairly blamed and stigmatized in the wake of mass shootings.

In his influential work on responsibility, Manual Vargas argues that our capacity for moral responsibility is influenced by the availability of “narratives, scripts, or cultural frameworks,” which comprise our “moral ecology” (2013: 246). Similarly, Jose Medina argues that our capacity for responsibility depends on the availability of social scripts, narratives, and discourses, which comprise our shared “social imagination” (2012). Our sensitivity to people’s moral and epistemic traits, on these views, is conditioned by salient social scripts – for example, scripts about putative associations between race and criminality, gender and credibly, and mental health and violence. These scripts are rooted in historical asymmetries of power, and they normalize and reinforce these asymmetries. Blaming disadvantaged groups reinforces the very scripts that oppress them.

Scapegoating targets socially marginalized groups because these groups are vulnerable, socioeconomically, politically, and epistemically. Therefore, they are easy targets. Perhaps the best way of framing scapegoating as a type of harm or injustice is to see it is as a kind of epistemic injustice, which involves giving someone a deflated credibility rating on the basis of identity prejudice (Fricker 2007). When we scapegoat someone by framing the person’s group identity or visible demographic attributes through dominant social narratives as a type of failing or liability, we are harming the scapegoating victim’s epistemic standing within the community – that is, we are committing testimonial injustice (Fricker 2007: 1). Similarly, when we fail to refute false stereotypes about marginalized social groups, we are withholding pertinent epistemic resources from those groups and normalizing oppressive scripts, which is what Alyssa Cirne calls “willful hermeneutical marginalization” (2012: 46) –  a second type of epistemic injustice. Both types of epistemic injustice are perpetrated by people with epistemic deficits.

The distinctive characteristic of scapegoating as a type of epistemic injustice is that scapegoating vilifies an epistemically vulnerable group by framing their identities and experiences through the dominant framework(s) of the privileged, and therefore frames their identities and experiences as essentially morally corrupt. As Gaile Pohlhais, Jr. (2014) describes epistemic injustice, its “primary harms” involve “othering” members of marginalized groups, specifically for purposes of “maintaining epistemic practices that make sense of the world as experienced from dominant subjectivities, but [does not grant the ‘othered’ individuals] the same epistemic support with regard to their lived experiences in the world” (2014: 105, emphasis mine). Scapegoating has this character of ‘othering’ epistemically vulnerable groups, but also vilifying these groups by framing them as responsible for a range of real or imagined social ills. Mentally ill people are responsible for mass shootings; Black and LatinX Americans are responsible for crime and socioeconomic inequality; women are to blame not earning as much as men, seeing that they are less competent. Nothing is the fault of the privileged on this top-down, non-reciprocal, hierarchical framework. Scapegoating reinforces social inequality by redistributing moral responsibility from the privileged to the least well-off, mirroring the flow of currency within the financial economy. This is a far cry from Rawls’ ideal of justice.

Scapegoating as an epistemic practice inflicts distinct  harms on scapegoating victims. Pohlhais, Jr. argues that epistemic injustice perpetrates two types of primary harm: it harms the individual as an epistemic agent, and it harms the epistemic community by withholding or suppressing valuable hermeneutical resources, particularly knowledge about the lived experiences of the oppressed – knowledge that the community is entitled to and requires in order to function well (in a truth-conducive way). Scapegoating is precisely this type of injustice – an injustice that validates the worldview and epistemic standing of the privileged and discredits the lived reality and epistemic standing of the victims, thereby harming the victims and the entirely epistemic community. These harms cannot be seen as equivalent, however. The victim is harmed in a particularly egregious way, as her epistemic standing is damaged, her testimony is discredited, she is prevented from pursuing epistemic projects that stem from her lived experiences, and she is denied the right to form epistemic alliances with other disadvantaged knowers (Pohlhaus, Jr. 2014: 110), and she is then subjected to “secondary harms” (Fricker 2007: 47), such as a loss of moral, socioeconomic, and political standing. This is why scapegoating – similar to gaslighting as described by Kate Abramson (2014) – is a particularly pernicious type of epistemic injustice: it inflicts distinct epistemic, moral, and existential harms on its victims. But unlike gaslighting, which pathologizes the victim, scapegoating vilifies the victim.

5. Who is responsible for scapegoating? Responsibility as indirect control

Who is responsible for scapegoating qua epistemic injustice? Fricker says that perpetrators of testimonial injustice are culpable, unless the hermeneutical resources required to accurately frame the victim’s experiences are socio-historically unavailable, in which case the perpetrator is a victim of “epistemic bad luck” (2007: 42). Sexual harassment, for example, was non-culpable (or less-than-full-culpable) before “sexual harassment” entered the English lexicon, and the same is true of testimonial injustice against victims of sexual harassment, who could not intelligibly frame their experiences (Fricker 2007: 148).

This is a strict view of culpability, as it sees culpability as dependent on control, such that we are only culpable for epistemic transgression that we could have avoided or prevented (or otherwise controlled), either directly or indirectly. (Direct control is too strict, since many mental states are cognitively impenetrable but amenable to indirect, non-immediate control via “life hacks,” such as intergroup contact, implementation intentions, and exposure to counter-stereotypcal images [viz., Holroyd 2012, Christiane Merritt: forthcoming]; thus, indirect control is the better criterion, and currently the more popular one). Some theorists don’t require any amount of control for responsibility, but even on the ‘control view,’ most Americans would turn out to be responsible for scapegoating, given that information about group-level injustice in America is openly discussed, widely disseminated, and accessible to anyone with an Internet connection or a library card. By all appearances, the control condition is met by most Americans on Fricker’s interpretation, since the “relevant concepts” for accurately framing the epistemic harms inflicted on scapegoating victims are “socio-historically available” (2007: 100). While I can’t speak to everyone’s specific epistemic position, I can say this: if you’re reading this blog post, you’re in an epistemic position to be held responsible for scapegoating, should you go ahead and scapegoat a member of a marginalized social group.

Many responsibility theorists subscribe to a version of the indirect-control view. (Fricker is a social epistemologists, not a responsibility theorists per se, though she writes about culpability). J. M. Fischer, the protagonist on the ‘deep control view’ (2006), has never, to my knowledge, written about epistemic ignorance (which is the basis of testimonial injustice), but most theorists who have written on this topic agree that ignorance is not an excuse for wrongdoing, since ignorance can be a culpable failing. People who fail to guard against ignorance are responsible for that epistemic vice and its downstream effects.

Manual Vargas and Jose Medina seem to agree with Fricker that culpability depends on access to epistemic resources, in addition to a functional adult brain. (Children, they would say, are not fully responsible). They are optimistic that ordinary people have the baseline cognitive capacity to sort through competing social scripts, narratives, and schemas, and appraise them for credibility. Unlike totalitarian regimes, liberal democracies involve a marketplace of ideas in which epistemic resources are widely available. That said, epistemic resources may vary by geographical location – for example, 28% of Americans living in rural areas have no access to the Internet, compared to only 23% of urban Americans [Molla 2017], and rural Americans also have less access to library books (Weingarten 2017. These epistemic factors might mitigate responsibility for resource-dependent epistemic deficits, but they don’t necessarily extinguish responsibility. If indirect control is all that is needed, then perhaps neurotypical adults would be expected to stop by a library at some point in their lives. All that we can say for sure is that responsibility is almost certainly extinguished in “epistemic black holes,” i.e., locations in which relevant concepts are completely absent. North Korea involves large areas of epistemic black holes; America involves relatively few. 

It is notable here that many cases of epistemic injustice are not motivated by simple ignorance, but, in Fricker’s view, by “motivated irrationality,” underpinned by “ethically noxious” motives (2007: 34). Scapegoating involves motivated irrationality in that it is motivated, as we saw, by a vested interest in protecting and perpetrating dominant frames of references. Jonathan Metzl notes that scapegoating narratives tend to use different frames of reference to explain the same behaviours in members of different social groups, even when there are no morally salient differences between the two. For example, when People of Color commit mass shootings, politicians and the media tend to frame the event as a collective or group-based problem – namely, a ‘problem with the Black community.’ This narrative has false predictive value because, if true, it would allow us to predict mass shootings on the basis of African descent. On the other hand, when white men commit mass shootings, politicians and the media tend to invoke an individualist or bad-apple framework, which allows them to frame the event as the decision of a mentally ill “lone wolf” (Metz 2017). The reason for this paradigm shift, says Metzl, is that white people identify with other white people and don’t want to see their image reflected back to them in the faces of white shooters, so they are reluctant to identify being white and male as a predictor of being a mass shooter, even though this paradigm would have much more predictive validity than their preferred scapegoating scripts. This exemplifies how scapegoating scripts rests on noxious motives – a vested interest in preserving one’s positive self-conception and privileged status as a white male.

Because scapegoating narratives, as such, involve not only pernicious consequences (the primary and secondary harms of epistemic injustice), but also noxious motives, they could be seen as blame-imputing on two counts: the agent is blameworthy for perpetrating certain harms, and perhaps also for acting on certain noxious motives. While there are debates about the moral status of implicit states (See Kelly & Roedder 2008), most people agree that a person can, under certain circumstances, be blameworthy for expressing morally problematic implicit states in his overt behaviour. Thus, scapegoaters might be blameworthy on both deontic and aretaic grounds, i.e., both for committing a moral transgression, and for expressing character flaws in their behaviour.

6. Responsibility as a social regulation mechanism (functionalism)

Many contemporary responsibility theorists reject the control condition, and subscribe to a ‘functionalist’ view on which blame is appropriate if this reaction would serve some positive social end (e.g., McGeer 2014, Bell 2014, Malle et al. 2014). Thus, people might be blameworthy even if they are irredeemable psychopaths. On this view, scapegoaters should be blamed and virtuous explainers praised, it seems, so as to establish a moral-epistemic ecology in which credible explanatory paradigms are salient, and harmful stereotypes are debunked. Blaming scapegoaters could be a way of condemning the expression of these harmful narratives and thwarting the spread of the “noxious” motives that support them. If so, then blaming scapegoaters is generally a good social policy.

This view also seems to imply that blaming public figures, whose speech is particularly visible, is an especially good social policy. Donald Trump is an example of a very public and very committed scapegoater. Trump, for example, has a habit of scapegoating Muslims for acts of terrorism, in spite of the fact that a majority of domestic terrorism is committed by non-Islamic right-wing extremists [Niewart et al. 2017]); but Trump was quick to swap the collectivist paradigm for an individualist one when he described the Las Vegas shooter as “‘a very sick man’ and a ‘very demented person,’ without mentioning anything about the shooter’s background or potential political ideology” (Metzl 2017).

Who is actually responsible for the American culture of school shootings? James Fallows argues that Mitch McConnel is perhaps more blameworthy than anyone, seeing that he blocked a bipartisan vote on gun control measures by leading a filibuster in 2013, and then Tweeted his “thoughts and prayers” to the victims of the Los Vegas shooting in 2017 (Fallows 2018). The ‘thoughts and prayers’ Tweet is a familiar obfuscatory tactic that substitutes a positive-thinking narrative for a causal explanation. The reason for McConnel’s decision is arguably his funding from the N.R.A. (though he is not even on the list of top-ten Senators and Congresspeople receiving N.R.A. funding [David Leonhardt et al. 2017]). McConnel, and other politicians who have accepted N.R.A. donations, then, seem to be blameworthy for thwarting gun control legislation, thereby perpetuating America’s gun culture, and for acting on ostensibly noxious (financial) motives.

On a functionalist picture, it makes sense to see Trump and McConnel as exceptionally blameworthy for scapegoating vulnerable groups and perpetrating false narratives (e.g., ‘thoughts and prayers’ are effective). But on the control view, they are potentially just as blameworthy. (I say ‘potentially’ because there are substantive questions about whether Trump has a functional adult brain, one that supports self-control [see Hamblin 2018]. In general, however, the the control view and the functionalist view converge in holding cognitively functional adults generally blameworthy for their overt scapegoating behaviours.

If we don’t reject scapegoating narratives about mass shootings and adopt evidence-based blaming practices and policies, then this Onion article might actually be our future:

Screen Shot 2018-02-19 at 5.59.39 PM.png


*I’m not committed to saying that mental illness is a valid construct, but I’m adopting this term from the research simply to pick out a social group that is especially disadvantaged.

The limits of blame: How to avoid blame fetishism

Tl;dr: What are the limits of appropriate or ‘fitting’ blame? Several theorists have raised worries about vicious judgmentalism, excessive blame, intrusive blame, moral grandstanding, and other instances of “blame fetishism” – blaming practices that exceed the limits of appropriate blame. In this paper, I identify eight constraints on appropriate blame, which, if observed, should help us avoid blame fetishism. The constraints are: (1) intelligibility, (2) situational uptake, (3) safety and wellbeing, (4) semantic resonance, (5) privacy, (6) proportionality, and (7) cultural sensitivity. These constraints are rooted in different aspects of blame; namely, since blame is essentially expressive or conversational, typically confrontational, and often punitive, it is subject to constraints (1)-(8).

  1. Constraints on blame

What are the limits of appropriate blame? Quite a few scholars have expressed worries about vicious judgmentalism, excessive blame, intrusive blame, moral grandstanding, and other practices that exceed the limits of judicious or ‘fitting’ blame. But it is exceedingly difficult to carve out the boundaries of appropriate blame.

Elsewhere, I have argued for an equality constraint on blame, as treating others as equals would promote (something approximating to) the ecological conditions of our early evolutionary ancestors, and thereby minimize the role of implicit bias in our ‘responsibility system,’ i.e., the system of relations in which we express the reactive attitudes in response to people’s morally salient characteristics (viz., McKenna 2013, Strawson 1963). Conditions of relational quality would enable us to respond sensitively to the morally-salient qualities of persons, as opposed to blaming and praising in response to morally-irrelevant traits (e.g., race, gender, SES) – a practice that would only reinforce existing social injustices.

The proposed model is a ‘functionalist’ one, in that it sees blame as serving a moral-social function. Functionalists hold that blame serves social purposes and has social conditions or constraints. According to Malle, Guglielmo, & Monro, “persuasive blame” has five conditions (2014: 172): it (i) gets the attention of norm violators, (ii) communicates information about norm violations, (iii) persuades others to accept the blaming judgment, (iv) promotes shared values, and (v) repairs damaged relationships. Insofar as blame is a feature of ordinary conversation, it makes sense to think that persuasiveness is a desideratum. I don’t dispute these conditions, but I believe that they, too, have constraints, particularly the constraint of promoting equality. A blaming judgment can serve to (i) communicate information about a (perceived) norm violation, (ii) communicate relevant information, (iii) persuade others, (iv) promote shared values, and (v) repair relationships, amongst an insular group of racists, without serving any ultimate moral purpose. Thus, conditions (i)-(v) must not only foster consensus, but also facilitate broad equality.

These constraints, together, help to delimit when it is appropriate to blame someone (appropriate contexts), as well as how one ought to express blame (appropriate contents). Some expressions of blame respect conditions (i)-(vi) more than others. If someone interrupts me and I say, “please don’t interrupt while I’m speaking,” this is a more satisfactory expression of blame than, “shut up, cow!” The first response is more (i) informative, (iii) persuasive, and (v) reparative than the latter, so to it is more successful in functionalist terms. To be clear, these should be seen as satisficing as opposed to maximizing conditions: while I should communication information about interrupting in my blaming response, I don’t need to write a dissertation on the harms of interrupting for the intellectual benefit of my interlocutor – this would probably be received as pompous and eccentric. As I see them, (i)-(v) provide thresholds that should be met by a satisfactory blaming response. These conditions can also compete: a more informative instance of blame might be less reparative, whereas a more concise response might elicit more sympathy from someone who’s in a hurry, fostering reconciliation. Thus, the pros and cons of the different conditions must be taken into consideration when formulating a rhetorically effective response.

Having said this, our perception of norm violations is not necessarily trustworthy: we tend to perceive norm violations (e.g., interrupting) differently depending on the demographic attributes of the speaker (gender, race, socioeconomic status). For example, when researchers studied people’s conversations, they found that men who interact with women interrupt 30% more often than when they interact with men; in the course of a 3-minute conversation, men interrupted women 2.1 times  (Hancock & Rubin 2014). This phenomenon is colloquially known as ‘manterrupting.’ Although interrupting (and even more so, ‘manterrupting’) is rude, men who speak more frequently are nonetheless perceived as more competent, while women who speak more frequently are perceived as less competent (Brescoll 2011), which means that manterrupting is likely to be perceived as a virtue only because the speaker is male, whereas interrupting by a woman is more likely to be seen as transgressive. This discrepancy exemplifies the importance of the equality constraint: our evaluative judgments and perceptions are biased by cultural stereotypes that harm oppressed social groups on average. (Implicit biases are triggered by speicifc activating conditions, but in general acculturated implicit biases are to the advantage of historically privileged groups). We must therefore be vigilant about the role of implicit biases in our communicative dispositions and jugdments. Unlike constraints (i)-(v), (vi) is indefeasible and domain-general, meaning that we must always promote equality, and there is no upper limit on the value of this condition. Equality prevents us from employing other conditions in biased ways, so it is, in a sense, an architectonic constraint.

Conditions (i)-(vi) thus provide a bulwark against a common worry about blame – namely, that focusing on blame might promote ‘blame fetishism,’ or a propensity for exorbitant blame. Malle and his colleagues’ work is helpful in this regard. But the proposed list of six constraints (including equality) is surely not exhaustive, and it is worth examining whether there might be other critical checks and balances. In the course of this paper, I will elucidate eight additional constraints, drawing on the philosophical literature and common sense. First, however, I will overview some salient about blame fetishism.

  1. Blame fetishism

Some people worry that fixating on blame can promote a kind of ‘blame fetishism.’ The worry is that talking about blame might leads to an unhealthy obsession with blame, a hyper-focus on putative norm violations, and a general propensity to blame people harshly, uncharitably, and ungenerously. Raymond Lau (2009) argues that we live in a “culture of blame” that denies the importance of chance in our lives, giving rise to excessive and obsessional blame. (Neil Levy says something similar about the denial of chance in his 2011 book, which defends responsibility elminativism). According to Lau, blame fetishism is a modern phenomenon connected with America’s culture of rampant litigation. Similarly, Paul Bloom thinks that we should blame people less often and less stringently by cultivating “rational empathy,” rooted in an understanding of our shared humanity, particularly our shared cognitive biases (2016). His theory has a “there but for the grace of god go I” flavor; he asks us to forgive people their enculturated biases. In a similar vein, Gary Watson admonishes us to avoid the “vice of judgmentalism” (2013), which includes being ungenerous and unaccepting of the faults of others. Tosi and Warmke (2016) criticize “moral grandstanding,” i.e., an inauthentic expression of blame used to gain attention and popularity, similar to self-centered ‘virtue signaling.’ These views all caution against the perils of blame fetishism.

Prior to any of these conversations, Nietzsche was skeptical of moral judgment, including blame. He criticized “master morality” for being dogmatic, limited, and incomplete: he wrote, “human history would be a really stupid affair without that spirit which entered it from the powerless” (Nietzsche 2009 [1887]: essay 1, sec. 7). But he also criticized its counterpart, “slave morality,” for being backward-looking, hostile, and incapable of overcoming its own limitations. Both “moralities,” he thought, rest on a kind of epistemic narcissism, and both are incapable of transcending their limitations to generate genuinely new discursive and existential possibilities. The role of blame within each system must be limited to the promulgation and validation of the insular norms of the system, which strengthens hostility to perceived outsiders (i.e., out-group bias), and increases social stratification. Blame, on this picture, is a dogmatic and polarizing element of a segregated society with two competing and irreconcilable moral frameworks.

I agree with the spirit of all these perspectives, insofar as I think that blame is typically expressed in patterns of thought and action (and social narratives and scripts) that oppress historically disenfranchised groups (see my previous work). In modern American society, mainstream blaming practices smack of “master morality,” in the sense that they recruit pernicious stereotypes and scapegoating narratives to punish, oppress, discredit, and marginalize the socially disadvantaged. (e.g., See Trump’s tweets for an example of scapegoating narratives that exploit cultural stereotypes to harm minorities [Cauley 2017]).

That said, I believe that the solution to this problem is not to eliminate blame, but rather, to shape blame in such a way as to foster social equality. This project admittedly involves the expression of hostility, resentment, disappointment, distrust, censure, and other negative ‘reactive attitudes’ (Strawson 1963) toward transgressors, particularly transgressors who abuse their institutional privileges; but this reactive response is not, contra Nietzsche, backward-looking and deconstructive; it is forward-looking and constructive, aimed at creating new social imaginaries (Medina 2012), engineering new moral ecologies (Vargas 2013), and fostering epistemic respect. It is geared toward ‘troubling’ extant moral discourses and producing new epistemic, existential, and performative possibilities (Butler 1990). Successful blaming practices foster epistemic and moral justice, and for this reason, they reduce historical binaries, exclusions, and fractures, and produce the preconditions for moral, political, and epistemic equality.

This is, of course, an (egalitarian) ideal of blame, not the reality, and I agree with skeptics that blame doesn’t foster equality in general. But I think that blame can be, not exactly “sanitized,” but rehabilitated (viz., McGeer 2008). To rehabilitate our blaming system, we need to be vigilant about how we deploy blame, and we need to be cautious about whether our expressions of blame respect success conditions (i)-(vi).

Admittedly, these conditions are not exhaustive. I’m not sure it would be possible to enumerate a set of necessary and sufficient conditions on successful blame. Conditions (i)-(v) are better seen as prototypical constraints, which are defeasible, can compete with one another, and are sensitive to context. (vi), on the other hand, is necessary and indefeasible. In addition to these conditions, philosophical theory and commonsense point to more specific limitations, which may or may not be subsumed under (i)-(vi) – I will not decide that here. One salient example is privacy: many theorists believe that we should limit the extent to which blame infringes on people’s privacy (e.g., Darwall 2011; Bell 2013). Another example is cultural variation: many theorists think that blame should be sensitive to cultural differences (e.g., Wong 2006). It is worth considering these types of commonsense constraints, which, if observed, could contribute to the project of rehabilitating blame and placing constraints on errant and excessive tokens of it.

Before proceeding with this inquiry, I will first explain what I take ‘blame’ to mean. The constraints on blame effectively fall out of one’s theoretical framework. Since my view is ‘expressivist’ (more on which in section 3), many of my constraints on blame are going to be extensions of the constraints on conversational contributions (e.g., intelligibility, uptake, semantic resonance); some will be constraints on confrontation (safety, wellbeing, privacy); and others will be constraints on punitive treatment (proportionality, cultural sensitivity). Since blame can have all of these features (conversational structure, confrontational interface, punitive effects), it is subject to all of these constraints. Blaming tokens that violate these constraints are prima facie inappropriate, unless they are justified by some feature of the context.

I hope that by illuminating these constraints, I can deflate the worry about blame fetishism. My arguments, of course, don’t establish that people won’t be blame fetishists; that’s up to them. I can only show that we don’t have to be blame fetishists, and we won’t be if we observe these checks and balances and hone epistemic virtue. I think that talking about blame is, in fact, more likely to democratize, domesticate, and enrich our system of blaming interactions, rather than corrupting it beyond its current level of depravity and inequity. Since our blaming system is currently strongly shaped by sexism, racism, transphobia, and other implicit and explicit biases, we should be having conversations about how to repair it.

  1. Blame as paradigmatically expressive

Although some theorists hold that blame can be a private judgment, most Strawsonians espouse expressivist accounts of blame (and praise), on which blame is paradigmatically an expressive exchange (e.g., McKenna 2012, 2013; McGeer 2013, Bell 2013, Malle et al. 2014). This doesn’t mean that blame cannot be unexpressed or unspoken, but such ‘internal,’ or merely thought, instances of blame are not the paradigmatic case. The reason for holding that blame is paradigmatically public is that blame cannot function to regulate interpersonal relationships if it is private, and Strawsonians subscribe to an essentially interpersonal account, so most expressivists (including those just cited) also subscribe to an expressivist model.

On the other hand, one might argue that blame is paradigmatically private because it is causally antecedent to public blame, and causal priority is a mark of paradigmatic-ness. This is not necessarily how blame works, though. In morally competent adults, an internal blaming judgment may precede the public expression of that judgment in many cases, but this is not necessarily the normal causal sequence: an overt expression of blame may be an automatic, knee-jerk response to blame-eliciting conditions, with the reflective judgment coming after, perhaps in the form of a post hoc rationalization (which is how Haidt thinks moral judgment typically works [2001]). On McGeer’s model (2008), blame is an evolutionary adaptation that is ‘canonically’ emotional and quite often automatic, though automatic blaming responses are susceptible to cognitive mediation. Public expressions of blame, on this view, are not necessarily caused by prior reflection – they may be automatic emotional responses to situational triggers. Some theorists construe blaming responses as habits (or habituated response patterns), which consist of (expressions of) neural patterns automatically activated in response to salient eliciting conditions (Wood 2017); but these neural structures are partly caused by reflective states and activated knowledge structures. Thus, habituated responses, even if automatic, are also ‘reflective’ in the sense that they are conditioned by the agent’s system of reflective judgments (system-2 processes) and learning history.

The most empirically accurate picture of blame on these models is as a hybrid of automatic (system-1) and reflective (system-2) judgments. To see blame as ‘paradigmatically reflective’ would be an oversimplification; blaming attitudes are states with complex neurocognitive profiles that implicate many brain structures and causal pathways. Therefore, if causal priority were relevant to the question of ‘paradigmaticness,’ this criterion wouldn’t settle the question of which type of blame (private or public) is paradigmatic.

In any case, expressivists don’t see pubic blame as paradigmatic because of its neurocognitive profile, but rather, because of its role in a network of social relationships. Expressivists are still interested in neurocognitive states, but only as parts of persons embedded in ecological systems with other persons and mediating environmental factors. The primary locus of analysis is the relationship between people’s reactive attitudes, their interpersonal relationships, local epistemic resources, political systems, social institutions, and other extended networks. Since I, too, am interested in blame as part of a system of (asymmetric) relationships and institutions, I take public blame to be the paradigmatic case, noting that private blame is a salient causal mechanism in the minds of persons, and worthy of attention in this capacity. Still, my primary interest is the causal role of expressions of blame in our complex social reality.

If blame is paradigmatically expressive, as I think it is, then it is subject to some of the same success conditions as expressive communication in general. Most people think that there are better and worse ways of communicating. Just because we can express a thought doesn’t mean that we should. I could give a reckless driver the finger, but I would probably be better off signaling my disapproval in a less hostile way. Blame can also be confrontational and punitive, insofar as it involves criticism, and these types of expressive interactions are subject to additional constraints. In what follows, I will outline some of the key constraints on these types of interactions. They are: (1) intelligibility, (2) situational uptake, (3) safety and wellbeing, (4) semantic resonance, (5) privacy, (6) proportionality, and (7) cultural sensitivity.

  1. Constraints on blame

(1) Intelligibility

If blame is part of a communicative exchange, then it must be subject to conversational maxims that govern effective communication – namely, Gricean maxims. These are:

“(A) The maxim of quantity, [whereby] one tries to be as informative as one possibly can, and gives as much information as is needed, and no more; (B) The maxim of quality, [whereby] one tries to be truthful, and does not give information that is false or that is not supported by evidence; (C) The maxim of relation, [whereby] one tries to be relevant, and says things that are pertinent to the discussion; and (D) The maxim of manner, [whereby] one tries to be as clear, as brief, and as orderly as one can in what one says, and where one avoids obscurity and ambiguity” (

This is a descriptive account that outlines the norms that govern natural conversation; but people who depart from these maxims are less intelligible to conversational partners, meaning that the account has normative upshot: we should abide by Gricean maxims if we wish to be adequately intelligible. Notably, blame must be intelligible to effectively regulate social interactions: unintelligible expressions of blame are hardly more causally efficacious than private blame, and to the extent that they are causally efficacious, they are confusing, disruptive, frustrating, and polarizing (i.e., not functional). Blame, then, should conform to Gricean maxims (all things being equal). We should not express blame in ways that are: (A) uninformative, (B) untruthful, (C) irrelevant/ill-founded, or (D) confusing.

This means that some common blaming practices are suboptimal or prohibited, such as: (A) Giving someone ‘the silent treatment’ without  providing an explanation (uninformative); (B) Scapegoating innocent victims (untruthful); (C) Blaming people for their sexual preferences (‘slut-shaming’), their weight (e.g., ‘body-shaming’), and other morally-irrelevant traits (ill-founded); and (D) Blaming people in abusive and insulting terms, such as ‘stupid,’ ‘lazy,’ and ‘crazy,’ which have no clear referent, and arguably no referent at all (confusing, misleading). Saying that someone is ‘crazy,’ for example, doesn’t direct attention to any discernable moral defect, and, since the term ‘crazy’ rests on a false, pathologizing narrative with gendered and racialized features, it appears to have no proper referent (Abramson 2014, Davis & Ernst 2017). This type of attribution is not only confusing, but confused and morally pernicious.

It’s worth noting that these constraints help functionalists escape standard objections to forward-looking theories of blame. Because successful blame respects Gricean maxims, there are inbuilt reasons not to blame innocent victims, even if doing so would prima facie advance some desired social end. The specter of victim-blaming is an objection to any forward-looking position, but expressivist accounts place restrictions on blame, rooted in intelligibility, social cohesion, and other social values (in my case, equality). Expressivism, then, is less susceptible to victim-blaming objections than many other forms of consequentialism. It’s also relatively immune to worries about gratuitous retributivism, since retribution is a condition of desert-based accounts, not functionalist accounts (though punishment could be a means to an end).

(2) Situational uptake

Even intelligible blame should not be expressed in every situation. Some intelligible (Gricean-maxim-respecting) expressions of blame cannot be grasped by the intended recipient due to situational constraints. While there may be agents who are globally unresponsive to blame – the primary candidate being psychopaths (viz., Levy 2007, Shoemaker 2011) – everyone is susceptible to situational sensitivity deficits.

Michael Smith (1995) presents an interesting example of someone afflicted by a (presumably) situational motivational deficit that prevents him from acting on his best judgment. This person (call him ‘Hostile’) cannot bring himself to shake the hand of his opponent (call her ‘Dejected’) after being badly trounced in a game of tennis. If he were to try to shake her hand, he would be compelled to smash her in the face with his racket in a fit of rage. Therefore, what Hostile ought to do, says Smith, is salute Dejected and walk off the court to cool down. (It seems that Hostile should also seek treatment for severe emotional dysregulation, as his response to losing a recreational sporting match indicates deeper emotional problems; but let’s say for the sake of argument that this is an uncharacteristic moment of rage). Hostile’s situational blame-insensitivity prima facie defeats Dejected’s reason to blame him (for suberogatory conduct, i.e., not being a good sport) while he is in that states of mind, since blaming him would only incite him to attack her. What she ought to do, then, is suspend blame until a more opportune time.

There are many situations in which people are temporarily unresponsive to otherwise intelligible blame. If, immediately after snubbing Dejected, Hostile had received a phone call informing him that his beloved mother had just passed away, this too would provide a reason for Dejected to suspend blame, seeing that Hostile is in no condition to respond. Any transient motivational deficit (grief, depression, anger) can provide a reason for the blamer to defer the expression of blame until a more uptake-conducive time.

On the other hand, if someone is consistently unresponsive to blame, then expressing resentment or censure toward the person precisely for having this character flaw could be justified. This response could signal to an audience that the person is morally incompetent and possibly dangerous. Flagging an unresponsive person to the community would satisfy conditions (i)-(v) relative to the community. Thus, in some cases, lack of sensitivity in the wrongdoer is not a reason to suspend blame, but rather, to express a blaming attitude (resentment, disapprobation) in public to flag the agent as a threat. (Consider the women who came forward to blame Harvey Weinstein, who is by all appearances unrepentant; their intention was patently not to enlighten the perpetrator, yet it would be odd to say that they were not, in any sense, blaming him).

 (3) Safety and wellbeing

As the previous example shows, it can be dangerous to express blame. Sometimes the danger comes from a hostile individual, and other times it comes from an oppressive political regime. In America, blaming an angry driver can result in a shooting. In North Korea, blaming the government is a death wish. Political dissidents in North Korea are imprisoned, tortured, and killed, and their children and grandchildren can be sentenced to intergenerational incarceration and torture. Some Americans, too, are unjustly incarcerated for their political activism, a notable example being Angela Davis (Davis 2016). In situations of imminent danger and political oppression, it may be reasonable to suppress blame for reasons of personal safety.

That said, suppressing blame for too long can undermine its regulatory force: silenced blame is private and politically inert, and blame could, in time, cease to exist even in the minds of the oppressed. (This seems to be a factor in North Korea, where people lack the epistemic resources to adequately frame their own political oppression, and do not seem to have a robust working concept of personal liberty [Kirby 2017]). The suppression of blame comes at a cost to the group: namely, it can damage the local social imaginary and moral ecology (viz., Medina 2013; Fricker 2007; Vargas 2013). Thus, while it may be reasonable to suppress public blame in the face of threats to personal safety and wellbeing, the value of survival must be weighed against the value of political activism. Angela Davis judged incarceration to be worth the price of speaking her mind, but her actions were arguably supererogatory as opposed to obligatory. North Koreans face much harsher penalties for political dissent. When someone’s survival is at stake, this may provide a compelling reason for the person to withhold blame, at least until the danger can mitigated.

Lack of uptake is another factor in North Korea: even if North Koreans protested the Kim regime, their speech would not receive significant uptake due to egregious political oppression. But if North Koreans could positively influence the regime, it’s not clear that they would have a duty, or even a strong reason, to do so at the expense of their lives and their family’s survival, and at risk of intergenerational torture.

 (4) Semantic resonance

There can also be constraints on the semantic content of blame, or the words used by the blamer. (This is related to intelligibility, but it specifically concerns the resonance of semantic contents). This aspect of blame is somewhat neglected in the literature. Scholars write extensively about the emotional and cognitive contents of blame, but say relatively little about its semantic contents. Word choice and word order, however, significantly mediate respondents’ receptivity to statement, whether moral or not. For example, when physicians framed a surgical intervention in two different ways, with the first group describing the surgical intervention as having a one-month survival rate of 90%, and the second group describing it as having one-month fatality rate of 10%, most patients (84%) in the first condition consented to the intervention, while only half of those in the second condition did the same (Kahneman 2011). The reason is that we respond more favorably to words like ‘survival’ than to words like ‘fatality.’ Similarly, respondents are more favorable to a ‘tax relief’ as opposed to a ‘tax cut,’ even if the proposed tax reduction is the same in both cases (Moscrop 2015). In other words, framing influences uptake. Successful blame, then, should avoid framing, priming, and word-order effects that undermine receptivity. For example, when blaming someone, we should avoid invoking gender and racial frameworks that promote inequality, such as “boys will be boys,” or “he’s a good boy” – language that relies on the myth of white male exceptionalism to systematically excuse white men. This language influences us to be less hostile to people who fit the favourable stereotype, and more hostile to those who do not.

The semantic content of blame also communicates relevant information about norm violations and character traits, and there are debates about what contents (aretaic or deontic) are admissible. McKenna (2018) maintains that blame can express different evaluative judgments – namely, judgments about norm violations (deontological judgments) and judgments about character flaws (aretaic judgments). He is neutral between these accounts – he thinks that we can blame people both for wrongful actions and for character failings, both for norm violations and for suberogatory actions. The unitary deontic view, says McKenna, doesn’t capture pervasive cases of blame, such as when we blame people for suberogatory actions that don’t violate any norms. To illustrate, if I snub my neighbour whenever she waves to me, I could be subject to (mild) blame on grounds of suberogatory conduct (being a bad neighbour). I’m not breaking a rule, but I’m exhibiting a character defect. It’s not clear why this would be an appropriate response in deontic terms, but the functionalist approach easily accommodates both types of judgment, since information about deontic properties and information about aretaic properties can be beneficial. Thus, both judgments have a place in our system of expressive relations. I can blame my neighbour for taking my designated parking spot (a norm violation) and for deliberately snubbing me (a character failing).

Another point of contention is whether people can be held responsible for actions over which they (putatively) have no control. In a debate between Manuel Vargas (2005) and J. M. Fischer (2011), Fischer argues that we cannot be responsible for uncontrollable mental states, whereas Vargas maintains that we can. My impression of this debate mirrors McKenna’s response to the deontological-aretaic divide: why not both? No one denies that it is permissible to blame people for norm violations over which they do have control, since blame in such cases is both putatively deserved (and thus justified on a desert framework), and putatively agency-enhancing (and thus justified on a forward-looking framework). Blaming people for uncontrollable character defects, in contrast, may not be ‘deserved’ if desert requires control, but it can easily be justified in functionalist terms, since it can serve positive ends such as (i) getting the attention of the blamee, (ii) conveying relevant information, (iii) persuade others, etc. Moreover, even if the transgressor is unresponsive to blame, publicly criticizing him could signal to the community that he is a threat, which could benefit vulnerable members of the community. Therefore, in functionalist terms, we can blame people both for controllable motivational deficits and uncontrollable ones.

That said, to be functionally justified, our blaming expressions should express judgments, words, phrases, and frameworks that facilitate intelligibility, pro-sociality, and other positive outcomes. That is, while various types of contents are admissible, we should choose our communicative vehicles with care, taking into account the likely consequences of our speech, particularly in light of pervasive epistemic injustice.

 (5) Privacy

Even if it is safe to express blame, it may be intrusive. Some philosophers hold that blame depends on an asymmetrical relation of authority between the blamer and the blamee. Blaming strangers, with whom we have no relationship, can seem intrusive (Darwall 2006). Macalister Bell (2013) argues that blame can be judiciously expressed outside of relationships of authority, provided that its expression serves a social purpose (e.g., motivating the transgressor, educating the community). That said, Bell retains a version of the privacy constraint, separated from the authority condition. On her view, we have independent reason to respect people’s privacy, insofar as they have a right to it. This preserves the intuition that it is better to blame people in privacy-respecting ways than in privacy-violating ways: for example, it’s better to accuse your spouse of adultery in private than to accuse him at his 40th birthday party in front of his family (all things being equal). The reason for exercising discretion is not that you don’t have the authority to blame your spouse at his birthday party – you might be within your moral rights to do so; rather, the reason is that infringing on someone’s privacy is a norm violation, and you should avoid committing norm violations in your blaming practice if you can. Ideally, we should blame people in ways that respect their rights.

That said, the duty to respect people’s privacy is defeasible, because not everyone has a right to privacy, or an equal right to equal privacy: people who pose a danger to the public arguably forfeit their right to privacy (e.g., serial sex offenders), as do public figures whose speech is on the public record (e.g., politicians’ Tweets, celebrities’ interviews). Hence, blame should be sensitive to privacy, but the right to privacy is not inviolable, equal between private citizens and public figures, or equal within these groups.

(6) Proportionality

If blame is to be persuasive, its semantic and affective contents should be proportional to the offense. (By semantic content, I am referring to the language used). Excessive blame tends to be unintelligible, confusing, and alienating. To give a famous example of semantically and emotionally excessive blame: Christian Bale blamed a sound technician for interrupting his scene during the filming of “Bat Man” by yelling at him. What Bale yelled was, “What the fuck is it with you? What don’t you fucking understand? You got any fucking idea about, hey, it’s fucking distracting having somebody walking up behind Bryce in the middle of the fucking scene? Give me a fucking answer!” (He continues like this (Telegraph transcript 2009). It’s pretty obvious that this wasn’t the most effective blaming response, the reason being that the emotions (rage) and the language (expletives) were too strong for the transgression. (Language and emotions are related because strong language often serves to convey strong emotions, but the two come apart: I can swear at someone unemotionally, and I can express strong emotions in polite language). Bale apologized (whether genuinely or not) in response to public scorn and ridicule. His transgression was excessively blaming someone for a benign infraction, which violates the proportionality constraint.

This isn’t to say that strong affect and language is always out of place. It’s simply to say that the language and emotional tenor of speech should fit the transgression. When it comes to systemic injustice, such as the cultural genocide perpetrated against Indigenous peoples, it’s difficult to imagine what could count as an “excessive” blaming response, short of violence. That said, the justifiable anger shared by oppressed groups can be misinterpreted by privileged people who are insensitive to the magnitude of the offense. Some people might respond better to a less emotional reaction. But this simply means that both types of response – emotional and unemotional – are fitting on different grounds. Indeed, these responses may be complementary, in that they serve different social purposes. (See articles on ‘tone policing’ (EF 2015) for more on why emotional responses to systemic injustice are valid and deserve social uptake).

Blame should thus be proportional to the offense, but our society tend to weigh offenses unfairly due to epistemic injustice, which results in the perverse marginalization of historically disenfranchised groups. Thus, we should be cautious to blame proportionally, and to interpret offenses in light of social injustice.


 (7) Context-appropriateness

Blame can also convey context-inappropriate emotions and judgments. Christian Bale’s rage, for example, was less suitable than mild irritation for that particular context, whereas mild irritation is an inappropriate response to cultural genocide. Blame’s emotional and cognitive contents should fit the situation, in addition to being proportional to the offense. The situation, then, places constraints on fitting blame.

Theorists debate whether some emotional contents may be categorically inappropriate, because they incite anti-sociality and social discord. Examples of potentially polarizing emotions include ridicule, anger, and contempt. We see these emotions on display in the public sphere, particularly in late-night talk shows. For example, Stephen Colbert regularly ridicules Donald Trump. When Trump tweeted that he had to “deal with fool[s]” like Bob Corker in the White House, Colbert responded that Trump’s real problem is “a crippling disease of Narcissistic Personality Disorder… ‘allegedly”” (Shapiro 2017). This criticism was, of course, unlikely to receive uptake from Trump or his base, but it was eminently well received by Colbert’s audience. Could Colbert’s public contempt for Trump promote positive social consequences? I think so. There is evidence that (a) ridicule “has significant potential to mobilize public interest constituencies, and to raise the consciousness of members of the public who might not otherwise be attuned to conventional policy discourse” (Grobosky 2013); that (b) anger is effective at motivating prosocial behaviour (Prinz & Nichols 2010), and that (c) contempt can consolidate in-group members and motivate transgressors to conform to community norms (Fischer & Giner-Sorolla 2016). These emotions, then, can promote moral ends. Even if these emotions are not always appropriate, they surely are not categorically harmful. Thus, in the right situations, these emotions can be apt contents of blame.


 (8) Cultural sensitivity

Cultural relativists, like Gilbert Harman (1996) and David B. Wong (1995, 2009), say that we can only express blame intra-culturally, not across cultures. Hence, I cannot blame people for committing non-consensual clitoridectomies on girls in non-Western cultures, or (perhaps) even in non-Western subcultures embedded in my own culture. Most theorists don’t agree with this position, for theoretical and practical reasons. In practical terms, relativism seems to promote cultural isolationism, and in theoretical terms, it fails to make sense of moral cross-cultural contamination – the sharing of moral knowledge across cultures, which is ubiquitous. Donald Davidson’s triangulation theory (1982) speaks against the very possibility of deep cultural relativism, or the idea that we can’t understand out-group beliefs and values (viz., Claudine & Myers 2018). Ultimately, there is no good reason to think that we can never blame members of other cultures in terms that are intelligible to them; and even if we couldn’t, it wouldn’t follow that we can’t blame out-group members for in-group purposes, such as to signal that certain practices (like female genital mutilation) are categorically unacceptable within our own culture.

Having said that, there is every reason to think that cross-cultural blaming judgments can be ignorant, condescending, oppressive, and otherwise epistemically and morally ignorant, seeing that blame has historically functioned to dehumanize perceived out-group members, and it still serves to oppress marginalized groups. This is seen, for example, in the representation of Asian Americans as a ‘model minority’ – a cultural stereotype that serves to ‘other’ and homogenize the target group (Chou et al. 2015).

Nonetheless, it can be appropriate to blame members of other cultures in culturally-sensitive ways. This requires having familiarity with the target culture. Familiarity comes in degrees: someone who grew up in a given culture is likely to be more sensitive to the culture’s norms and practices than an outsider, but we can gain sensitivity to a culture’s norms and practices through familiarity with insiders and their epistemic products. Thus, it is possible to aptly blame members of other cultures if we cultivate adequate sensitivity. This constraint offers protection against familiar ethnocentric biases in our intergroup blaming practices.

  1. Final comments: Equality and coalescence

Blame, as I have described it, is a paradigmatically expressive practice, governed by communicative norms of intelligibility, uptake, and semantic resonance. Blame is not just any type of expressive practice, though; it can be antagonistic, intrusive, hostile, and punitive, and as a result, it is also properly governed by norms of safety and wellbeing, privacy, and proportionality. Finally, since blame often punishes people on ethnocentric and xenophobic grounds, it should be governed by an attitude of cultural sensitivity.

I want make an important qualification before closing. Although blame is often hostile, and is in fact described by Macalister Bell (2013) as paradigmatically hostile, I prefer to think of it as essentially argumentative or dissensual. Blame, in essence, identifies and responds to a moral transgression or character flaw, and thus contains or expresses criticism, but it does not follow that blame essentially contains or expresses hostility. While we often think of argumentation itself as hostile, many arguments are merely dissensual, and either unemotional or emotionally positive. Arguments are often deployed in a spirit of epistemic respect and earnest curiosity, as argumentation theorists have argued at some length. Indeed, while arguments are essentially dissensual, they also essentially function to promote coalescence, or agreement based on shared goals, according the preeminent argumentation scholar Michael Gilbert (2013). That is, although argumentation is dissensual by its very nature, it ideally promotes agreement. I concur with Gilbert that argument paradigmatically functions to foster coalescence between argumentative partners, but I would add that, in order to serve this end, there must be an even playing field between argumentative partners – that is, there must be relational equality. Analogously, I think that the paradigmatic role of blame in our system of moral relationships is to foster coalescence around our shared moral goals, through a process of argumentation in conditions of epistemic justice. For this process to obtain, we must repair rifts, exclusions, and marginalizations in our moral ecology. Ideally, blame functions to foster coalescence through a system of equal relationships. By enhancing equality, we enhance our chances of virtuous or morally responsible coalescence.

The project of repairing our moral relationships goes hand in hand with the project of trying to observe the constraints on blame identified here. To limit the risk of blame fetishism, we should observe conditions (i)-(vi), as well as conditions (1)-(8). By doing this, we increase the chances that our blaming attitudes will hit their mark, i.e., the morally salient qualities of persons.


(Really good book, by the way).


Can children be responsible agents? (tl;dr: No).

Philosophers tend to see young children as paradigmatic cases of non-responsible agents. Strawson, for example, says that very young children are non-responsible, but older children as “penumbral” cases, as they are acquiring the capacity for responsibility (1963). Strawson also cites people with severe mental disorders as paradigmatically non-responsible. These two types of agent reside in a sphere outside of the reactive attitudes (blame, praise).

In my contribution to the Bloomsbury Companion to Philosophy of Psychiatry, I argued that people with mental disorders are not, in fact, non-responsible, contrary to the received philosophical wisdom. Indeed, seeing psychiatric service users are non-responsible is incompatible with the dominant model of therapeutic treatment, the person-centered model, on which service users are treated as presumptively capable of exercising responsibility over their treatment-relevant choices, as well as a non-trivial range of non-therapeutic choices (Rogers 1986). Why have many philosophers, then, seen service users (‘the insane’) as non-responsible? One likely explanation is culture-wide implicit bias against persons with psychological disorders. (Strawson wrote his seminal work in the 1960s, remember). Another likely reason, I believe, is the tendency to see responsibility (tacitly, perhaps) as a modular capacity rather than a complex suite of capacities involving many cognitive processes distributed across the brain. On the modular view, moral processing is similar to colour processing, which involves a small constellation of cortical regions, where damage to any one can severely impair the perception of colour.

On the complex-capacity model of responsibility, a person can lack one of the capacities paradigmatically implicated in responsible agency (e.g., perspective-taking), but posses many others (e.g., self-efficacy). Consider two cases that have received much philosophical scrutiny (e.g., Jeanette Kennett 2002): psychopaths and people with autism spectrum disorder (ASD). Both groups share deficits in empathic processing and traits of alexithymia (the inability to describe emotions in the self), but psychopaths have difficulty resonating with others’ emotions, whereas people with ASD have difficulty with cognitive perspective-taking and emotional recognition, and not vice versa (Morosan et al. 2017; Lockwood et al. 2013). On the basis of these cognitive profiles, we can infer that people with ASD may be forgiven for failing to register someone’s emotional distress, but not for failing to care about a person’s emotional distress once registered, since ASD impairs the first capacity (emotional recognition) but not the latter (resonating with emotions). Psychopaths, meanwhile, have much more severe moral deficits – they register distress in others, but do not care about it. This raises the legitimate question of whether psychopaths are morally responsible agents in any non-trivial sense. Aside from the psychopathic population (which is responsible for a vastly disproportionate amount of crimes according to the author of the Psychopathy Checklist PCL-R, Robert Hare), persons with psychological disorders across all DSM-5 categories possess many of the paradigmatic capacities implicated in responsibility. In general, deficits in responsible agency are local and highly specific, not global, and acquired deficits may be responsive to therapeutic interventions. Thus, it makes sense to hold people with psychological disorders responsible for actions not impaired by those disorders.

Young children are a different case because they have deficits in all responsibility-relevant capacities, as they are not fully cognitively developed. This has led some people to jokingly observe that “all children are little sociopaths.” But children really do have morally-relevant cognitive deficits, including theory-of-mind deficits (Wellman et al. 2001), perspective-taking deficits, and planned problem-solving deficits (Tecwyn et al. 2014). That said, children have different capacities depending on their age, cognitive architecture, and learning environment. Does this mean that children are quasi-responsible?

I am inclined to say that young children, unlike adults with domain-specific cognitive deficits, are not responsible, full stop. Of course, at some indeterminate point, children become responsible, but young children, in my view, are non-responsible. When we blame and praise children, we are, I believe, using operant conditioning to shape children’s cognitive architecture to discriminate different moral stimuli and respond sensitively to discriminable stimuli in the future and across various contexts. Young children unquestionably possess the neurological states required to respond to conditioning (e.g., nucleus basalis neurons and dopamine receptors). Even non-human animals possess these states, which is why they are capable of learning conditioned responses.

There is another critical element to this picture, however. Once children begin to develop the neural correlates of moral responsibility (over and above the neural structures implicated in conditioned learning), they still are not responsible for their behaviour on my view, because they are not responsible for the acquisition (or not) of these higher-order neural structures. Their caretakers are responsible for conditioning them to respond sensitively to moral stimuli, thereby acquiring adult moral capacities. If children grow up with deficit in moral cognition due to caretaker neglect, this is their caretaker’s fault. (The correct specification of ‘caretaker’ is complicated, but for my purposes, it can be seen as encompassing anyone responsible for the child’s wellbeing). Thus, children are not responsible for their behaviour even if they possess some morally-relevant capacities, because any non-congenital deficits in their cognitive architecture are their caretakers’ fault, and any congenital deficits are the result of the natural lottery, i.e., not their fault. So, if a child, say, burns down a bee farm, bankrupting the owners and killing half-a-million bees, the caretakers are responsible, not the child. The caretakers are responsible to pay restitution, and are also the correct targets of moral blame.

The reason children are not responsible for their behaviour when adults with equivalent cognitive deficits are, is that children’s are not autonomous: they are under their caretakers’ directorship. Children cannot autonomously choose their learning environment, peer group, school – the salient features of their ‘moral ecology,’ so to speak (Vargas 2016), whereas adults (in a pluralistic democratic state) can choose one moral ecology over another. Because children lack autonomy vis-a-vis their moral ecology, they lack responsibility for their actions, even if they have many of the neural structures implicated in the capacity for moral responsibility. In other words, if child C and adult A have similar responsibility deficits D, child C will not be responsible for moral infractions caused by D, whereas adult A may be responsible moral infractions caused by D (depending on other conditions, such as whether A could have taken steps to remediate D or prevent the transgression by other means). Thus, even if C and A are cognitively identical, their responsibility status differs due to different background conditions.

It is only once children have autonomy comparable to most adults that they become responsible.

Similar considerations apply to adults – namely, adults can be differentially responsible depending on their access to different moral ecologies. (For a full argument to this effect, see Ciurria 2016 in the Journal of the APA). Yet, the differences amongst adults will be a matter of degree, not of kind; while adult P may be less responsible than adult Q for moral infraction I when P and Q have identical cognitive architectures but differential access to agency-enhancing moral ecologies, child C is not responsible for I across all moral ecologies. Unlike children, no cognitively-normal adult is non-responsible by virtue of ecological deprivation, because all societies include a non-trivial amount of cultural pluralism (with the possible exception of North Korea). In America, there is extensive pluralism, so no adult can escape blame (or praise) by saying that she couldn’t have done otherwise.

Thank you for reading my post.



Blame as social cognition: The Path Model

Tl;rd: Malle, Gugliemo, and Monroe (2014) offer the Path Model as a representation of the cognitive architecture of blame. On the Path Model, blaming cognition involves information-processing over criterial information in a paradigmatic sequence of steps. This sequence is based on subjects’ responses to surveys. Malle et al. say that biases and affect, which are not represented on the model, do not play a ‘paradigmatic role’ in blaming cognition. In this post, I contend that these states do play a paradigmatic role in the blaming architecture of some people, particularly those high in implicit bias and low in emotional regulation. This is an expectable result of living in conditions of socioeconomic inequality, which give rise to implicit bias. Subjects’ responses to surveys reveal their explicit reasoning, not their underlying information-processing sequence, so these responses have no bearing at all on whether blaming cognition is, in general, paradigmatically irrational, i.e., rooted in criteria-irrelevant inputs such as implicit bias and affect.

  1. Introduction

Much has been written on the nature, norms, conditions, and psychology of blame, but until recently there were no neurocognitive models of this capacity. Malle, Guglielmo, and Monroe (2014)[1] have filled this gap by offering a compelling model of the cognitive architecture of blame, which they define as a moral-social judgment that regulates social behaviour. This definition of blame is consistent with Strawsonian (interpersonal, functionalist) theories of blame defended by the likes of Michael McKenna (2012), Victoria McGeer (2013), Macalaster Bell (2013), and Angela Smith (2013). The Path Model (henceforth, Path for short) is therefore a fruitful point of entry for thinking about the cognitive architecture of blame, and for considering whether an information-processing model accurately maps the neurocognitive processes involved in blaming.

The model is called Path because it identifies hierarchically-structured information-processing pathways that play a role in paradigmatic blaming judgments. On Path, blamers typically (1) detect a norm-violating event E, (2) evaluate E for agent causation (versus object causation), (3) evaluate whether E was intentional, and then investigate (4a) the agent’s reasons for causing the event (if E was intentional), or (4b) the agent’s obligation to have prevented E from occurring, and (4c) the agent’s capacity to have prevented E from occurring (if E was unintentional). These inputs give rise to a dichotomous yes-or-no or a scalar more-or-less judgment of blame, depending on the informational inputs. This sequence is represented in the graph below:




Path is supported by evidence that ordinary judgments of blame involve the processing of criterial information concerning agent causation, intentionality, reasons, obligation, and capacity, in the suggested sequential order. Path allows that the criterial information can be processed quickly and automatically, or slowly and reflectively, but in either case it follows the same sequence.

Path has elicited various criticisms, including that it does not sufficiently explain judgments of transgression severity, it involves overly-stringent criteria, its classification of morally-eligible agents is too narrow (because it excludes non-human animals), it too quickly discounts blame-first models on which blame is typically fast and automatic as opposed to slow and deliberate (Goodwin 2014), it treats victim-blaming as rational, and (relatedly) it omits the role of ideology and motivated cognition in blaming (Niemi & Young 2014). These criticisms can perhaps be accommodated by the stipulation that Path represents paradigmatic blaming cognition, not every type of blaming response. Yet research shows that victim-blaming is fairly common, calling into question whether Path’s criterial information-processing sequence is, indeed, typical.

In this paper, I build on these criticisms by highlighting likely causal pathways for two ‘extra-evidential’ (i.e., criteria-irrelevant) inputs into the Path system: implicit bias and implicit affect. These causal factors, I argue, tend to drive information-processing down the ‘wrong’ pathways – that is, pathways unsupported by, and contrary to, objective evidence – giving rise to irrational (criteria-irrelevant) blaming judgments. I also argue that implicit states may preempt irrational blaming judgments non-sequentially (bypassing all sequential information-processing), and these ‘blame-first’ responses may be quite common, depending on the individual’s cognitive architecture and the saliency of triggers. (Different people have different levels of implicit bias, and implicit biases are activated in response to triggers such as visible demographic attributes). The characteristic role of implicit bias and implicit affect in the Path system, I argue, is to generate irrational blaming judgments, and these judgments are more common than the un-amended Path Model suggests. The idea that irrational blame is common is supported by the fact that implicit bias and implicit affect play a substantive role in social cognition (Nosek et al. 2011; Danziger et al. 2010) – for example, mediating mirror neuron activation and mimicry (Gallese et al. 2004), empathy (Mathur et al. 2014), and prosocial behaviour (Stepanikova et al. 2011). Implicit bias and implicit affect typically conflict with the agent’s explicit and rational beliefs.

The claim that we often, and perhaps paradigmatically, blame people irrationally is hardly surprising, given that (1) we live in conditions of inequality in which most people have implicit biases that track stereotypical representations of historically disenfranchised groups (women, African Americans, sexual minorities, etc.); (2) these biases mediate social cognition (Nosek et al. 2011); and (3) implicit affect mediates judgments of guilt (e.g., Danziger et al. 2010). By mapping these implicit inputs onto the Path system, we gain insight into the characteristic distorting role of implicit states in standard blaming cognition, embedded in conditions of inequality.

In what follows, I outline Malle, Guglielmo, and Monroe’s defense of Path, including important caveats that they make (section 2); I show how implicit biases can distort sequential reasoning at the ‘capacity,’ ‘obligation,’ and ‘reason’ stage of information processing (section 3); I show that affect can generate distorted judgments of blame and proportionality (section 4); I argue that implicit states can potentially trigger non-sequential (blame-first) judgments, making these judgments less computationally taxing over time, and thus more habitual (section 5); I argue that Path represents explicit moral reasoning, which does not track computational processing in neural systems – processing that may include implicit and criteria-irrelevant inputs (section 6); and I contend that, because Path does not chart the role of implicit bias in blaming cognition, it more closely reflects the paradigmatic blaming cognition of our early evolutionary ancestors (who enjoyed relatively egalitarian conditions), as opposed to modern-day humans who inhabit conditions of extreme inequality (section 7). Charting implicit states on the Path Model provides, I believe, a more accurate representation of our (largely irrational) blaming practices.

  1. Path: clarifications and caveats

It would be wrong to say that Malle et. al. deny that affect and implicit bias play any role in blaming cognition. Although they do not mention implicit bias, they admit that Path is incomplete in various ways, including that it does not show “the role and impact of affect on the information-processing chain” (177). They grant that affect is a mediating factor that “signals [the] seriousness” of the response (171) and potentially “amplifies” moral judgments (166). Psychopaths and patients with lesions in their ventromedial prefrontal cortex (resulting in ‘acquired psychopathy’) can recognize wrongdoing, but do not care about it (153). These agents cannot blame others interpersonally, because they merely impartially register that wrongdoing has been done; they do not engage in interpersonal, regulatory blaming. The authors emphasize that paradigmatic blame is not merely a private judgment, but a social interaction that serves to regulate social behaviour (149). Affect, then, plays a role in blaming cognition in neurotypical humans.

That said, Path favours “thought over emotional intensity” (171), in part because intense emotions tend to produce “less paradigmatic” blaming responses, such as yelling and personal attacks, as opposed to “more paradigmatic” blaming responses, such as rebuking, reproaching, and accusing (171). “Paradigmatic blame” is picked out by subjects’ pretheoretical intuitions about the “social acceptability” and “similarly to blame” of various moral criticisms; the most paradigmatic cases are “rational, calm, [and] constructive criticism” (171). This is why Path shows only criterial information – it charts the sequential decision-making processes involved in explicit moral reasoning. Non-criterial inputs and non-conscious processes are not figured in the diagram.

In addition to allowing that affect can play a limited role in blame, Malle et al. admit that biases can also play a limited role in blame, seeing that human judgment is “defeasible,” “fallible,” and influenced by emotion, motivation, and context (152). That said, the authors reject influential theories that purport to show that biases characteristically distort sequential reasoning; for instance, ‘outcome bias,’ in which worse action-outcomes bias blamers in favour of harsher judgments, and ‘character bias,’ in which dislikable character traits bias blamers in favour of harsher jugdments, are rejected in favour of the explanation that subjects infer that worse outcomes and character flaws are caused by negligence and extreme motives in the wrongdoer’s intentions, which is an ‘evidential’ (criteria-relevant) input at the ‘intentionality’ stage of the path sequence, not an ‘extra-evidential’ input (161). Hence, these putative biases are not biases at all, but criteria-relevant inputs that support the blaming judgment. This revisionary interpretation supports the view that blaming judgments are typically produced by the sequential processing of criteria-relevant information. Thus, these judgments are ‘rational’ in the sense that they are supported by evidence internal to the blaming system – evidence that inductively supports the judgment.

That said, Malle et al. allow that biases can enter the Path Model, albeit under special circumstances, particularly conditions that favour parsimonious computation. Blamers tend to make parsimonious inferences that minimize computational load under pressure, and these inferences are inappropriate if the value of parsimony is given more weight than the value of sufficient information and inferential accuracy. Sometimes, too, people rely on heuristics or ‘preset values’ activated by existing knowledge structures to generate fast, computationally simple blaming judgment. If these heuristics are unreliable or criteria-irrelevant, they may give rise to biased judgments. The use of heuristics is sometimes necessary due to situational pressures, but heuristics do not, according to Malle et al., play a characteristic role in blaming cognition.

As noted above, Path is impartial about whether information processing is conscious or unconscious, automatic or controlled. While some processes may be deliberate, others are automatic. For example, upon witnessing a bank robbery, I might automatically judge that the robber is blameworthy, because the relevant information (re. agent causation, intentionality, etc.) is readily available and the needed inferences are computationally simple. Ambiguous cases, by contrast, involve slower and more deliberate computational processing. (Was the bank robber acting voluntarily or under duress? If I am not sure, I will explicitly reflect on the ‘intentionality’ criterion). Automatic processing still follows the path sequence, but this method is more prone to error.

Malle et al. also admit that intuitions can play a role in the path system, in much the same way that heuristics can. For example, people are highly sensitive to the perception of negative events, which activates attention and linguistic centers, giving rise to rapid evaluative responses, (i.e., intuitions) (153). These intuitions ‘code’ the negative event as particularly serious. Intuitions, then, enter the path system at the ‘event detection’ stage, generating a stronger downstream judgment of blame. Malle et al. reject that idea that intuitions can produce paradigmatic blaming judgments on their own, independent of the information-processing stages, in opposition to Haidt’s ‘social intuitionist model,’ which holds that moral judgments are typically direct and non-sequential (2001). In section 4, I will argue that Malle et al. give us no good reason to think that criterial information-processing is a paradigmatic case of blaming cognition.

Although Malle et al. reject the social intuition model, Path does allow for intuitions, emotions, and biases to enter the path system when the context favours parsimonious computation. These allowances somewhat insulate the model from the obvious criticism that it is ‘too rational’; the authors can simply respond that it is incomplete, and criteria-irrelevant inputs (e.g., affect) are recognized as mediating factors, even if they are not ‘paradigmatic’ causes of blame.

The rationality objection is also amenable to the reply that Path models “information integration,” “not rationality constraints” (Malle et al. 2014: 177); the rationality objection is, in effect, a category mistake. To this, one can object that the integration of information can still generate irrational judgments, if the informational content misrepresents the world, driving informational processing down the (objectively) wrong pathway(s). For instance, if the blamer incorrectly perceives an agent as having an obligation to prevent a norm violation, the encoding of incorrect information at the ‘obligation’ stage will drive information-processing down the wrong pathway, to the ‘capacity’ stage, instead of generating a ‘no/low blame’ judgment. The role of distorting inputs can give rise to false and oppressive blaming habits, particularly when these inputs are implicit. This is not to say that the sequences in Path are out of order, but it is to say that non-evidential information – which does not provide a reason to blame – enters the Path system at various criterial stages and diverts information-processing down the objectively wrong pathways.

Although the final blaming judgment is supported by information internal to the system, the judgment is still irrational in the sense that it misrepresents its objects (salient events, capacities, obligations, etc.), and in the sense that it conflicts with the blamer’s practical commitments and explicit beliefs, such as the commitment to judge people on the basis of relevant criteria, to successfully navigating social relationships, and so on. (Implicit biases, notaby, conflict with the agent’s explicit beliefs by their very nature). A mismatch between (1) a judgment and the world, and/or (2) a judgment and the agent’s other beliefs and commitments, renders a judgment irrational on many theories of rationality (e.g., Brogaard 2014). It is plausible, then, to think that blaming judgments based on non-criterial inputs are irrational. Even though the judgment arises from sequential processing over criterial information (in some cases), the role of criteria-irrelevant implicit inputs drives sequential processing down the wrong pathways, giving rise to substantively irrational judgments. To illustrate: if I judge a person to be blameworthy due to implicit racial bias in my cognitive architecture, I blame the person on the basis of criteria-irrelevant content (racial bias), which I process as if it were relevant to one or more criterial stage(s).

Whether irrational blaming judgments are common is a substantive empirical question. I will argue that social psychology research and evolutionary psychology support the view that irrational blame is fairly common in our society, although blaming habits differ from one person to another.

In the next section, I will outline like points of entry for implicit biases in the Path system. If we chart these inputs on the Path Model, we shed light on the myriad ways in which blaming cognition can go wrong. This in turn helps to explain the pervasiveness of irrational blaming in our culture. For brevity, I will focus on only three criterial stages: capacity, obligations, and reasons.

  1. Criteria-irrelevant implicit biases

3.1. Capacity

‘Capacity’ in the Path Model refers to the capacity to prevent negative outcomes. Malle et al. give a few examples of this stage in the path sequence. First, they say that obesity, if caused by an uncontrollable medical condition, is perceived as not blameworthy (Weiner 1995, cited in Malle et al. 2014: 155). Second, if a “rape victim” is perceived as having been capable of taking precautions to avoid rape, she is deemed blameworthy (Davis et al. 1996, cited in Malle et al. 2014: 155). Third, if a driver’s capacity to have taken a different course of action is made salient, the driver is seen as blameworthy for causing a subsequent accident. In each case, if the capacity condition is met (e.g., if the obesity is controllable), the person is judged ‘blameworthy/high in blameworthiness.’

It is telling that two out of three of the prototypical judgments made by respondents are putatively biased. Contemporary feminists like Roxane Gay (2017) would say that obesity is not a moral failing; rather, the perception of obesity as a moral failing is motivated by weight bias – in many cases, implicit weight bias, which is as common as implicit racial bias (i.e., pervasive) (Skinner et al. 2017; KISRE 2015). Implicit weight bias can give rise to the false perception that people can and should exercise the capacity to avoid getting fat, although no such obligation exists, and it is far from clear that such a capacity exists in ordinary people (according to Gay). Implicit weight bias is not relevant to the ‘capacity’ criterion, but it drives information-processing down the ‘capacity’ path to a ‘blame’ judgment. This judgment is unwarranted (albeit supported by false criterial beliefs), in that it misrepresents the obligations and capacities of its target, and it conflicts with the explicit beliefs of the blamer (about her motivational profile and values).

Alternately, the critical mistake may have occurred at the ‘event detection’ stage, where the event (being fat) was coded as morally significant as opposed to morally irrelevant, triggering information-processing over the (objectively irrelevant) criterial stages of ‘agent causality,’ ‘intentionality,’ ‘obligation,’ and ‘capacity.’ Implicit bias likely triggers incorrect processing all the way down the line, with the misperception of the event injecting false information into the system.

Notably, if the blamer were to explain her judgment to another person, she would only have access to the explicit informational contents of her cognitive judgment, not the implicit content – the implicit bias. Hence, she would have no choice but to offer an incomplete explanation of her judgment, as opposed to explicating the salient causal mechanisms. If implicit states play a paradigmatic role in blaming cognition, then people’s self-reports of their information-processing sequence will typically omit salient mechanisms. In some cases, blamers will confabulate criterial explanations. For instance, when people discriminate against overweight people, they tend to invoke false criteria-relevant information in defense of their judgments e.g., the person was ‘incompetent’ (Levine et al. 2015). The criterial information misrepresents the agent’s qualities.

Malle et al. hold that in order to blame people, we must have “access to the informational basis of [our] blame judgments” (149). Prima facie, this seems to suggest that we must have access to the causal mechanisms that produce our blaming judgments, but this can’t be right, since those mechanisms can be implicit and unavailable to consciousness. Thus, ‘real’ blaming judgments include post hoc rationalizations and confabulations, completely detached from causally salient information-processing mechanisms. Confabulatory blaming judgments, moreover, are public, social, and regulatory in nature (although they have negative social consequences), so they function in much the same way as criteria-relevant blaming judgments – via the exchange of reasons generated via (implicit and explicit) processes of social cognition. Judgments induced by implicit bias, then, are legitimate, albeit substantively irrational, cases of blame.

Similar considerations apply to the blaming of the rape victim: the view that a person can proactively prevent rape is false and mediated by implicit biases. Men shown videos with high levels of female sexual objectification exhibit high implicit rape-myth acceptance, and also tend to excuse the rapist and blame the victim when asked to respond to a hypothetical rape scenario (Melina & Sandra 2012). The likely explanation is that the activated implicit states generate a false judgment about the capacity of the rape victim (to avoid getting raped), or a false judgment about the cause of the event (the victim’s behaviour). The salient cause of the judgment is, unbeknownst to the respondent, implicit rape-myth acceptance. This implicit state is criteria-irrelevant, but it generates a false criterial belief, giving rise to an irrational downstream blaming judgment. The researchers see this as a real, but biased, case of blame, which I think is the best (functional) explanation, since the blaming response is generated by social cognition and overtly expressed.

Implicit racial bias can also influence processing at the ‘capacity’ stage. Research shows that adults view Black girls as less innocent and more adult-like than white girls, resulting in more frequent and severe punishments in schools and the juvenile justice system (Epstein et al. 2015). Excessive and unnecessary blaming judgments about Black girls tend to be activated by implicit racial bias. This can result in the incorrect perception that Black girls are capable of preventing negative outcomes, because Black girls are seen as having capacities similar to adults. Biased blamers, then, may process information incorrectly at the ‘capacity’ stage due to activated implicit racial bias. This gives rise to false and excessively harsh blame directed toward Black girls.

Notably, implicit biases are difficult to mitigate or extinguish, in part because they are automatic and implicit reactions to salient demographic attributes and situational cues. They are, on one model, akin to habits, which are enduring because they are only partially cognitively penetrable (Devine 1989; Devine et al. 2012; Horoyd 2012). On dual systems theory, implicit biases are part of the fast, implicit processing system, which is “highly contextual and only changes in an enduring way after considerable time, effort, and/or intensity of experience” (Devine et al. 2012: 1268). Remediating interventions can be effective, but only if the agent is motivated to learn about activating contexts and committed to practicing self-regulation (ibid.). It is reasonable to infer that, when implicit biases enter the path system, driving informational processing down the same incorrect pathway(s) over and over in response to triggers, those connections become less computationally taxing and more accessible. Thus, the processing sequence becomes a habitual response, forming a ‘paradigmatic’ processing pattern. In people high in implicit bias, then, paradigmatic blaming responses may be habitual patterns, triggered by the saliency of implicit biases and well-worn pathways in response to frequently-encountered triggers.

3.2. Obligation

‘Obligation’ in Path refers to an obligation to have prevented a norm-violating event. We have already seen examples of processing over (false) obligation information, leading to irrational blaming judgments (e.g., of fat people). There are many other examples of obligation misperceptions. To give just one: most people assumed that wives had an obligation to have sex with their husbands until at least the late 1980s, when marital rape was finally criminalized in most (but not all) states. The tendency to blame spousal rape victims is mediated by implicit rape-myth acceptance and implicit gender bias (Grubb & Turner 2012). Thus, people who used to blame wives for spousal rape on the basis of misperceptions of obligation and capacity – for example, the misperception that wives have a duty to sexually gratify their husbands, and that doing so will prevent spousal rape – are irrationally blaming on the basis of implicit bias. This blaming response was culturally normative, revealing that irrational judgments can be paradigmatic. To this day, there is pervasive ignorance about consent and the culpability of rapists.

3.3. Reasons

‘Reasons’ refers to an agent’s reasons for acting. Processing at this stage can also be distorted by implicit bias. To give an example, African American men face harsher sentences than white men for the same federal crimes, controlling for criminal history, age, education, and citizenship (Pryor Jr. et al. 2017). By contrast, we tend to see adult white men as younger and more innocent than they really are (Fang 2017). What accounts for this discrepancy? A likely explanation is that judges are more likely to impute guilty intentions (mens rea) to Black men, and thus to encode false information at the ‘reasons’ stage. Implicit racial bias, then, can enter blaming cognition at the ‘reasons’ stage, producing false and excessively harsh blaming judgments toward Black men.

Once again, only the criterial information is introspectively available to the blamer, so the blamer’s explanation will not track the judgment’s underlying causal mechanisms (implicit racial bias). In people high in implicit racial bias, irrational blaming responses may be habitual response patterns. These blaming patterns are irrational, but an entrenched part of the blaming system.

3.4. Intermediary conclusions

Charting the likely role of implicit biases in blaming cognition illustrates that blaming judgments can be irrational just in case they misrepresent their object and/or conflict with the blamer’s practical commitments and explicit beliefs. But this does not yet show that blaming judgments are paradigmatically irrational in general. Admittedly, it is difficult to quantify the role of implicit bias in blaming cognition in the general population, especially given that implicit states interact with each person’s broader cognitive architecture and cultural factors, resulting in significant intersubjective differences. That said, most white people score positively for implicit bias on implicit association tests, and there is evidence that implicit biases are valid and predictive constructs that give rise to measurable prejudice effects (Greenwald et al. 1995; Green et al. 2007; Bertrand & Mullainathan 2004). Importantly, however, the strength and saliency of implicit biases depends on the agent’s cognitive profile and environment. The most that we can say is that for many people, when triggers are salient, the ‘paradigmatic blaming judgment’ will be an irrational one. This is a more nuanced view of the ‘paradigmatic’ operation of the blaming system than the idealized information-processing view, which only charts explicit beliefs. Our implicit states, informed by cultural norms, can be much darker than our explicit value judgments.

We cannot reasonably say that blaming judgments must provide accurate explanations of the information-processing sequence involved in blaming, since, even on the Path Model, these processes can be implicit, automatic, and unavailable to consciousness. Thus, blaming responses that involve confabulations, post hoc rationalizations, and other misrepresentations, must be seen as genuine blame. (The alternative would be to say that we can never know when anyone is blaming anyone). Moreover, confabulatory blaming responses satisfy the conditions of being public, social, and regulatory, even though they do not promote the goals of social cohesion and trust.

  1. Criteria-irrelevant affective states

Affective states also play a characteristic role in blaming cognition, and while this role is not always distorting – indeed, affect is likely required to code transgressions as morally salient, as Malle et al. admit – it often plays a distorting role, particularly when the affective input is implicit.

We can see this in judicial decision-making, at the stage of judging whether the defendant is culpable, and at the stage of determining sentencing. As Terry Maroney observes, affective mediation is:

observed both when the juror’s emotion is integral to the case—that is, prompted by case-relevant information, such as gruesome testimony about a victim’s injuries—[and] incidental to the case—that is, prompted by information extrinsic to the juror’s judgment task, such as anger at a fellow juror for rude behavior, anxiety over being pulled away from work and home obligations, or disgust with unsanitary courtroom conditions” (Feigenson 2010:52).

In the second case (incidental causes), the affective input is criteria-irrelevant, but gives rise to an overt judgment of culpability. When a juror, for instance, judges a defendant guilty (or especially guilty) due to anger toward another juror, the affective input is irrelevant to the obligations, capacities, or reasons of the defendant, but that input is causally efficacious nonetheless. Research shows that mock jurors primed with anger cues are likely to judge the defendant with more certainty, are less likely to evaluate evidence systematically, are more likely to rely on heuristics, and they tend to consider less information when coming to a decision (Tiedens and Linton, 2001; Lerner, Goldberg, & Tetlock, 1998). In some cases, anger may activate mediating heuristics. For instance, anger may trigger the use of a representative heuristic – a common decision-making short-cut – inducing the blamer to judge someone as a ‘representative criminal’ (or not) based on salient cultural stereotypes and the target’s visible attributes. This heuristic is not activated by parsimony constraints, but by affect under ordinary conditions. Even if the activated emotions are consciously available to the blamer (e.g., she is aware that she is angry), the causal antecedent of the anger may not be; that is, she may not realize that she is blaming the defendant because she is angry at someone else. This type of affective ignorance is a standard feature of human psychology, not a context-specific processing error (Dan Haybron 2008). Hence, affectively-triggered, irrational blaming responses may be quite common, particularly in people lacking emotional regulation and source attribution competency. In these groups, irrational, emotionally-charged blame may be a habitual response pattern.

  1. Non-sequential, fast blame

Haidt provides an alternative to the Path model. On his view, blame is typically generated quickly and automatically by intuitions. Malle et al. call this a ‘blame-first’ theory, and they deny that blame-first theories can explain paradigmatic cases of blame. This is because blame-first responses are triggered by first-personally opaque causal mechanisms (e.g., intuitions), which cannot be articulated in public blaming exchanges; thus, they cannot regulate social interactions through the exchange of criterial information. According to Malle et al., the social intuition model (henceforth, SIM) explains ‘moral judgments,’ not blame as a regulatory social process.

A notable case of intuitive blaming is moral dumbfounding, in which the blamer generates an explicit moral judgment on the basis of intuitions that are unconscious. In dumbfounding cases, there is a direct link from the intuition to the moral judgment, such that the production of the judgment bypasses criterial information-processing. The intuition, that is, preempts a judgment automatically and non-sequentially. Malle et al. object that intuitive moral judgments are not instances of blame, but rather of moral judgments (about the moral valence of actions), whereas blame involves an exchange of reasons. Since dumbfounded agents cannot articulate the “informational basis of their blame judgments (e.g., inferences of intentionality or preventability [etc.]),” they cannot “demand, offer, and negotiate such information as warrants for [our] acts of blaming” (Malle et al. 2014: 149).

This objection is not quite fair to SIM, seeing that dumbfounded blamers typically eventually offer reasons in support of their judgments, even if these reasons are post hoc and possibly confabulatory in nature. The blamer’s overt explanation of her judgment fails to track the judgment’s underlying computational mechanisms. It is often a rational reconstruction of those processes, based on available evidence and the agent’s subjective goals. This is no different from path-generated sequential judgments, in the sense that sequential processing can also be implicit and automatic, and thus introspectively opaque. Thus, our explanation of our judgment is likely to involve rational reconstruction in both cases. Malle et al. give us no reason to think that blamers’ self-reports of their blaming judgments accurately track the underlying causal mechanisms. Thus, we have no good reason to differentiate between intuitive judgments explained by post hoc rationalizations, and sequential-criterial judgments that look identical from the blamer’s perspective. Although intuitive judgments are generated by criteria-irrelevant inputs, we have seen that sequential information-processing also typically implicates criteria-irrelevant inputs, with the public explanation being divorced from underlying causal mechanisms. Thus, the distinction between intuitive judgments and sequential judgments is unmotivated.

It is also reasonable to think that implicit biases, like intuitions, might give rise to blame-first responses. To see the parallels between the two cases, consider a typical case of moral dumbfounding: when people are asked if it is wrong for a brother and sister to sleep together even once, assuming that there are no adverse consequences, many people intuitively judge the action to be wrong and the actors to be blameworthy (Haidt 2000). If asked to explain why, they provide post hoc rationalizations that invoke relevant (but causally inert) criteria. Similarly, people who form judgments on the basis of implicit bias tend to offer confabulated explanations. For instance, when subjects were separated into two groups and asked to evaluate a male and female candidate for police chief, where the male candidate was described as ‘street smart’ but lacking in formal education to the first group, and the female candidate was described as ‘street smart’ but lacking in formal education two the second group, both groups favoured the male candidate due to implicit gender bias. When asked to explain their choice, however, they invoked the male candidate’s qualifications (either street smarts or credentials), unaware that the qualification made no difference (Uhlmann & Cohen 2005). That is, they offered a post hoc rationalization. It is reasonable to think that implicit gender bias directly triggered a judgment about the candidates’ job suitability, similar to intuitions, with the criterial judgment being post hoc.

Admittedly, the study involved judgments of job suitability, but we can see how blame can be confabulatory in the same way: if someone were to blame a female police officer for incompetency on the basis of implicit gender bias, the person would be blaming on criteria-irrelevant grounds (implicit bias), but providing criterial information. Survey results provide no evidence to the effect that implicit states cannot generate blaming judgments non-sequentially, since these results track subjects’ explicit inferential processes, not their neurocognitive processes.

  1. What’s wrong? Path represents explicit, not implicit, reasoning

Path is based on abundant empirical evidence, and it correctly models explicit moral reasoning. The sequential steps are based on subjects’ responses to surveys. Evidence for the ‘capacity’ criterion, for example, comes from studies showing that foreseeability, or the capacity to predict, and potentially avoid, aversive outcomes, is deemed significant by subjects asked to judge the blameworthiness of theoretical actors in hypothetical scenarios. When the subjects were invited to ask questions about these scenarios, they typically asked questions about capacity prior to asking questions about obligation, revealing a default reasoning sequence from ‘capacity’ to ‘obligation.’ (The same result obtained when subjects were offered different types of information). The survey results supported the other ‘default’ pathways on the Path Model.

I have no doubt that Path accurately traces folks’ explicit reasoning about people’s blameworthiness, but it is a leap to say that it captures their actual information-processing sequence in their neurocognitive systems. This is because our first-person reports about our cognitive processes, and our actual cognitive processes, do not coincide. We lack reliable conscious access to our implicit states, and when we lack relevant information about our actions, we tend to confabulate intelligible but inaccurate explanations. According to Peter Carruthers (2010), all of our self-reports are inferential – none are direct – and thus we are just as likely to mis-attribute internal states to ourselves as we are to mis-attribute internal states to others, ceteris paribus. Carruthers also believes that confabulation is pervasive, based on converging research, including split-brain studies showing that the left hemisphere systematically confabulates intelligible explanations of our behaviour, and other data [Carruthers 2010: 39-42]). Other neuroscientists might not go this far, but most agree that a considerable amount of cognitive processing is unconscious (Bargh & Morsella 2008). We have no reason, then, to think that survey results about blame-relevant criterial information match the computational processes that produce judgments of blame. Indeed, given that people paradigmatically blame rape victims and fat people, it is reasonable to think that implicit biases play a paradigmatic role in generating blame.

Malle et al. admit that biases play a role in cases of parsimonious computation – a small number of cases – but given that social cognition is strongly mediated by inegalitarian cultural norms, it is much more plausible to think that implicit biases play a paradigmatic role in blaming cognition, which explains why blame typically targets and harms historically disenfranchised groups.

  1. Blame as an evolved function

Malle et al. say that blame’s cognitive architecture evolved as a mechanism for regulating social behaviour in our early evolutionary history. This cognitive architecture allowed us to express blame interpersonally in traditional, close-knit communities, regulating the behaviour of in-group members. In order for blame to serve this evolutionary function, they say, certain preconditions must be met, including (i) the overt communication of blame, (ii) the intelligible delivery of blame, (iii) the treatment of the offender as a member of a shared community, (iv) the possibility of repair, and (v) the presentation of reasons for acting (Malle et al. 2014: 172-173). Blaming judgments that meet these criteria are more likely to enhance the moral fitness of the group.

Elsewhere, I have disputed the idea that blame must be addressed primarily to the wrongdoer; I think that blame serves a variety of social functions independent of the wrongdoer’s responsiveness (or lack thereof). But here, I want to question whether blame paradigmatically serves its evolutionary function in modern times. If blame is often, and perhaps paradigmatically, irrational due to pervasive implicit bias, then blame most likely does not enhance moral-group fitness.

Malle et al. never say whether they think that blame functions to regulate behaviour in a fitness-enhancing way in the year 2018. They would presumably be amenable to the idea that blame does not effectively serve this function, insofar as the postulated preconditions are not met – specifically, blame is not communicated overtly, intelligibly, in such a way as to offer reasons to the wrongdoer, and so on. Perhaps these conditions have been compromised by modern cultural norms and technological innovations. Maybe we are more likely to ‘troll’ each other on the Internet, as opposed to confronting each other in civil discourse, contrary to the intelligibility constraint; maybe some of us are too politically oppressed to publicly express our blaming judgments, in violation of the communication constraint. Differences between the modern ecology and our ancestral ecology could potentially undermine the efficacy of modern blaming practices.

These are interesting departures from ancestral blaming norms, but I am more interested in (what I take to be) a more fundamental difference – the emergence of gross inequality, and the corresponding emergence of pervasive implicit bias. As I said above, implicit bias most likely plays a characteristic role in blaming cognition – for instance, mediating mirror neuron activation and mimicry (Spalding 2013), empathy (Mathur et al. 2014), and prosocial behaviour (Stpanikova et al. 2011). The content of our implicit biases is not innate – it is a result of cultural exposure. Because implicit biases are not introspectable, when we act on them, we tend to offer confabulatory reasons for our behaviour. Malle et al. should be amenable to the idea that the exchange of confabulatory reasons can be an impediment to the effective communication of blame, which is a precondition for effective blaming. Yet we often offer confabulatory reasons unwittingly, and we should not discount these common exchanges from the category of blame. These judgments regulate behaviour, even if they do not facilitate social cohesion and repair. If we discount these blaming judgments, we restrict the scope of blame to ‘sanitized’ judgements, leaving out a large proportion of our day-to-day blaming interactions (McGeer 2008). A psychologically realistic model of blame must include these ordinary cases of irrational blame. Sanitized blame is a blaming practice perpetuated by hyper-rational agents, or neurotypical agents living in conditions of extreme equality, not creatures like us, living in conditions like ours.

Malle et al. agree with me that our early evolutionary ecology was highly egalitarian compared to modern times. This means that there was relatively little implicit bias within the group. In modern pluralistic democracies, implicit bias is pervasive and highly impactful on the life prospects of historically disadvantaged groups within the broader collective. A precondition for a functioning, collective blaming practice, then, may be an increase of equality. More precisely, we may need to enhance equality in order for blaming cognition to function as it did in early evolutionary times, i.e., in such a way as to enhance social cohesion and repair of ruptured relationships within the collective. By omitting the role of implicit bias and affect from Path (even provisionally), the authors fail to capture the deep irrationality of modern blaming practices.

Equality, in my view, is a central precondition for a rational and reparative blaming practice. It may even be a precondition for preconditions such as community and repair, inasmuch as inegalitarian (irrational, biased) blaming practices tend to undermine mutual trust, cohesion, and respect.

  1. Concluding remarks

I have argued that Malle et al.’s model of blaming cognition charts explicit blaming judgments, but not underlying computational processes. Omitted from Path are implicit biases and implicit affect, which significantly mediate social cognition, giving rise to criteria-irrelevant, and thus irrational, blaming judgments (in the presence of triggers). I have proposed likely points of entry for implicit bias and affect into the path system, as well as proposing that cognitive biases may preempt blaming judgments on their own, independent of any sequential processing. Whether the Path Model captures ‘paradigmatic’ blame is not determined by survey responses, which trace the respondent’s explicit (possibly confabulatory) reasoning processes. It is eminently likely that implicit biases and affect play a systematic role in ordinary blaming cognition, though the saliency of these states depends on background cognitive and environmental conditions, which are highly subjective. Yet given the pervasiveness of victim-blaming judgments in our culture, it is likely that implicit bias plays a significant role in the blaming cognition of ordinary people. The prototypicality of irrational blame, moreover, helps to explain modern conditions of extreme inequality: those conditions are caused by, and reinforce, victim-blaming habits. Irrational blame and conditions of inequality, in other words, go hand in hand.

[1] Henceforth, references to Malle, Gugliemo, and Monroe (2014) will be denoted by ‘Malle et al.’ or, if the reference is obvious, only pagination.

Blame: It’s nature, targets, and functions


Blame: Its nature, targets, and functions.


This paper asks three questions: (1) What is holding-responsible, particularly blaming, as an expressive practice? (2) Who is an appropriate target of blame? And (3) What is the function of blame? I answer that (i) Blaming is an expressive practice, in which (somewhat surprisingly perhaps) blame can be expressed unconsciously by the blamer; (ii) the appropriate target of the reactive attitudes (resentment, disapprobation, indignation) is the wrongdoer, but the appropriate target of the conative force of blame – its demand for a response – is a suitably sensitive audience, not necessarily the wrongdoer; and (iii) one of blame’s primary functions is to enhance relational equality, because relational equality is a precondition for a fair and rational blaming practice.

Key words: responsibility; blame; reactive attitudes; implicit bias; eliminitivism

  1. Introduction

In this paper, I ask three questions: (1) What is holding-responsible, particularly blaming, as an expressive practice? (2) Who is an appropriate target of blame? And (3) What is the function of blame? (What is it for? Why do we do it? Why should we do it?).

I come to the following conclusions.

(1) Responsibility is an expression of the reactive attitudes, which, on various proposals, have emotional contents (emotionalism), cognitive contents (cognitivism), or both. I show that, because emotions and (cognitive) judgments can, in some cases, be unconscious, blame qua emotional/cognitive response can be unconscious. This is not necessarily a refutation of extant theories, but an under-appreciated implication of neuroscientific data on the nature of emotions, judgments, and implicit states implicated in emotional and cognitive activations. Although there is an abundance of scholarship on responsibility for unconscious states, very little has been published on unconscious blame, even though this type of blame is extremely consequential.

(2) Although Strawson says that moral incompetents are not apt targets of the reactive attitudes, I argue that moral incompetents can be apt targets of the emotional and cognitive contents of blame – resentment, disapprobation, negative valuation – but not apt targets of the conative orientation of, or demand implicit in, blame. Instead, the proper target of blame’s demands and solicitations is a suitably sensitive audience. Thus, blame’s emotional and cognitive contents, on the one hand, and blame’s conative orientation, on the other, come apart, and moral incompetents are only partially outside the scope of the reactive attitudes. This view vindicates contemporary theories on which blame can be fittingly addressed to the community (e.g., Bell 2014).

(3) Blame is most likely multi-functional, because versatile (multi-directional) blaming practices have more traction, potentially motivating more people to heed its demands. That said, surely one of the most urgent and immediate functions of our blaming practice is to enhance relational equality, or equality of moral and epistemic esteem and standing (Anderson 2015). The reason is that relational equality is a prerequisite to other posited aims – e.g., registering one’s protest against wrongdoing (Smith 2013), resisting wrongdoing (Bell 2013) – since our perception of wrongdoing is distorted by implicit biases rooted in pervasive cultural stereotypes. So, relational equality must be one of blame’s utmost normative functions – the function that it should, and must, serve, if it is to function fairly and rationally on a large-scale interpersonal level.

  1. Responsibility as an expressive practice: What is ‘an expression’?

One of the most influential accounts of moral responsibility in philosophy is Strawson’s view (1963), on which responsibility is an interpersonal practice in which we express the reactive attitudes, such as blame, praise, approbation, disapprobation, resentment, and so on. For this paper, I will focus on the negative reactive attitudes, particularly blame.

Many contemporary theorists construe Strawsonian holding-responsible as an expressive or communicative exchange, in which an agent expresses a reactive attitude toward another agent in response to the target’s quality of will (or other morally salient quality). McKenna (2014), for example, describes blame as part of a conversational practice with three paradigmatic stages: (1) agent A performs a morally problematic action, (2) agent B expresses a negative reactive attitude to A, and (3) A responds to B. In an ideal conversational exchange, A’s expression and B’s response will be fitting, meaning, in McKenna’s terms, intelligible to the recipient and appropriate to the context. If the wrongdoer is morally incompetent, she is not an apt target for blame.

Other theorists have different ideas about the expressive function of blame. Angela Smith argues that blame is an expressive practice that registers one’s protest against someone’s morally problematic treatment of oneself or others (2013). Macalister Bell holds that blame expresses a hostile attitude toward others, as a means of resisting wrongdoing (2013). She then identifies a number of sub-functions that subserve the resistance aim: registering the damage done to our relationships, educating its target, motivating offenders to avoid recidivism, educating members of the moral community, and expressing or affirming one’s own values. Christopher Bennett describes blame as a way of expressing disapproval through symbolic gestures that register the offender’s damage to important relationships (2013). Victoria McGeer describes blame as an expressive practice that produces prosocial behaviours by drawing attention to past offenses (2013: 171). These accounts differ in their construal of the function that expressions of blame paradigmatically serve, but they agree with the proposition that blame is an expressive practice.

Let’s assume that the expressivist view is right. At this juncture, we can ask: what does it mean to express blame?

This is a germane question because there is disagreement about what cognitive states are implicated in holding-responsible, and thus what cognitive states blame can express. Many expressivists think that emotions are a “canonical,” if not necessary, part of our responsibility practice (McGeer 2013: 168). McGeer argues that creatures like us, with our evolutionary history, cannot entirely expunge emotions, including ‘uncivil’ emotions, from blame. She sees blame as characteristically emotionally-charged, as well as “rapid and involuntary” (2013: 172), though (possibly) reflectively mediated and thus ‘domesticatable.’ Many people would urge that holding-responsible also implicates judgments or cognitive contents, which mediate (and potentially ‘domesticate’) our knee-jerk, affective responses (e.g., Smith 2013; Sher 2010; McKenna 2013). If we grant that reactive attitudes typically implicate emotional and/or cognitive contents, then (in either case) we might be tempted to assume that they are typically conscious, or at least consciously available. We know when we are blaming someone.  

(Many theorists hold that the reactive attitudes also have a ‘conative orientation,’ or an intention to produce some type of change or response in another person. I will hold off on discussing the conative view until the next section, where it will be more relevant).

In the first place, one might assume, as Williams James did, that an ‘unconscious emotion’ is a contradiction in terms (1884) – that there is a tangible ‘feeling-that’ quality to emotional states. On this assumption, unconscious blame is impossible. One might assume, too, that judgments are reflective in nature – that they involve a conscious representation of their object. If these assumptions are right, then perhaps we can be temporarily unaware of our blaming attitudes due to inattention or a transient distraction, but these states are nonetheless available to consciousness.

Here are some typical examples of blame that seem to affirm this picture. Someone loses confidence in his friend Joe after learning that Joe ridiculed him behind his back, and he modifies the relationship accordingly (Bennett 2012, citing Scanlon 2008). Isa blames her friend Marie for accepting a disempowering relationship with an abusive man (Bell 2014). A college students posts a ‘NO TOUCHY’ post-it on her pizza box to deter her friend from eating the leftovers, exemplifying ‘anticipatory blame’ (MacNamara 2014: 160). In all of these examples, unless the blamer is acting in a somnambulant state, we must imagine the person consciously blaming the target, i.e., expressing consciously-felt emotions and (most likely) consciously-represented judgments. For instance, Isa, if questioned, would probably say that she reproaches Marie, and disapproves of Marie’s decision to stay with her partner. She has conscious access to her blame-constitutive emotional and cognitive states.

Emotions and judgments, however, are not necessarily available to consciousness. (I’ll discuss emotions first and judgments shortly thereafter). Mounting neuroscientific evidence supports the existence of unconscious/implicit emotions, or states with the neurocognitive profile of an emotion that are not available to conscious awareness. Smith and Lane offer as evidence “emotionally relevant stimuli presented so briefly that perceptual awareness is not possible, [which nevertheless] reliably influence preferences (Zajonc, 1980), consummatory behavior (Winkielman and Berridge, 2004), and… other emotion-related physiological/ behavioral reactions (Tamietto and de Gelder, 2010)” (2016: 17). A growing literature, they add, “has also established that unconscious or implicit attitudes and beliefs have a profound effect on social behavior (McConnell and Leibold, 2001)” (2016: 217) – they are causally efficacious. Joseph LeDoux, a neuroscientist and writer, has argued that the activation of emotional circuits can trigger overt behaviour without giving rise to any conscious feeling, if those activations do not interact with neural systems implicated in conscious processing (2012). These activations are called ‘implicit emotions.’ To give some concrete examples, research subjects exhibit facial expressions, physiological arousal, and afferent feedback constitutive of fear, without subjectively representing fear (Khilstrom et al. 2000); in dissociation patients, conscious awareness of an emotion can be severed from conscious processes, rendering the emotion subjectively opaque (Bucci 2016); in patients with chronic health problems, activation in emotional circuits often presents as somatic symptoms with no conscious neural correlates – no consciousness of the emotion (Kroenke 2003; Konnopka et al., 2012; Sharpe and Carson, 2001).

These may not be canonical cases of emotions, but they are still valid cases, and these unconscious emotional activations can motivate overt behaviour. If there can be motivationally efficacious unconscious emotions, there can, on an affective view, be unconscious blame. For instance, if I implicitly fear, resent, or experience somatic distress in response to someone’s salient attributes, I may express those unconscious neurocognitive states toward their object (in negative ways fitting of the emotion) without realizing it. That is, I may (affectively) unconsciously blame the person. This doesn’t mean that unconscious blame is the norm, but it can happen.

Judgments can also be unconscious. Many theorists now see them as observable dispositions or functional states, which can be unconscious neurocognitive states. On the dispositionalist view, if I consciously believe that I value my friendship with Greg, but I have a disposition to avoid interacting with Greg, I may dispositionally judge Greg to be an inadequate friend, though I consciously represent him as a good friend (depending on relevant background factors). On the functionalist view, a judgment plays a reliable causal role in a certain pattern of behaviour. There is evidence that unconscious states can play a reliable role in producing goal-oriented behaviours. For example, unobtrusive priming of cooperation (using words like ‘dependable,’ ‘helpful,’ and ‘supportive’) caused subjects assigned to fisher roles in a resource-management game to cooperate to maintain the critical limit of 70 fish in the lake (Bragh et al. 2001: 1018). Notably, there was no difference between a group given unobtrusive primes, and a group given explicit instructions, suggesting that non-consciously activated goals can play just as reliable a role in the production of cooperative goal-oriented activity as the conscious intention to cooperate.

Similarly, unobtrusive priming of high-performance behaviour (using words like ‘win,’ ‘compete,’ and ‘success’) caused subjects to resume an intellectually demanding task (Scrabble) after a disruption, even when they were given the chance to play a less-demanding, ‘fun’ task instead (Bragh et al. 2001). This suggests that non-consciously activated goals are fairly robust: they persist over time. Indeed, non-consciously activated goals “show the same quality of persistence over time toward the desired end state, and of overcoming obstacles in the way, as… consciously set goals” (Bragh et al. 2001: 1015). That is, they have the quality of a disposition, and they can be just as causally efficacious at producing goal-oriented behaviour as conscious intentions.

On a dispositionalist picture, then, we can interpret the goal-pursuing behaviour provoked by non-conciously activated goals in each task as a judgment that cooperation/high performance (depending on the task) is worth pursuing, even if the judgment was not reflectively available to the actor at the time of action. Ditto for the functionalist interpretation. Granted, the value of the primed goal may have been consciously available to the subjects—they may have honestly affirmed that they did, e.g., consciously value cooperation in the resource-management task. Still, if one of them were to deny this, we would have reason to doubt their conscious appraisal, and favour the functionalist/dispostionalist interpretation on which they (implicitly) did value the goal.

On the cognitivist view, then, blame may sometimes manifest as an unconscious negative judgment. If I persistently fail to cooperate with certain types of people due to non-consciously activated states, perhaps I (functionally) negatively appraise, and thus blame, those people.

Another reason to doubt that judgments are necessarily conscious is that most people take judgments to be “intertwined” with emotions, both conscious and unconscious, meaning that judgments sometimes contain unconscious affective states (Chekov 2009).

Another compelling example of unconscious blame is blame motivated by implicit bias, where the normative significance of the action is first-personally opaque. Although implicit bias is a contested notion, the general consensus is that implicit biases are implicit associations between concepts, manifested in overt behaviour (but see Levy 2015 & Mandelbaum 2013 for exceptions, to be discussed shortly). Implicit racial bias, for example, is tracked by the Harvard Implicit Association Test by measuring subjects’ reaction times (‘response latencies’) to positive and negative words (‘smile,’ ‘honest,’ ‘disaster,’ ‘agony’) after priming the subject with a black face or a white face. Most white people are faster to pair black faces with negative words and white faces with positive words than vice versa. This means that they score positively for implicit racial bias.

It is debatable whether IAT scores translate into real-world discrimination against Black people (i.e., overt manifestations), but the reason many people think they do is that anti-black discrimination is a pervasive part of the cultural landscape, even though most people would (presumably sincerely) deny being racist. To give just a few examples of this phenomenon: Employers who had advertised an employment equity policy nonetheless favoured resumes with European American names over identical resumes with African American names by a ratio of 2:1 (Bertrand & Mullainathan 2004); managers discriminated against job applicants who used African American Vernacular English rather than Standard English in phone interviews, rating those applicants as less intelligent, less ambitious, and less qualified (Henderson 2001); adults view black girls as less innocent and more adult-like than white girls, resulting in more frequent and severe punishments in schools and the juvenile justice system (Epstein et al. 2015); African American men face harsher sentences than white men for the same federal crimes, controlling for criminal history, age, education, and citizenship (USSC 2017).

These studies show that America’s expressed attitude (i.e., overt disposition) toward black people as a group ranges from avoidant to punitive. We avoid renting to them, hiring them, and giving them equal educational opportunities; we punish them, even as young children, by barring them from equal participation in our most valued social institutions (schools, employment, housing, civilian society). Should we say that, if we engage in these avoidant, punitive, and generally exclusionary practices, implicitly rather than explicitly, they are not expressions of blame? If we are not committed to the requirement of consciousness, this constraint is unmotivated.

Neil Levy (2015) disputes the implicit-association model (as well as the implicit-belief model) and proposes that implicit biases are sui generis states, with some belief-like and some association-like properties. Like beliefs (but unlike implicit associations), implicit biases have some propositional structure, on his view, based on studies showing that previous exposure to counter-stereotypical images (women in a kitchen, men in an office) can produce counter-stereotypical responses on implicit association tests: subjects who had been exposed to stereotypical images were faster to associate competency/agency words with female faces than male faces (De Lemus et al. 2013). This suggests that implicit biases interact with the propositional structure of beliefs, albeit not inferentially – these states are still unconscious processes. Levy, in fact, holds that implicit biases, while not associations, are both unconscious, and motivationally efficacious (2015: 809). This theory, then, still supports the view that blame can be implicit, inasmuch as it can involve implicit biases that render its normative significance first-personally opaque. When we express or enact implicit biases, we do not realize that we are doing so, meaning that the evaluative significance of our behaviour is unavailable to reflective retrieval; we are not aware that are treating others in (often) aversive, avoidant, and punitive ways.

Notably, our emotions and judgments can be informed by implicit bias. Subjects exposed to a bad smell exhibited harsher moral judgments than controls, showing that implicit disgust can mediate explicit moral judgments (Schnall et al. 2008). Inducing disgust in subjects also produces intuitive disapproval of gay people, even in subjects who deny having any anti-gay bias (Inbar et al. 2018). In these cases, the subject is aware of harbouring aversive emotions and judgments, but not of the causal antecedents of those states. These are cases of ‘moral dumbfouning’ (Haidt 2001): the subject is aware of her emotion/judgment, but not of why she has it. On the other hand, if someone displays aversive behaviour in response to a prime without knowing it, the person’s dispositional emotion/judgment is first-personally opaque. This is a paradigmatic case of ‘unconscious blame,’ but moral-dumbfounding cases of blame are also possible.

We can define ‘unconscious blame’ as a negative overt response to another person, motivated by unconscious neurocognitive states, which are not directly available to consciousness, nor immediately reflectively retrievable. These negative responses are typically patterned because the implicit emotional, cognitive, and conceptual contents of blame, as we have seen, tend to be reliably activated by certain stimuli (unobtrusive primes, salient cultural stereotypes), and to persist in spite of disruptions. However, un-patterned responses should not necessarily be discounted as candidates for blame, even if they are not the norm. Expressions of unconscious blame are ‘negative’ or ‘aversive’ in that they express a ‘negative stance’ toward their target, as expressed in avoidant, punitive, or otherwise antisocial, patterns or instances of behaviour.

This account departs from the Strawsonian view that blame is a response to an agent’s quality of will, since blame in these cases responds to salient triggers and subjective states, not the target’s qualities. But this view allows that blame is typically a response to an agent’s perceived quality of will, mediated by the perceiver’s implicit and explicit emotions, judgments, and attitudes, and situational triggers. When blame is well-placed or ‘fitting,’ it accurately responds to the wrongdoer’s objective quality of will. In practice, blame frequently misses its mark due to implicit states. Unbiased blame is non-discriminatory, while biased blame is bigoted in various ways.

The idea that holding-responsible (including blaming) can be unconscious is not exactly revisionary. Expressivists in particular grant that our responsibility practice regularly involves manifestations of unconscious attitudes, though they focus on unconscious states at the ‘moral contribution’ stage (‘stage 1’ above). McKenna, for example, says that committing an unintentional offense is blame-able (2014); Smith holds that forgetting about a close friend’s birthday warrants moral approbation (2005); and Sher says that forgetting a family pet in the backseat of a hot car is blameworthy (2010). These are ‘omissions cases,’ in which an omission (to do what one ought, take proper precautions, exercise due moral vigilance, or what have you) reflects a negative quality of will, opening the person to blame. Many people now take this ‘anti-reflecitivist’ stance on responsible agency, according to which people can be responsible for unconscious transgressions (Doris 2016). That said, few theorists have discussed the role of unconscious states in stages two and three of our responsibility practice: the stage at which we express the reactive attitudes (‘moral address’), and the stage at which we respond to these attitudes (‘moral accounting’). This is not to say that theorists deny that we can express and respond to these attitudes unconsciously, but if they believe that we can, they don’t explicitly say so.

This points to an asymmetry in the literature, in which an abundance of work has been done on whether one can be blameworthy for an unconscious violation, whereas relatively little has been said about whether one can express and respond to the reactive attitudes unconsciously. If I am right, then we can, at least, unconsciously express these attitudes. When we unconsciously express a reactive attitude, we are unaware of some normatively salient feature of our behaviour.

Here is an example (taken from an earlier paper [citation withheld for anonymity]). Physicians prescribe fewer pain-killers to Black patients than white patients when both exhibit the same symptoms (Silverstein 2013; Cleeland et al. 1994). This may be related to the fact that white observers show less physiological arousal in response to Black people’s pain than white people’s pain, with their level of arousal correlating with their level of implicit racial bias (Forgiarini et al. 2011). That is, physicians may prescribe fewer pain-killers to Black patients due to implicit racial bias. If so, then they are, on scrutiny, treating Black patients in a punitive way, denying them the medical treatment they need or deserve on the physician’s evaluation of equivalent suffering in white patients. They may also distrust their Black patients’ ability to take pain-killers responsibly compared to their white patients.[1] If this analysis is right, then physicians who manifest racial prescription bias are unconsciously blaming (i.e., punishing, distrusting, implicitly judging as undeserving or incompetent) their Black patients. They are not consciously aware of doing this, but their overt behaviour nonetheless expresses blame to Black patients.

Again, I am not saying that theorists would not count this as an instance of expressive blame; I am saying that there is little discussion of such cases – cases that are extremely consequential in their impact on the life prospects of historically disadvantaged groups.

The view that implicitly biased prescription practices can count as blame requires a move away from an intuitive perspective on blame: the first-person standpoint. Physicians motivated by implicit racial bias are not aware that they are expressing these biases in their prescriptive habits, much less that they are unconsciously blaming their Black patients, so we cannot rely on the first-person standpoint to deliver accurate evaluative descriptions of the subject’s own expressive habits. (This is a rejection of what Sher [2010] calls ‘the searchlight view,’ applied here to blaming rather than to blameworthiness). We cannot necessarily rely on the average observer’s appraisal either, since most people are not in a good position to evaluate whether a perceived pattern of behaviour is an instance of implicit bias. To evaluate whether a person P’s pattern of behaviour counts as blame, we need to assess P’s stated intentions and overt behaviour, and compare these data to relevant psychological and sociological research or trends. We should, in a sense, adopt a ‘heterophenomological perspective’ to blame, in Dennett’s terms (1991) – a perspective that compares a piece of phenomenological data against a range of intersubjective and empirical data. A physician’s prescriptive practices can be evaluated for implicit bias by observing the person’s treatment of different patients, in comparison with national prescription trends. This allows us (if there is sufficient data) to determine if unconscious blame is explanatorily potent in a particular case. (Identifying unconscious blame is easier on a population level than an individual level – that is, it is easier to determine whether a certain social group unconsciously blames another social group, than whether a certain individual unconsciously blames another individual in response to the person’s identity or group affiliation. Nonetheless, the population trends are manifestations of individual patterns of behaviour).

This is a suggested methodology for identifying unconscious blame. It does not bear on the question of whether unconscious blame is a valid construct. This question has already been settled by the discussion about the nature of emotions and judgments, which can be implicit.

  1. Fittingness constraints: Communication to whom?

Assuming that responsibility is an expressive practice, another pertinent question is: What is its appropriate target?

Strawson (1963) argued that the reactive attitudes respond to an agent’s quality of will, unless there are salient extenuating conditions. Excusing conditions mollify the reactive attitudes, rendering them less severe, whereas exempting conditions trigger a complete suspension of the reactive attitudes, and a switch to ‘the objective attitude,’ i.e., an emotionally detached, typically avoidant, stance. Adopting this stance involves treating the target as, “perhaps, an object of social policy; as a subject for what, in a wide range of sense, might be called treatment; as something certainly to be taken account, perhaps precautionary account, of; to be managed or handled or cured or trained; perhaps simply to be avoided” (1963: 66). We refuse to engage emotionally with the target in the objective stance, and we aim to exclude the person from ‘the moral community,’ the group of morally-responsive agents. This is a fitting attitude toward moral incompetents, those who are deeply or constitutionally incapable of responding to moral claims.

If the objective attitude is an emotionally detached stance, then it excludes blame, and other reactive attitudes, from its scope. The ordinary defense for this position is that blame toward moral incompetents has no point. The point of blame, many assume, is to elicit a fitting emotional reaction in the target, in response to a salient moral demand. This is its ‘conative orientation’ (or one of its dominant conative orientations, if there are several): blame solicits a fitting response from the target. This conative aspect of blame can be retrospective or prospective, or both. Sher, for instance, describes the conative orientation of blame as a desire that the wrongdoer “have responded, or that he be disposed to respond, to what we consider a compelling moral reason” (2006: 105). There is a retrospective desire that the agent have behaved otherwise, a prospective desire that he (be disposed to) respond appropriately in the future. Others see blame as not merely desiring, but demanding, a response. We can call this a ‘strongly conative’ orientation: the blamer doesn’t just wish the wrongdoer had acted otherwise, she calls for action. This strong conative orientation is required, or presupposed, by many influential theories.

McKenna, for instance, construes blame as, in effect, a request for a ‘moral account’ from the transgressor (2014). Similarly, McNamara says that blame solicits a response from the wrongdoer, and that “successful” cases of blame are those that “receive a response” (2012: 159).

Others see this constraint as only partial, or defeasible. Smith’s protest account interprets blame as “implicitly seeking a response” from the object of protest (2013: 40), consistent with Strawson and McKenna. Absent this eliciting function, blame could be “a one-sided adjustment of attitudes,” and, thus, “deeply non-relational” (41) – an unacceptable conclusion. But Smith thinks that blame can also serve another function: to register a complaint against someone’s attitude – though this might be an ‘imperfect’ or ‘incomplete’ case. McGeer similarly argues that blame can serve two functions: regulating people’s behavior, and appraising an action as wrong. Blame, then, does not necessarily solicit a response from the wrongdoer – it can serve a purely signaling function. Bell (2013), too, thinks that blame can serve goals other than appealing to the wrongdoer: it can function to motivate and educate the community and to signal one’s values to the group.

These last three views (Smith’s, McGeer’s, and Bell’s), which we can call ‘multi-functional accounts of blame’ (because they see blame as serving more than one aim), mark a departure from the Strawsonian ‘fittingness’ constraint, on which blame is fitting only if the proper target – namely, the wrongdoer – is suitably sensitive to the demand. The justification for this departure is that blame’s proper ‘target’ is not necessarily, or exclusively, the wrongdoer; rather, the target of the conative orientation of blame, which seeks a response, must be some other individual or group capable of responding appropriately (with sympathy, vicarious indignation, political activism). The ‘conative target,’ in other words, is a sensitive moral audience, those to whom the expressive act is communicated, and from whom uptake is demanded, sought, or anticipated. This condition satisfies blame’s expressive ‘point’ – it communicates with a fit respondent. That said, the wrongdoer is surely an appropriate target of blame’s emotional/cognitive contents (resentment, distrust, disapproval), for, whom else could be? If we cannot extirpate the emotional contents of blame from human psychology, they must have some target, and this must be the wrongdoer. Yet the morally incompetent wrongdoer is not an apt target for blame’s conative orientation.

This multi-directional view of blame’s ‘point’ respects the relational element of Strawson’s theory, but it expands the scope of the reactive attitudes beyond the relationship between the complainant and the accused; it recognizes that blame can be expressed to the community, in relationship with the community. The incompetent wrongdoer, however, is not completely outside of blame’s scope; this person is an apt target of blame-constitutive emotions and judgments, though not of blame’s demand for a suitable response. That is, on my interpretation, the emotional/cognitive contents of blame, on the one hand, and the conative force of blame, on the other, come apart. In separating these two elements, we can see how the conative element of blame can be fittingly addressed to an audience, in keeping with the multi-functional view.

This interpretation implies that moral incompetents are not exempt from blame in the way that Strawson envisioned; they may be exempt from blame’s conative aspect, but they are still apt targets of blame’s emotional and cognitive contents. That is, when we adopt the objective attitude toward someone, we only partially suspend the reactive attitudes; we suspend a given attitude’s conative force, but we do not suspend the attitude’s emotional and cognitive contents. We are permitted to be resentful, reproachful, and censorious toward morally incompetent wrongdoers.

This is, I think, a more realistic view of blame’s psychological profile. Do the families of psychopathic (non-responsive on many views) murder victims suspend their resentment toward the psychopath? No, but they might turn to the community and the justice system for an appropriate response to their understandable emotions. (An appropriate response would be sympathetic resentment, sequestration, and other actions that ‘take the side of’ the victims and protect the community from harm. Note that there are additional fittingness constraints on blame and punishment, such as proportionality and intelligibility, which I do not have time to address here).

The multi-functional view also calls into question McKenna’s model. If the audience can be the target of blame’s demands, then the ‘moral accounting’ stage must be expanded to include a range of fitting responses from variously situated respondents. This is because the audience is not in a position to give a ‘moral account’ (apology, explanation, restitution) on behalf of the transgressor – only the transgressor himself can do this. Thus, the ‘moral accounting’ stage must be expanded to include other types of response – vicarious resentment, collective action, and so on.

This means that we can tweak Strawson and McKenna to make them compatible with a multi-functional view, by separating the emotional/cognitive and conative elements of blame. But why should we think, all things considered, that blame can be addressed to an audience?

There are normative and descriptive reasons on offer. The descriptive reason is that (1) this is how blame actually functions, and (2) we should endorse a descriptively accurate theory. McGeer provides an evolutionary (descriptive) account of blame, on which the reactive attitude evolved as rapid and involuntary responses to perceived transgressions. (However, as ‘mentalizing creatures,’ we are capable of reflecting on and revising these responses, within limits set by evolution). These evolved responses “play a critical role in regulating behaviour by way of making salient the demands that shared norms place on our actions and attitudes” (2014: 183). They promote pro-social behaviour by ‘coding’ certain practices as wrong, and ‘signaling’ disapproval of those practices. They can serve these ‘coding’ and ‘signaling’ functions in a variety of ways: not simply by demanding a response from a transgressor, but also by demanding recognition from the community. We are constrained in how far we can suppress and alter these responses, but we can implement plans and policies to help ‘guide’ them in pro-social ways. In sum, emotions, while somewhat plaint, are an inextricable part of our blaming psychology.

Notably, even if blame does not function well in modern society compared to our early evolutionary history (more on which shortly), it remains true that emotion-laden blame is an ineradicable part of our moral psychology, though it may ‘misfire’ due to differences between modern and ancient ecologies. The direction of blame and the existence of blame are different things: while we can perhaps ‘refine’ blame to minimize its harmful effects, we cannot eliminate it.

The normative reason in favour of multi-functionalism is that blame should serve a multitude of functions, as these functions help to regulate human behaviour in positive ways. Bell points out that blame can serve a variety of (local) functions (perhaps in the service of a general, over-arching function), aside from eliciting a reaction from the transgressor; it can also educate, motivate, and register a moral complaint with, the moral community. Indeed, if we think of holding-responsible as a functional practice that serves the moral ends of the community, it is, I think, arbitrary to cut off any local aims that subserve this goal. To be effective, expressive practices must be versatile, appealing to and ‘connecting with’ as many people as possible. Any number of specific aims, then, may be compatible with the general aim of “responsibilizing” people (Pettit 2007).

Bell advances this argument as a way of defeating the view that blame is ‘positional,’ or fitting only toward those over whom we have moral authority, rooted in our relationship with the blamee (viz., Darwall 2006, Cohen 2006). Against this view, Bell claims that we can blame strangers, co-conspirators, the deceased, and (as I have urged) moral incompetents. The rationale for this departure from the ‘positional’ view is that blame is multi-functional, and not all of its functions are indexed to authority relations. We have responsibilities, not only as members of specific relationships, but “as critics” and “third parties” (2012: 265). Indeed, we witness wrongdoing every day; the fact that we may not be in a position to solicit a response from the wrongdoer does not entail that we cannot, in a meaningful sense, blame the person. Arguably, when we ignore the transgressions of moral incompetents, we make ourselves complicit in their actions.

I agree with Bell’s objection to the ‘positionality’ constraint, insofar as I agree that morally insensitive agents can be resented, distrusted, and censured, but I do not think that we thereby need to abandon the intuitive idea that the conative orientation of blame must be directed to a responsive target. While the emotional and cognitive contents of blame are fitting for the transgressor, the proper recipient for the demand for a response may very well be someone else.

If this is right, then the answer to the question, ‘who is an apt target of blame?’ is broader and more nuanced than Strawson envisioned. Moral incompetents are apt targets of blame-constitutive emotions and judgments, but not of moral demands; these demands should be addressed to a suitably sensitive audience.

  1. Functionalism: Communication to what end?

We have seen a variety of functionalist interpretations. McGeer, Bell, and Smith agree that blame’s functions include (1) registering someone’s behaviour or quality of will as morally problematic, and (2) soliciting a response from the wrongdoer or the community. I have called these views ‘multifunctional’ simply because they allow that blame can function not merely to demand a response from the wrongdoer, but to do other things as well. These theorists, however, posit distinct higher-order aims that blame is supposed to serve, under which other (subordinate) aims can be subsumed. For Bell, blame’s main function is to “resist wrongdoing” (266); for Smith, it is to protest (or “register one’s protest of”) someone’s treatment of people (27); and for McGeer, it is to enhance the moral fitness of (or ‘responsibilize’) the group. These aims overlap: protest and resistance are similar – perhaps even coextensive – concepts, and the aim of responsibilizing the group surely encompasses resisting and protesting violations, whatever else it may entail.

Rather than arbitrating which of these views is superior, in this section I will argue that each is too thin to serve as a normatively adequate theory of blame, and that any adequate theory of blame must make relational equality a priority. The reason is that our perception of wrongdoing will be distorted as long as relational inequality is the norm. If our blaming responses register, protest, or resist ‘wrongdoing’ by our lights (as they must), they will systematically miss their mark, as ‘our lights’ are coloured by the conditions of social injustice in which we live (viz. Fricker 2007, Medina 2013). Repairing relational inequality, then, must be one of our explicit aims and priorities as members of the moral community. If we do not effectively prioritize this aim, we risk blaming people in unjust and irrational ways – ways that harm marginalized groups.

Bell and Smith posit similar aims for blame (protesting and resisting wrongdoing), while McGeer thinks that blame registers offenses and responsibilizes people. I don’t doubt that blame should do these things, but in our society, it does not advance these aims effectively. This is because our perception of wrongdoing is distorted by implicit biases informed by cultural stereotypes, collectively speaking. (Individuals are biased to a greater or lesser degree depending on their social position, learning history, neurocognitive profile, etc. What follows here is a population analysis, abstracting away, for the most part, from individual perception. This analysis is compatible with the idea that some people are moral-epistemic saints, impervious to implicit bias; but, on a standpoint epistemological picture, most people’s perception will be affected by implicit biases rooted in cultural stereotypes and scripts, since individual perception reflects background epistemic conditions, or what Jose Medina calls ‘the social imagination’ [2012]).

As we saw in section 2, physicians systematically distrust and punish Black patients; educators systematically punish and sanction young Black girls; employers systematically distrust and avoid Black job applicants; the justice system systematically gives Black men harsher prison sentences. These are examples of misplaced blame – specifically, misplaced unconscious blame – in which our (collective) blaming reactions subvert their putative function(s), viz., protesting and resisting wrongdoing, and responsibilizing people. These attitudes, in effect, malfunction in current social conditions, making society worse off and less responsible. We are less responsible qua blamers, blamees, and third-party critics, as we are more mired in distorting cultural stereotypes and scripts that prevent us from discerning who objectively deserves blame (or protest or avoidance), and who does not. The more our blaming practices miss the mark, the harder it is for us to accurately perceive wrongdoing, and to differentiate morally salient qualities from morally neutral demographic attributes that happen to be stigmatized by patriarchal-colonial-cissexist culture. The more likely we are, in other words, to illicitly blame and punish members of culturally disadvantaged groups, and to illicitly praise and reward members of socially privileged groups –  to get blame wrong. Meanwhile, the underprivileged are deprived of equal access to responsibilizing institutions, such as quality education, housing, and lucrative jobs, which is an affront to their dignity as persons. This not to say that these individuals are particularly epistemically flawed, but rather, it is to emphasize that they lack access to resources that enhance responsibility – for example, safe housing in which to responsibly raise children, occupations with which to responsibly pay off one’s student loans, etc.

Because our blaming practice is systematically biased against historically disenfranchised groups, some theorists have adopted an eliminativist position, arguing that we ought to eliminate blame because it does more harm than good (Waller 2016, Levy 2012). In other words, on a descriptively accurate account, taking proper account of modern social conditions, it is accurate to say that blame serves the function of unfairly punishing and oppressing minorities. Prima facie, this is the opposite of McGeer’s evolutionary account, on which blame serves to enhance moral-group fitness. But the two views are compatible if we see McGeer as talking about blame’s function in the Paleolithic era, in which societies were more egalitarian (Dyble et al. 2015), and Waller as talking about blame in modern times, in which inequality is systemic – particularly in America, the most economically unequal developed country (Allianz 2016). Perhaps blame’s evolved function has been co-opted by wealthy colonial-patriarchal-cissexistto advance the material and political interests of the elite. If so, then blame no longer serves its evolved function; it serves a man-made function: to promote and reinforce modern asymmetries of power.

I think that something like this story is right, but I agree with McGeer that blame is a canonical feature of human moral psychology, not something that we can eliminate, though we can ‘domesticate it’ with social engineering and careful planning. So, eliminativism is not a viable option, pace Waller. I also think that partially eliminating blame would be a mistake, as it would prevent us from responding sensitively to the suffering of the oppressed – something that we cannot and should not do. Since, on the most optimistic projection, we will not eliminate global inequality in our lifetime (indeed, income inequality in America is steadily growing (Saez 2016)) – it would be wrong, and probably psychologically impossibly, to completely suspend the reactive attitudes. It would be wrong because the detachment involved in a full suspension of the reactive attitudes would constitute callous indifference to the suffering of the oppressed; and achieving this state is most likely psychologically impossible, in any case, because we are not, as McGeer drives home, the kind of creatures who can sustain indifference in the face of undeserved suffering (with the exception of psychopaths). When we see a starving child, we feel both sympathy for the child and outrage against the person or people responsible for the child’s plight. This is how we are built.

Eliminativism also faces an aggregation problem, since not everyone can be expected to endorse it as a policy, and if only a few conscientious eliminativists suspended their blame (assuming this were possible), this would take the pressure of public condemnation off the worst offenders.

Rather than eliminating blame, I agree with McGeer that we should try to domesticate it, and I submit that this domesticating project involves promoting ‘relational equality,’ or equality of “authority, esteem, and standing” (Anderson 2015: 65). The reason is that equality is a prerequisite to holding people responsible in a fair and rational manner, given that inequality creates stereotypes and pernicious social scripts that impair our ability to recognize wrongdoing. Specifically, inequality creates implicit biases that prevent us, as a society, from distinguishing ‘wrong’ from ‘stigmatized,’ ‘blameworthy’ from ‘socially marginalized.’ Thus, relational equality must be an explicit aim, or end, of our blaming practice. This does not preclude the aims of protesting and resisting wrongdoing and registering wrongdoing, but it is temporally prior to the effective (perfect) attainment of these ends: if we do not eliminate inequality-based biases, we cannot accurately perceive people’s morally salient qualities and accurately determine who warrants blame. We do not need to completely eliminate inequality prior to blaming, but we must reduce the influence of inequality on our blaming practice if blame is to hit its mark every time.

I say ‘relational equality’ because I have been describing blame as a relational practice (between blamer and blamee, or blamer and society, or some combination of these), and relational equality, as envisioned by Elizabeth Anderson, is also a relational notion: it seeks to promote fair and equitable interactions across a variety of humanly-valuable dimensions. In particular, relational equality entails not only equality of resources, but also equality of epistemic and moral standing, esteem, and regard. To blame people fairly, we must blame them in a way befitting their (objective) moral and epistemic standing – a way that enhances relational equality. When we blame people on the basis of implicit bias, we unfairly morally disdain them; we treat them as morally and epistemically ‘beneath us.’ When we avoid, ostracize, or exclude people for no good reason, we treat them as morally unworthy of our interest, attention, and cooperation; we take ‘the objective attitude’ toward them without warrant, twisting it into a tool of objectification and marginalization. These misguided (but common) blaming practices are based on, and partly constituted by, relationships of unequal epistemic and moral standing. Notably, Anderson’s view is a response to classic egalitarian theories, which focus on distributing resources fairly, but which tend to treat the recipients with an attitude of “contemptuous pity” (Anderson 2000: 6), characterizing them as irresponsible, lazy, and poorly endowed. Anderson argues that the point of egalitarianism should be to allocate goods in the spirit of respect for human dignity, not to transfer funds from the ‘responsible’ to the ‘irresponsible,’ ‘lazy,’ and ‘pitiable’ – that is, she rejects the premise of classic egalitarian reasoning. I am suggesting that this should also be the guiding aim of blame: to distribute moral and epistemic regard fairly, so as to respect the objective moral and epistemic qualities of members of the moral community. Enhancing relational equality and enhancing the rationality of our blaming practice, on this view, go hand in hand.

Although I have been focusing on population-level blaming practices, the goal of enhancing relational equality can, and should, be implemented on both a personal and an institutional level. Individuals can take steps to try to mitigate their implicit biases, and institutions and governments can implement policies to reduce the adverse effects of implicit bias on the common good. As I have discussed these approaches elsewhere, I will not elaborate here, except to note that I believe that individuals should try to cultivate epistemic virtues (Fricker 2007) and vigilance (Murray 2017) to the greatest extent possible, while institutions and government should implement epistemically responsible protocols (Longing 2008). Even if the institutional approach is more efficacious (as McGeer [2013] suggests), the first method is indispensable, given that fully responsible social institutions are not on the horizon. While we are waiting for effective social engineering policies to come into effect, we should work on our own moral character.

  1. Conclusions

I have argued that blame functions to express emotional and cognitive states to the wrongdoer, but its demands are properly addressed to a suitably sensitive audience. Blame can be expressed unconsciously, as evidenced by neurocognitive, psychological, and sociological data. Because expressive blame as a dispositional response is influenced by implicit biases and other morally-irrelevant situational factors, we should strive to blame people in a way that enhances relational equality, or equality of standing, authority, and esteem. Thus, a prerequisite for rational blame, and one of its proper aims, must be relational equality. This might sound circular, but the same can be said of the heart: pumping oxygenated blood throughout the body is a prerequisite for a fit heart, and this is also the heart’s function. Similarly, relational equality is a prerequisite for a fit blaming practice, and it is also the function (or one of the utmost functions) of this practice. A fit moral community, in which people hold each other responsible on fair and rational grounds, aims to promote relational equality, and the more they do this, the closer they get to this aim.

[1] Distrust is a reactive attitude on McGeer’s view (2008).