History and Context of Campus Expression in North America

Although critiques of higher education predate the sociocultural upheavals of the 1960s (Buckley, Reference Buckley1951), this period is key to understanding the alleged campus crisis. An obvious exemplar is the Free Speech Movement on the University of California, Berkeley, campus in the early 1960s. Student activists, some of whom had recently returned from civil rights activism (volunteering in the Freedom Summer of 1964), were incensed that the Berkeley administration mostly forbade political expression on campus. Ensuing protests and sit-ins attracted thousands of students, eventually led to the administration relenting, and galvanized student movements across the country and around the world (Cohen, Reference Cohen2009). Student activists were demanding that the university loosen its expressive restrictions and sought to expose inconsistencies between the university's self-proclaimed mission to pursue truth and knowledge without constraint, its institutional rules and the capricious behaviour of administrators.

According to Mario Savio, the most recognized face of the Free Speech Movement, expressive freedom was “the most dangerous [right] . . . in the last analysis, because if the thing you tell people about is bad enough then it . . . leads immediately to advocacy, and action” (Cohen, Reference Cohen2009: 79). During negotiations, Savio was explicit about demanding legitimate reasons for the authority wielded by the university, especially when it made policy decisions with little to no discernable rationale (83). Thus, the Free Speech Movement practised a radical immanent critique of the institution, arguing that expressive restrictions were incompatible with its self-avowed mission (the pursuit of truth and knowledge). As a result of this and other student upheavals, campuses quickly became synonymous with the spectre of youthful protest in the popular imagination. Further, these movements contributed to the impression that campuses ought to be exemplars of expressive freedom by retaining their unique form of collegial autonomy and a singular mission.

Approximately five decades later, UC Berkeley was again the fulcrum of a debate about the state of expressive freedom on campus. In response to an event featuring Milo Yiannopoulos, far-right provocateur (and self-proclaimed “troll”), several hundred protestors descended upon Berkeley in early 2017. Some of the protestors used black bloc tactics, engaged in skirmishes with Yiannopoulos supporters and caused damage to (and then set fire to) campus and police property estimated at more than $100,000 (Fuller, Reference Fuller2017). The debacle resulted in a predictable chorus of criticism. Many noted the historical irony, which was obviously Yiannopoulos's strategy all along (Wong and Levin, Reference Wong and Levin2017). Even President Donald Trump chimed in on Twitter, warning that his administration could withhold federal funding if colleges and universities didn't protect expressive freedom. Two years later, he signed an executive order mandating just that, although the order merely underlines existing protections (Thomason, Reference Thomason2019).

Confrontations like this—between student protestors and edgy political-pundits- turned-entrepreneurs—are now commonplace, with the latter relishing an opportunity to portray the college campus as a bastion of progressive intolerance and to attract some sympathetic media coverage. Interestingly, student protestors were again practising a radical immanent critique of the institution. This time, however, they argued that restrictions on expression were compatible with another self-avowed mission: providing a safe and hospitable environment for the institution's community members. Whereas the Free Speech Movement asserted expressive freedom as a non-interference principle that befitted less institutional control, contemporary student protestors demanded precisely the opposite: an administration that would more readily interfere with expressive freedom if warranted.

Similar events across North America likewise suggest that changes in approaches to expressive freedom on campus are afoot. The most widely reported events typically revolve around a controversial speaker being deplatformed as a result of venues being pressured to cancel and/or raucous and disruptive protests (Heinze, Reference Heinze2019a; Smith, Reference Smith2020). At UC Berkeley, other conservative pundits stirred controversy after the Yiannopolous riot, including Ann Coulter and Ben Shapiro (Panzar and Tchekmedyian, Reference Panzar and Tchekmedyian2017; Peters and Fuller, Reference Fuller2017). At Middlebury College, Charles Murray was violently deplatformed (Saul, Reference Saul2017). At Yale University and the Evergreen State College, students berated their administrators for allegedly failing to create a more hospitable environment for minorities, and the terse confrontations went viral (Friedersdorf, Reference Friedersdorf2015; Hartcollis, Reference Hartocollis2017).

In Canada, similar examples include a fire alarm effectively cancelling an event featuring Faith Goldy at Wilfrid Laurier University and an appearance of Ezra Levant at Ryerson University being disrupted after it was relocated due to security concerns (Booth, Reference Booth2018; Malyk, Reference Malyk2017). Ironically, one of the most high-profile incidences of deplatforming involved an event titled “The Stifling of Free Speech on University Campuses.” In response to an inundation of complaints and an inability to provide adequate security, Ryerson University cancelled the event, and it was subsequently moved to a different venue (Hauen, Reference Hauen2017). The original lineup for the event featured Goldy (who was later excluded due to her extremism) and University of Toronto professor and clinical psychologist Jordan Peterson.

Nothing would solidify the perception that something was terribly amiss on Canadian campuses more than Peterson himself. In the fall of 2016, when he declared his refusal to use gender-neutral pronouns, as allegedly compelled by new legislation (via Bill C-16), he quickly became the cause célèbre of campus critics everywhere (Peterson, Reference Peterson2016; Bartlett, Reference Bartlett2018). Later, Wilfrid Laurier teaching assistant Lindsay Shepherd would magnify Peterson (and the alleged campus crisis) by showing a clip of him on TVO's The Agenda to her introductory communications seminar and then being critically interrogated by superiors in her department (Hutchins, Reference Hutchins2017).

To be clear, these events are neither indicative of a crisis (Sachs, Reference Sachs2018) nor novel when put in historical context (Smith, Reference Smith2020). But they do vividly demonstrate that approaches to expressive freedom may be changing in the contemporary moment. Rather than speculatively extrapolating from these events, it would be more productive for scholars to analyze the ideas and arguments that can be marshalled to justify expressive limits on campus. Before analyzing some of these novel conceptualizations of harm, I will discuss two explanatory frameworks catalyzed by recent events.

Two Theses of the Campus Crisis

As noted above, two explanatory frameworks for understanding the alleged campus crisis predominate, the snowflake thesis and the backlash thesis, but both have some inherent limitations that political theory can address. The snowflake thesis has gained impressive momentum since 2015 (Lukianoff and Haidt, Reference Lukianoff and Haidt2015). It has the advantages of piggybacking on long-standing anxieties about youth culture, decades of punditry alleging that higher education has been captured by progressives, and a flurry of controversies demonstrating the sometimes absurd nature of campus unrest. Although there is an explicit recognition that scholarship is consequential for evolving approaches to the question of expressive freedom on campus, if and when Lukianoff and Haidt engage with theory, it is strikingly shallow.

Although the authors do offer brief criticisms of concept creep (that is, words being understood as literal violence), Kimberlé Crenshaw's version of intersectionality in theory and practice, and Herbert Marcuse's concept of repressive tolerance (Lukianoff and Haidt, Reference Lukianoff and Haidt2018: 24–26, 64–71), they more often frame the problem in terms of individual psychosocial deficiencies of those who oppose potentially harmful expression on campus. As a result, Lukianoff and Haidt's analysis ends up pathologizing students with potentially legitimate grievances about expressive freedom; notably, the book includes an appendix on cognitive behavioural therapy, ostensibly to provide a redress for some of the cognitive distortions underpinning campus unrest.

The snowflake thesis predominates despite being based on scant empirical data (Dea, Reference Dea2018; Sachs, Reference Sachs2018), but arguments against it have sometimes relied upon flimsy argumentation and analysis too. The major countervailing argument, the backlash thesis, draws attention to the long-standing efforts to resist diversity on campus. In the wake of the 1960s, various social movements transformed into academic disciplines, commensurate with novel theories and methodologies grounded by collective identity and experience (Ferguson, Reference Ferguson2012). Student populations also became increasingly diverse, as universities loosened their exclusionary norms (Baer, Reference Baer2019; Ben-Porath, Reference Ben-Porath2017). These developments meant that individuals and groups once easily ignored are now an important part of academic life, even though access to academia remains uneven (Henry et al., Reference Henry, Dua, James, Kobayashi, Li, Ramos and Smith2017).

But these marginal gains for equity-seeking groups have not been without resistance. A host of critics have argued that scholarship grounded in identity-based particularism perverts the pursuit of truth and knowledge (Bawer, Reference Bawer2012; MacDonald, Reference MacDonald2018; Scruton, Reference Scruton2015). In this context, the “weaponization” of expressive freedom sits comfortably within a broader ongoing backlash against diversity on campus (Liptak, Reference Liptak2018; Malik, Reference Malik2019; Manne and Stanley Reference Manne and Stanley2015; Moskowitz, Reference Moskowitz2019; Picazo, Reference Picazo2017; Zine, Reference Zine2018). The term expressive freedom itself is sometimes considered a “red herring,” “Trojan horse” or “dog whistle” that facilitates the normalization of far-right politics and discourse on campus (Climenhaga, Reference Climenhaga2019; Levenson, Reference Levenson2017; Press Progress, 2019; Rangwala, Reference Rangwala2019).

While the backlash thesis has some explanatory potential, it is an oversimplification to frame all (or even most) invocations of expressive freedom on campus as a backlash to diversity. Likewise, the reflexive association between a strong defence of expressive freedom and classical liberal theory (or conservative and libertarian thought) is largely unearned (Heinze, Reference Heinze2016a, Reference Heinze2016b, Reference Heinze2017, Reference Heinze, Stern, Del Mar and Meyler2019b). In the same way that there are potentially sound reasons for challenging a generous margin of appreciation for expression on campus, there are potentially sound reasons for challenging expressive restrictions on campus, and from a wide range of perspectives. A corrective for these partial explanations, then, is to focus on the ideas at the root of the alleged campus crisis (Whittington, Reference Whittington2018: 57), something for which political theory is well positioned.

Conceptualizing Harm in Political Theory

What can account for these markedly different approaches to expressive freedom in academic contexts? The answer, I think, is that elastic conceptualizations of harm are increasingly legitimizing expressive restrictions, despite there being reasonable disagreements to have about definitively drawing these boundaries.

Because harm is expression's inseparable restraint in theory and practice, a specific conceptualization of harm is indispensable for understanding issues related to expressive freedom. For example, in Canada, section 2(b) of the Charter protects a wide scope of expression, but (if it is nonviolent) it is tempered by limits informed by the connection between expression and reasonably anticipated harms (Cameron, Reference Cameron1997, Reference Cameron2012; Moon, Reference Moon2000). In the most recent ruling on the constitutional validity of hate speech proscriptions, Saskatchewan (Human Rights Commission) v. Whatcott (2013), the Supreme Court unanimously decided that “a reasonable apprehension of societal harm” is the barometer for a legislative prerogative in this context (para. 135). Although some scholars disagree over the speculative connection between expression and harm in Canadian law (Braun, Reference Braun2004; Cameron, Reference Cameron2013; Heinze, Reference Heinze2016a; Newman, Reference Newman2017; Schutten and Haigh, Reference Schutten and Haigh2015; Sumner, Reference Sumner2004; Zwibel, Reference Zwibel2013), similar debates in political theory are perhaps even more intricate and contentious.

The reflexive conceptualization of expression and harm in liberal democracies (including Canada) is best encapsulated by the work of John Stuart Mill (Heinze, Reference Heinze2016a: 62; Moon, Reference Moon2000: 9–12). Mill's truth-based defence provides a wide margin of appreciation, warning that all forms of censorship are, in practice, an assumption of infallibility (Mill, Reference Mill, Philp and Rosen2015: 19). Although he also grants government the authority to wield coercive power over the individual to prevent harm done to others, harm is largely understood as material, demonstrable through direct causation and experienced at the individual level.

Since then, however, various strands of interdisciplinary theory have offered some robust and nuanced (re)conceptualizations of harm. For example, harm can also be psychological (Fanon Reference Fanon2008; Lawrence, Reference Lawrence1987), environmental (Dworkin, Reference Dworkin1981; MacKinnon, Reference MacKinnon1993; Waldron, Reference Waldron2012), epistemic (Medina, Reference Medina2013; Spivak, Reference Spivak, Nelson and Grossberg1988), experienced at the group level (Delgado and Stefancic, Reference Delgado and Stefancic2004, Reference Delgado and Stefancic2018; Matsuda et al., Reference Matsuda, Lawrence, Delgado and Crenshaw1993) and need not have a clear chain of causation (Gelber, Reference Gelber2002; Gelber and McNamara, Reference Gelber and McNamara2016; Langton, Reference Langton1993; McGowan, Reference McGowan2019; Tsesis, Reference Tsesis2002). In practice, these competing conceptualizations of harm contest the margin of appreciation that typically applies to liminal expression, sitting uneasily between the merely offensive and objectionable, on the one hand, and what is explicitly proscribed by law, on the other. Unsurprisingly, this type of liminal expression is the catalyst for almost every contemporary expressive freedom controversy on campus.

Many of the debates associated with the alleged campus crisis hinge less on whether individuals have a legal or institutional right to expressive freedom and more on whether their expression has caused sufficient harm to legitimize expressive restrictions. How one conceptualizes harm will, then, automatically condition the margin of appreciation for expression and can further condition expression itself (that is, self-censorship), even when the latter falls well within legal or institutional thresholds. Consequently, well-meaning but elastic conceptualizations of harm are laden with unintended consequences. While they might admirably sensitize one to various forms of harm that are well below the aforementioned thresholds, they might also unduly restrict expressive freedom and open inquiry, and especially so in an academic environment.

As Richard Moon (Reference Moon2019) rightly points out, if the theoretical assumptions that undergird justifications for expressive freedom are unwarranted—namely, individual rationality and autonomy, good faith epistemic engagement, and equitable access and distribution in the marketplace of ideas—then a re-examination of the principle is in order if it is to remain relevant in changing circumstances. In this sense, the university campus is an ideal litmus test for expressive freedom writ large (Heinze, Reference Heinze2016a: 177) and, further, consequential for expressive freedom for society more broadly (Whittington, Reference Whittington2018).

Despite its obvious imperfections, an academic environment still bears the closest resemblance to the ideal conditions in which expressive freedom can flourish. Even if one argues that these assumptions are unfounded, and thus bring the justifications into disrepute, if expressive freedom is impossible and/or undesirable in an academic environment, it is likely impossible and/or undesirable anywhere else. This raises the stakes of scholarship on the topic, so to speak, because if the diagnoses related to expressive freedom hold on campus, they will certainly hold elsewhere.

Epistemic Injustice

Miranda Fricker's Epistemic Injustice: Power and the Ethics of Knowing (Reference Fricker2007) is now a referential text, one that has helped spawn an expanding subfield of similar work (Dotson, Reference Dotson2011, Reference Dotson2014; Kidd et al., Reference Kidd, Medina and Pohlhaus2017; Medina, Reference Medina2013; Pohlhaus, Reference Pohlhaus2011) and ought to be a pillar of contemporary expressive freedom theory. Essentially, she argues that “testimonial exchange” features the potential for biased filtering. On the receiving end of an exchange, individuals “use social stereotypes as heuristics in their spontaneous assessments of their interlocutor's credibility” (16–17). If prejudice does, in fact, lead to an interpretive loss of credibility in the speaker, two things subsequently occur: (a) an “epistemic dysfunction” whereby a credibility deficit results in less knowledge created; and (b) the individual who suffers the deficit is harmed in their “capacity as a knower” (17). Credibility deficits are most troubling when they are persistent and systematic, in the sense that they are linked to broader social signs, meanings, interpretations, and so on, that have prejudicial effects based upon one's identity-related social situatedness (that is, disproportionately less power):

Primarily, the harm that accrues through epistemic injustice is an individual “wronged qua giver of knowledge” (Fricker, Reference Fricker2007: 45). Since knowing is “a capacity essential to human value,” it is a harm borne by those whose contributions are assigned less credibility for reasons extrinsic to the epistemic engagement (and it might additionally entail being “symbolically degraded qua human”) (44). At the least, prejudice uncharitably filters potential contributions to knowledge and at most outright obstructs them; therefore, it should be considered “a serious form of unfreedom in our collective speech situation” (43). One might also argue that secondary forms of harm similarly accrue at a broader level for anyone whose understanding is unnecessarily limited (and even for systems of knowledge). In this sense, “prejudice presents an obstacle to truth, either directly by causing the hearer to miss out on a particular truth, or indirectly by creating blockages in the circulation of critical ideas” (43). Finally, there are additional secondary harms associated with epistemic injustice that might include cascading perceptions of untrustworthiness and an epistemic inferiority complex (46–48).

However, not all disproportionate assignments of credibility are necessarily suspect; Fricker also outlines the parameters for “innocent error” (Reference Fricker2007: 21). This occurs when an individual assigns less credibility to a speaker than they ought to due to “a false belief about the speaker's level of expertise and/or motives” (21). Importantly, a “non-culpable mistake” such as this does not inflict harm because the root of epistemic harm is a specifically prejudicial judgment of the hearer (22). Instead, it might be “an unlucky epistemic mistake” (21) or “collective epistemic bad luck” (32). She uses the example of someone who has “shifty eyes” while presenting testimony and of a hearer who, in turn, uses a potentially reliable stereotype to infer untrustworthiness. Likewise, in a more fitting academic context, she highlights potential testimonial exchanges in which a credibility deficit does not automatically translate into harm (22). For example, because academics routinely assess each other's credibility, one might erroneously assume that another lacks credibility, expertise or authority. Fricker argues that this should be considered “a very weak sense of injustice,” as the misperception lacks “something ethically bad about the hearer's misjudgment” (22). By contrast, the main forms of epistemically and ethically culpable practices are related to what she terms “identity-prejudicial credibility deficits” (28). Thus, “an unreliable empirical generalization” itself may or may not be a case of epistemic injustice (32).

Crucially, Fricker suggests that testimonial injustice can result in epistemic exchanges that are not strictly testimonial and that the concept should be interpreted to include “all cases of telling” (Reference Fricker2007: 60). These instances might include, for example, “when a speaker simply expresses a personal opinion to a hearer, or airs a value judgement, or tries out a new idea or hypothesis on a given audience” (60). Within an academic context, however, there is a crucial distinction to be made between testimony—understood as the recounting of personal evidence and/or experience—and argumentation. Despite the former being easily discounted due to the credibility deficits Fricker analyzes, it is relatively less open to disagreement. When recounting a personal experience (that is, “X happened to me” or “this is what it is like to be X in Y situation”), there is less room for interpretation because it cannot be falsified in the same way that an argument might be. Part of experience is obviously subjective, but a great part of testimony is whether the event or experience actually transpired in the way recounted, which is why the assignment of credibility is so crucial. A slight exception might be if one imputes motivation or causation related to their experience. Or someone might be persuaded that their interpretation or account of events is distorted and/or unnecessarily unsympathetic—in a way similar to gaslighting—which could potentially be a form of epistemic (hermeneutical) injustice (McKinnon, Reference McKinnon, James, Kidd and Pohlhaus2017).

Argumentation, on the other hand, is much more open-ended, in the sense that it invites contestation by design, and particularly in an academic context. Because argumentation is always susceptible to potential disagreement, it is practically the prerequisite for scholarship. For example, the particular right of expressive freedom for academics—academic freedom—sharpens expressive freedom with institutional protections. Similar to expressive freedom (in a general sense), it is a counter-majoritarian protection (Fish, Reference Fish2014; Horn, Reference Horn1999a). According to James L. Turk, academic freedom “is a special right of academics—a right to freedom from prescribed orthodoxy in their teaching, research, and lives as academics” (Reference Turk2014: 11). Therefore, provided that neither institutional policy nor applicable law has been violated, academic freedom ought to insulate scholars from undue interference and subject their work only to the judgment of their peers. Without a distinction between testimony and argumentation, elastic conceptualizations of harm might legitimize expressive injunctions well within the boundaries of expressive freedom (and academic freedom) and similarly delegitimize interrogative practices that are necessary to assess scholarly merit.

Although non-academics do not enjoy these special protections—and thus a distinction between expressive freedom and academic freedom is merited (Dea, Reference Dea2018; Fish, Reference Fish2019; Scott, Reference Scott2019)—expressive freedom is still a pillar of campus life that animates the unique role of the university in society, as even a cursory review of institutional mission statements will quickly reveal. Further, one might argue that expressive freedom derives its legitimacy less from external legal frameworks than from the fact that “it is constitutive of the institution” (Whittington, Reference Whittington2018: 29). This is at least one reason why student agitation, including the Free Speech Movement, was able to gradually expand the scope of expressive freedom for students on campus (Horn, Reference Horn1999b).

If this distinction between testimony and argumentation holds, I argue that it would be unreasonable to adopt the same standards for epistemic injustice in both domains. Within the domain of argumentation, in particular, there is a responsibility to carve out a space for reasonable disagreement. The key difference is that while an individual's identity and/or experience should not be “up for debate,” their arguments (that is, analysis, recommendations, proposals, suggestions, hypotheses, etc.) absolutely are, and it needs to be the case that an individual's argumentation can be challenged without necessarily invalidating their identity and/or experience. Similarly, while one easily strays into unsavoury territory by evaluating the merit of an identity and/or experience, universities are an enterprise dedicated to the evaluation of argumentative merit (Fish, Reference Fish2019; Heinze, Reference Heinze2016a: 176–77; Scott, Reference Scott2019). Therefore, credibility deficits related to one's argumentation are not exactly analogous to credibility deficits related to one's testimony.

Aiding this distinction is another made by Sigal Ben-Porath in her recent book Free Speech on Campus (Reference Ben-Porath2017). In assessing the alleged campus crisis, she advocates for a theoretical position parallel to those who argue for hate speech restrictions in liberal democracies (see Waldron, Reference Waldron2012). Essentially, the dichotomous framing of protection from harm and expressive freedom belies the mutually constitutive nature of the two concepts (Ben-Porath, Reference Ben-Porath2017: 42–43). In order for epistemic exchange to actually mirror the ideal of expressive freedom, individuals must enjoy democratic equality. Thus, well-defined expressive restrictions may be consonant with expressive freedom, rather than an aberration. Nonetheless, she is also attendant to the ways expressive freedom and open inquiry can be limited by elastic conceptualizations of harm.

In this spirit, she argues that one ought to distinguish between two forms of safety on campus: dignitary and intellectual. Dignitary safety is the ability for professors and students to be accorded both equality and dignity in their academic pursuits. This type of safety is arguably the prerequisite for a properly functioning academic environment. Hopefully, if this is satisfied, members of the community are equally “invited to contribute to a discussion as a valued participant” (Ben-Porath, Reference Ben-Porath2017: 62). Intellectual safety, on the other hand, is to have one's concepts, ideas, worldviews and opinions remain intact, without needing to judge their merit through epistemic contestation. It includes a “refusal to listen to challenges to one's views or to consider opposing viewpoints” and is ultimately antithetical to the mission of higher education (62).

Ben-Porath's modest remedy is a framework of “inclusive freedom” that balances the putative tension between protection from harm and valuing expressive freedom in an academic context. Importantly, she also claims that not every invocation of harm is necessarily legitimate, in the sense that it is automatically sufficient to justify expressive limits on campus. In other words, while one may genuinely experience harm, the invocation of harm should not be uncritically received as an expressive injunction. In fact, there may be many academic contexts in which some discomfort could be considered good pedagogy, and educational environments ought to cultivate spaces of reasonable disagreement. Likewise, an identity reliably commensurate with disproportionately less identity-based power does not automatically translate into argumentative potency or legitimacy (Ben-Porath, Reference Ben-Porath2017: 68). Instead, “dignitary safety . . . should be understood as an aspect of access” (68). Put more simply, identity itself cannot insulate an individual from legitimate criticism, provided it does not cause dignitary harm, and dignitary harm (as outlined by Ben-Porath) is a more useful framework for drawing lines about what types of expression are (il)legitimate on campus. While not prescriptive, one potential way to distinguish dignitary harm is to take into consideration a distinction between testimony and argumentation. Credibility deficits arising from testimony would constitute dignitary harm. Credibility deficits arising from argumentation, however, would not constitute dignitary harm; they would merely constitute a challenge to intellectual safety.

But if identity and argumentation are interwoven, as many scholars might argue, is it epistemically unjust if a person regarded someone working within a specific framework as intellectually untrustworthy? How might one draw the line between an identity-based prejudice and an argument-based prejudice? The problem, then, is that neatly delineating testimony and argumentation is no simple task. The example mentioned earlier can be slightly altered to clarify. Considering that so much of scholarship is divided into various theoretical and methodological camps, it is likely that credibility is spontaneously assigned prejudicially in academic exchanges as a rule and not as an exception. As Fricker (Reference Fricker2007) notes, “A hearer could seemingly perpetrate a testimonial injustice without harbouring any prejudice at all” (41).

However, for Fricker, the determinative factor in assessing the moral culpability associated with potential credibility deficits is prejudice (and especially prejudice related to disproportionate identity-based power). Assigning a default credibility deficit to an interlocutor within (academic) testimonial exchange would be non-culpable if it is based upon one's well-considered beliefs about an inferior theory, framework, methodology, approach, and so on. However, this becomes more complicated when these are intimately interwoven with personal experience and/or group identity. For example, is it possible to know if prejudice is the genuine catalyst for assigning a credibility deficit? If someone—to take another example—finds feminist theorists or critical race theorists genuinely unpersuasive because of their scholarly argumentation (rather than the normative ends attached to such scholarship), can they do so without moral culpability? What would be the difference between finding a specific theory and/or methodology suspect because it lacks sufficient merit, in that person's view, and suspect because it is associated with a stereotypical (negative) conjuring of an identity group?

Although it is impossible to conclusively separate scholarly merit and the conjuring of negative stereotypes (and the former can surely masquerade as the latter), a distinction between testimony and argumentation can substantially address the problem. The ethical imperative or virtue associated with Fricker's (Reference Fricker2007) framework is to maintain a position of “critical openness” (66) that subsumes the potential for credibility deficits via (un)conscious prejudice (92–93). Thus, in order for good communication to transpire, “the hearer must exercise a certain reflexive critical awareness” (92). In its most basic form, individuals should overcome the default skepticism they might normally exercise when engaging in epistemic exchanges. This is a reasonable expectation, and Fricker is correct to say that denying some the ability to contribute to the creation of knowledge (and the necessary trust and good will to foster epistemic exchange) is a serious injustice (58–59). However, it is not self-evident how this theoretical approach might coalesce with an academic environment in which expression typically enjoys a wide margin of (moral) appreciation. For example, would a broad position of default skepticism in the academy be problematic? I think not, and for this reason, the same ethical position cannot extend from the realm of testimony into argumentation. However, there is a crucial caveat to be made: a distinction between testimony and argumentation can solely apply in circumstances in which a credibility deficit can be attributed to an anticipated argumentative position and not to an identity category.

For good reasons, an expectation of falsifiability would be unreasonable in the domain of testimony. It would inevitably lead to the harm of invalidating someone's identity and/or experience, such as attempts to convince a person that their perceptions are mistaken. One can legitimately question experiential interpretation, but an adversarial method, for example, would not necessarily refine testimony to be more “truthful.” Due to the particularities of an academic environment, and especially the reflexive nature of the adversarial method of knowledge production, the suspension of default skepticism is as difficult as it is undesirable. Even amid the lack of a “simple universal characterization of good scientific reasoning” (Moulton, Reference Moulton, Sandra and Hintikka2003: 152), there is still good reason to prioritize argumentation, because it is a form of “quality control” (ensuring scholarship is properly vetted by relevant experts) and the foundation of institutional legitimacy (how the scholarly enterprise justifies itself).

As Fricker (Reference Fricker2007) mentions, there are epistemic exchanges in which it would be completely reasonable to assign an interlocutor a default credibility deficit (35). For example, if someone were known to contribute unreliable information, it would be reasonable to assign that person less epistemic credibility. Likewise, if someone were known to contribute reliable information, it would be reasonable to assign that person more epistemic credibility. While extrinsic factors might influence one's spontaneous assignment of credibility (such as appearance and/or behaviour, etc.), it is not unreasonable to assume that the reliability of a person's (argumentative) epistemic contribution is the primary catalyst for such a determination. In such cases, a credibility deficit would not constitute harm; it is instead “epistemic bad luck,” even if the subject of the credibility deficit has disproportionately less identity-based power.

This raises at least two potential problems. First, there is the obvious difficulty of how to understand credibility deficits that arise when there is significant overlap between one's identity (and commensurate identity-based power) and one's argumentation. Importantly, an academic environment that thrives on reasonable disagreement does not necessarily require the suspension of default credibility deficits in order to foster good epistemic practices. The barometer should not be whether individuals suffer credibility deficits but, rather, whether interlocutors are ultimately impervious to superior reasoning and argumentation. In other words, the fact that an interlocutor might assign a speaker a default credibility deficit based on the speaker's perceived argumentative position does not mean that the interlocutor is impervious to that speaker's reasoning or argumentation. While it would be reasonable to say that such a speaker might face a disproportionate burden in imparting their position because of a credibility deficit, the deficit itself would not be unjust, provided it is not catalyzed by their identity. In sum, while it would be harmful to assign less credibility to an interlocutor due primarily to their identity (especially if they hold disproportionately less identity-based power), it would not be harmful to assign less credibility to an interlocutor due primarily to their argumentation.

Second, could it be reliably predicted whether a credibility deficit is catalyzed by an interlocutor's argumentation and not their identity? I think yes, because an academic environment is nominally premised upon (argumentative) merit. While a hearer may assign a credibility deficit during epistemic exchanges, that hearer ultimately bears the onus of presenting reasons why a specific argumentative position deserves less credibility. Thus, the test case for predicting the catalyst for (un)just credibility deficits is whether superior argumentation and reasoning is able to shift one's view (Fricker, Reference Fricker2007: 34). If an inferior argument persists despite compelling epistemic contributions, one may reasonably conclude that something extrinsic to the merits of the contribution (such as prejudice and/or stereotypes) is part of the interpretive calculus.

Argumentational Injustice

Stephanie Kapusta (Reference Kapusta2018) builds upon Fricker's concept of epistemic injustice, specifically examining the nexus of identity and argumentation. In 2017, the feminist philosophy field was polarized by a peer-reviewed journal article in Hypatia (written by Rebecca Tuvel) that analogized transgender and transracial identities, framed in part by the radically different reactions to Caitlyn Jenner and Rachel Dolezal's self-identifications. According to Tuvel's critics, her analysis was not just lacking in rigour but harmful (Schuessler, Reference Schuessler2017; Tuvel, Reference Tuvel2017; Winnubst, Reference Winnubst2017). Kapusta responds directly to this controversy by focussing on the ways in which marginalized identities might experience disproportionate burdens within epistemic exchange (“argumentational injustice”).

She argues persuasively that philosophical engagement ought to be considered a form of “argumentational work” (Kapusta, Reference Kapusta2018: 65), with naturally associated benefits that make it a desirable pursuit. These are typically “non-monetary ‘goods’” that include “excellence in performing relevant tasks, social status, community and collaboration, and a sense of self-respect” (65). However, there are differential costs and benefits associated with epistemic engagement that can be traced to identity. In her words: “There is an unwarranted selective exposure of some of the arguers to the risk of these harms, and this exposure occurs due to the fact that they are members of a marginalized group, that is, socially situated within systems of subordination” (66). Particularly in academic disciplines where an individual researcher or writer's own identity is relevant as it relates to disadvantage, marginalization and/or exclusion, it is more difficult to construe intellectual engagement as a disinterested or detached endeavour:

The potential harm, then, is the experience of disproportionate burdens that flow naturally from certain forms of argumentation; not only is someone marginally situated less likely to derive expected goods from epistemic engagement but they are also more likely to experience psychological harm (Kapusta, Reference Kapusta2018: 64). It is not primarily the assignment of credibility within epistemic exchange that is privileged per se (although this is certainly relevant) but rather the reasonably expected identity-related costs and benefits that might be sufficient conditions for “argumentational injustice.” As a result, there are two types of injustice that potentially flow from argumentational work: (a) some will face relatively higher “burdens of argumentational engagement” commensurate with their social situatedness, and (b) arguers might experience psychological harm (“harm injustice”) as a result of epistemic engagement (62). Avoiding such pitfalls requires, according to Kapusta, engagement with academic literatures that engage diverse identities and experiences, in addition to being attentive to the concerns of marginalized individuals so as to not perpetuate disproportionate burdens (70).

Kapusta entertains a number of potential objections to her position (Reference Kapusta2018: 68–69), but I think there are at least two additional (but no less relevant) difficulties that flow from her argument. First, it is entirely possible that an individual invokes harm (and experiences this harm as something intimately connected to their identity) to describe an epistemic engagement where there is simply reasonable disagreement. Considering that harm is innately subjective, it is always possible that what is experienced as harm can be amplified or dampened based upon an individual's personal circumstances and experiences. That said, while one ought not simply discount experiences of harm (as this would constitute dignitary harm), invocations of harm are not an automatic indication that a specific epistemic exchange is illegitimate and/or that expressive freedom ought to be restricted. Further, since the invocation of harm might lead to some form of punitive consequences, precision in the description of harm is paramount. Relevant questions might include who has been harmed, whether the epistemic exchange is linked to broader patterns of marginalization, and the type and level of harm (potentially ranging from mild conversational discomfort to palpable psychological trauma).

Second, it seems untenable that certain arguments would be relatively more immune from scholarly criticism simply because specific arguers had borne disproportionate burdens of argumentation. A reasonable ethical imperative is taking these burdens into consideration within epistemic engagement, but drawing a more definitive line between violations of dignitary safety and intellectual safety is warranted. After all, one might unfortunately face disproportionate burdens of argumentation but still present arguments that are worthy of critique. In consideration of these two difficulties, I again think there is a sharp distinction to be made between testimony and argumentation. In this instance, it is more specifically the difference between one's argumentation being invalidated and one's identity being invalidated; the former does not automatically constitute the latter, even when there is a strong correlation between the two. Although Kapusta's work is framed as a good epistemic practice and not a prescriptive injunction, it still runs the risk of positing an elastic conceptualization of harm by conflating disproportionate argumentative burdens and dignitary harm. While the latter might render expressive restrictions appropriate, evidence of the former alone would not be a sufficient condition for limiting expressive freedom in an academic context.

Epistemic Exploitation

Nora Berenstain (Reference Berenstain2016) similarly builds upon Fricker's concept of epistemic injustice with a specific focus on the position of default skepticism. According to her, “epistemic exploitation” occurs when marginalized individuals have an explanatory obligation placed upon them if and when they invoke harm (that is, testimony of disadvantage, marginalization and/or exclusion). When invocations of harm are greeted with skepticism rather than acknowledged as “contributions to knowledge,” a privileging occurs whereby some have access to the epistemic labour of others without any innate value attached to it (586). In this sense, identity again conditions epistemic exchange because harm may be experienced doubly: first as a result of one's identity and again in needing to overcome default skepticism to gain credibility and legitimacy. In her words:

Accordingly, warranting explanation is a typical method of diversion or distraction, conveniently eliding any responsibility or implication in the conditions that might generate such harm. The onus of surmounting this burden of proof often relegates complaint to the less valued epistemic class of conjecture, personal affect and partial observation. To compound this uneven demand of epistemic labour, the potential consequences of non-performance further constitutes a double bind. If a marginalized individual chooses not to substantiate their claims, so to speak, they may cement their image as opportunistic, angry, aggressive or as acting in bad faith. Nonetheless, if they choose to perform this ostensibly “exploitative demand,” there is by no means a guarantee that the result will be positive for either interlocutor (Berenstain, Reference Berenstain2016: 576). On this point, Berenstain and Kapusta overlap seamlessly. The result of these epistemic exchanges may, in fact, be a disheartening confirmation of ignorance and wasted emotional, psychological and physical energy. Most importantly, Berenstain argues that those who defend interrogative practices—being a “devil's advocate” or “offering alternative explanations,” for example—are masquerading epistemic exploitation as a “virtuous epistemic practice related to the pursuit of truth” (571).

Clearly, Berenstain's point is merited; good epistemic practices would surely include being cognizant of disproportionate identity-based burdens. But is harm self-evident when an individual invokes it? Further, is every invocation of harm legitimate in the sense that default skepticism is rarely, if ever, warranted? In practice, particularly for an academic environment, this conceptualization of harm poses some difficulties. Principally, it posits harm as something entirely subjective. Of course, in an academic environment, the object of moral concern ought to be potentially harmful expression that jeopardizes full participation in the community. Thus, this should not be read as an invitation to automatically question the identities and experiences of individuals if and when they present painful testimony. In this case, a “virtuous epistemic practice” would be empathy and humility and decidedly not default skepticism. But due to the institutional mandate to foster spaces of open inquiry on campus, drawing a neat line between mere discomfort (as a potentially unavoidable externality) and harm (which might merit an institutional response) is no easy task. Unfortunately, there will be occasions in which otherwise reasonable disagreement is subjectively experienced as harm. In these situations, it would be presumptuous to assume that specific identity categories (even those that reliably correlate with relatively less identity-based power) automatically translate into credibility and/or argumentative superiority (or that skepticism itself constitutes an oppressive act).

The obvious retort is that not all potential harms are equal. As all three of the interrelated concepts—epistemic injustice, argumentational injustice and epistemic exploitation—make clear, identity conditions one's epistemic exchanges and may disproportionately burden some over others. Accordingly, the harm experienced by those with relatively less identity-based power takes normative precedence over those with relatively more. If one truly cares about equity within educational contexts, this principle or position is unassailable. Its negation would seriously jeopardize the prerequisite equality for open and equal learning opportunities. However, again, a distinction between testimony and argumentation is useful in parsing this complexity in order to do justice to both expressive freedom and equitable learning environments.

While default skepticism is typically inappropriate in the context of presenting testimony related to one's identity and experience, interrogative practices (like those earlier mentioned) are not necessarily inappropriate in the context of argumentation (even if related to potentially painful experiences). In this sense, it ought to be reasonable to assume that interrogative practices can engage with argumentation (that is, reasonable disagreement) without impinging upon the moral and intellectual worth of epistemic interlocutors. Since judging argumentative merit (by some standard) is required of all scholarly pursuits, it remains unclear how critical but legitimate questions could be raised without doing some harm, by Berenstain's standard.

The primary target of Berenstain's position is default skepticism, a pillar of political theory but no less vital to a range of other disciplines. But here, again, a distinction between testimony and argumentation is useful to address the elasticity of harm within Berenstain's framework. When marginalized individuals provide testimony related to the oppression they face, an ethical epistemic practice would be the “reflexive critical awareness” suggested by Fricker (Reference Fricker2007: 92). But while one should hopefully extend the same critical awareness to marginalized individuals when they make arguments, it is not necessarily an oppressive practice to interrogate specifically argumentative claims (even if based to a large degree on identity and/or experience). There is no doubt that an onus placed on marginalized individuals to substantiate testimony would be unjust if it were disproportionately burdensome and/or catalyzed by identity-based credibility deficits. In an academic environment, however, identity and/or experience cannot insulate someone from scrutiny and criticism if and when they are making specifically argumentative claims, nor should invocations of harm be an automatic expressive injunction.