1.1. Open-mindedness

Let’s begin with our proposed account of open-mindedness. What does it consist of? Following Baehr (Reference Baehr2011a, Reference Baehr2011b), Riggs (Reference Riggs and Battaly2019), and the standard view in virtue epistemology, we assume open-mindedness is a disposition or trait, though we don’t assume it is a virtue (more on this below). We likewise assume that (at a minimum) open-minded people are not closed-minded in Battaly’s (Reference Battaly2018a) sense. Open-minded people do not tend to dismiss relevant ideas and evidence that conflict with their own beliefs. In contrast, when competing ideas and evidence cross their paths, they tend to engage seriously with them (Baehr Reference Baehr2011b: 151). They tend to “hear them out,” make an effort to understand them and evaluate them on their merits. They neither dismiss nor ignore the competing ideas and evidence that they encounter. Accordingly, we can begin with the following suggestion:

What is wrong with this initial suggestion? For starters, it isn’t sufficient for open-mindedness, since it allows those who hide from encounters with competing ideas to be open-minded. As Medina has forcefully argued, people who “need not to know”—i.e., who are willfully ignorant of competing ideas and deliberately wall themselves off—are closed-minded rather than open-minded (Medina Reference Medina2013: 34). Clearly, this applies to people who intentionally hide in like-minded echo chambers (Nguyen Reference Nguyen2020). But, in our view, people who unwittingly get stuck in epistemic bubbles suffer a similar fate due to no fault of their own. One of the insights of feminist virtue theory (Daukas Reference Daukas and Battaly2019; Dillon Reference Dillon, Crasnow and Superson2012) is that our environments impact our traits, and can do so despite our intentions; our environments can prevent us from having some traits and set us up to have others, e.g., they can make us epistemically unjust even while we are desperately trying not to be (Fricker Reference Fricker2007: 37). In short, the point is that to be open-minded, it won’t be enough to seriously engage with relevant competing ideas when one encounters them. To be open-minded, one must encounter them. More specifically, one must be in an environment that affords opportunities to encounter them in the first place. And, once in such an environment, one must encounter them and engage seriously with them. Accordingly, people who fail to encounter competing ideas, and thereby fail to engage seriously with competing ideas, won’t be open-minded, at least in their current environment. For this reason, we think people who hide in echo chambers or are stuck in epistemic bubbles aren’t open-minded, the former because they intentionally eschew opportunities to encounter competing ideas, and the latter insofar as they are stuck in environments that don’t afford them such opportunities—they unwittingly fail to encounter competing ideas.

Granted, there is more than one way to encounter relevant competing ideas and evidence: by crossing paths with those generated by other agents, and also by thinking them up ourselves. But, in our view, people who only encounter and engage seriously with relevant competitors they generate themselves aren’t sufficiently open-minded either. Of course, they will be more open-minded than people who fail to encounter and engage seriously with any relevant competitors at all. But in environments that afford ample opportunities to encounter and engage seriously with the relevant competing ideas of others, failures to do so will be failures of open-mindedness.Footnote ³ Case in point: the White supremacist who intentionally walls himself off from our arguments against his beliefs is (in doing so) failing to be open-minded, even if he manages to generate and engage seriously with some relevant competing ideas on his own. Even if he is disposed to engage seriously with a sub-set of relevant competitors—those he generates himself—he isn’t disposed to engage seriously with relevant competitors more broadly, and it is the latter disposition that is partly constitutive of open-mindedness, and that figures in (OM1).

To frame this differently, the above emphasizes interacting with the relevant competitors of other agents for good reason. Returning to the insights of feminist epistemology, none of us can generate all of the relevant competitors to our ideas ourselves (much less generate all the relevant competing evidence).Footnote ⁴ Happily, we don’t need to, since once we get outside our own epistemic bubbles and echo chambers, our shared epistemic environment in the real world affords us ample opportunities to encounter and engage with the relevant competitors of others. If, as argued above, our failures to do so would be failures of open-mindedness, then there is an important sense in which open-mindedness is (1) partly other-oriented and partly constituted by other-oriented engagement. We think this marks a key difference between open-mindedness and intellectual humility, and return to it below. For now, suppose we eliminate the offending conditional in (OM1) and make its other-orientation explicit, bringing us to:

But, this still isn’t sufficient for open-mindedness, since one can engage seriously with a broad range of relevant competitors, including those supplied by other agents while remaining closed-minded. To engage seriously with competing ideas and evidence, one must evaluate them on their merits by, e.g., asking relevant questions, evaluating whether arguments for them are valid or strong, offering relevant rebuttals and counter-evidence, and so forth. But, one can do all of these things while remaining closed-minded. Borrowing an example from Battaly (Reference Battaly2018a: 280), imagine a conspiracy theorist who listens to you, correctly represents your view, and offers relevant counter-evidence, but only because she wants to change your mind and not because she is willing to revise her belief. This conspiracy theorist isn’t interested in re-opening her own inquiry: she isn’t being open-minded; she is only engaging seriously with your ideas because she wants to change them. This shows us that open-mindedness is also partly constituted by (2) a willingness to revise one’s beliefs, and more specifically, by a willingness to revise one’s beliefs due to one’s tendency to engage seriously with relevant competitors (rather than due to some extraneous factor). We think this marks a second key difference between open-mindedness and intellectual humility. Accordingly, suppose we adopt:

This still isn’t quite right. The problem with (OM3) is that it restricts the scope of open-mindedness to cases in which we already have beliefs, and further restricts its locus to ideas and evidence that compete with those beliefs. But, the first thing to note is that we needn’t already have beliefs about a topic to be open- or closed-minded in the way we approach it. Suppose, for instance, that we are inquiring into a topic for the first time, and haven’t yet formed any beliefs about it. We can still be open- or closed-minded in the range of ideas and evidence we consider. Second, we can also be open- or closed-minded in the questions we ask about it, the methods we use to explore it, the sources and traditions we consult, and so forth. Accordingly, the locus of open- and closed-mindedness isn’t exhausted by ideas and evidence and should be expanded to include what Battaly (Reference Battaly2018a) terms “intellectual options,” e.g., questions, methods, and sources. In short: in addition to excluding agents who are closed-minded about ideas and evidence that compete with their extant beliefs, our analysis needs to exclude agents who are closed-minded with respect to these other intellectual options, both in cases where they have extant beliefs and in cases where they don’t. With that in mind, we propose the following as a working analysis of open-mindedness:

Note that in cases where we already have extant beliefs about a topic, including cases in which we reject extremist views, we won’t be open-minded unless: (1) we engage seriously with others at least insofar as they are sources of relevant competitors to our beliefs; and (2) we are willing to revise our beliefs, and more specifically, willing to revise them due to our tendency to engage seriously with relevant competitors.

1.2. Intellectual humility

Let’s turn to intellectual humility and what it consists of. Following Whitcomb et al. (Reference Whitcomb, Battaly, Baehr and Howard-Snyder2017), we assume that intellectual humility is a disposition or trait that needn’t be an intellectual virtue (see below). We further assume that (at a minimum) intellectually humble people are not arrogant, at least insofar as they don’t overlook or downplay their own cognitive limitations, e.g., their ignorance and gaps in knowledge, their cognitive mistakes, and their deficits in cognitive skills. Quite the contrary, intellectually humble people are attentive to and own their cognitive limitations (Whitcomb et al. Reference Whitcomb, Battaly, Baehr and Howard-Snyder2017).

Roughly, intellectually humble people are attentive to their cognitive limitations in the sense that their limitations come to mind; i.e., they aren’t oblivious to their limitations. But, attentiveness won’t be sufficient for intellectual humility, since agents whose cognitive limitations come to mind can still be in denial about them (i.e., refusing to believe they have them), complacent about them (i.e., refusing to care about them), or even hostile about them. Accordingly, intellectually humble agents must also own their cognitive limitations, where which characteristically involves: admitting them to themselves (rather than denying them), caring about them and taking them seriously (rather than being complacent), and feeling dismay or regret (rather than hostility) about them (Whitcomb et al. Reference Whitcomb, Battaly, Baehr and Howard-Snyder2017: 517–519). In short, we follow the limitations-owning analysis of Whitcomb et al. (Reference Whitcomb, Battaly, Baehr and Howard-Snyder2017), whereby:

Notably, owning one’s limitations doesn’t require success in changing them or having control over them. But, it does require caring about them and taking them seriously by trying to change (or mitigate) them. Accordingly, we can own limitations (e.g., implicit biases) that can only be changed gradually, and that (e.g., color-blindness) can’t be changed at all. In our view, agents own limitations like these by admitting, caring about, and trying to change or mitigate them (e.g., via recommended strategies), even if they haven’t succeeded in changing or mitigating them.

There are two crucial points to note about intellectual humility (IH*) as we understand it. First, it consists of a self-oriented, intra-personal, or inwardly-directed disposition, rather than an other-oriented, inter-personal, or outwardly-directed disposition. To explain, intellectual humility is a disposition toward one’s own cognitive limitations. It is a stance one can take (or fail to take) toward cognitive features oneself (Whitcomb et al. Reference Whitcomb, Battaly, Baehr and Howard-Snyder2017; Tanesini Reference Tanesini2021). We think this is the right way to understand intellectual humility since we can’t be intellectually humble about someone else’s cognitive limitations.Footnote ⁵ That is a category mistake. For instance, I can’t be intellectually humble about President Biden’s errors in reasoning. I can be (e.g.) empathetic or understanding about Biden’s errors, or (e.g.) disappointed, frustrated, or annoyed by them. But, I can’t be intellectually humble about them because (I recognize that) they aren’t mine to be humble about. In short, intellectual humility is a stance we can take (or fail to take) toward our own limitations or what we view as extensions of them; it isn’t a stance we can take (or fail to take) toward the limitations of others.

Nor does intellectual humility, in and of itself, require engagement with others. As an inwardly-directed and self-oriented disposition, IH* does not require (1) a disposition to engage with others or with the intellectual options they supply. We think this, too, is the right way to understand intellectual humility, since there is nothing internally inconsistent or self-defeating about an isolated inquirer who is nevertheless intellectually humble. Adapting an example from Tanesini, a scientist who finds herself isolated and alone, with nobody to consult, can still be intellectually humble in conducting the lines of inquiry she carries out on her own. She can recognize that she “has made a mistake when an experiment produces results that are not credible,” and can own her flawed experimental design by changing it (Tanesini Reference Tanesini2016: 82).Footnote ⁶ In other words, intellectual humility is something an isolated agent can have when conducting inquiries by herself.Footnote ⁷ It can manifest in behaviors that are solely intrapersonal.

But, while intellectual humility doesn’t itself require engagement with others, in contexts where one is already engaging with others for independent reasons, it can be manifested in interpersonal behaviors. To explain, in situations where we are already engaging with others—perhaps because we are already in a conversation—intellectual humility can be manifested in interpersonal behaviors like admitting one’s limitations to others and deferring to others (in addition to intrapersonal behaviors like admitting them to oneself). In our view, though tendencies to perform these interpersonal behaviors aren’t required for having IH*, since we needn’t be in contexts of engagement to have it, they can be manifestations of IH* given that (and because) the context is already one of engagement. Here, the point is that while IH* can be manifested in interpersonal behaviors, it needn’t be, and doesn’t by itself entail interpersonal behaviors.

Intellectual humility can also be causally connected to other-oriented engagement. Engaging can be a “downstream” product of intellectual humility, e.g., when one seeks out others to gain knowledge as a result of intrapersonally admitting one’s own ignorance (Krumrei-Mancuso and Worthington Reference Krumrei-Mancuso, Worthington, Ottati and Stern2023). It can likewise be an “upstream” contributor to intellectual humility. We might even causally need to engage open-mindedly with others to become aware of our stealthy limitations, in which case we should expect open-minded engagement to be part of the standard causal story of the development of IH* (Baehr Reference Baehr2022). But, this doesn’t mean open-minded engagement is partly constitutive of intellectual humility; it just helps us develop intellectual humility.

One might object to our intrapersonal view of intellectual humility on the grounds that arrogance is interpersonal. The thought is that if “arrogance is essentially an interpersonal matter” (Tiberius and Walker Reference Tiberius and Walker1998: 381), then humility should be, too. Else, one risks the compatibility of humility and arrogance. Additionally, intrapersonal views like ours risk mistaking arrogance for a self-oriented disposition and precluding other-oriented manifestations of it—such as interrupting others and treating them with disdain. In reply, we do think arrogance consists of a self-oriented disposition to either be oblivious to one’s limitations or under-own those that come to mind (e.g., by refusing to believe one has them, failing to take them seriously, or getting angry about them). Accordingly, we think arrogance doesn’t itself require interpersonal behaviors like interrupting others. Our isolated scientist could just as easily be arrogant in conducting her inquiries. She could be oblivious to, or in denial about, her errors, or unwilling to take the flaws in her design seriously. She could “stand in perfect isolation, absolutely indifferent to the behaviors of others, and yet be arrogant” (Tanesini Reference Tanesini2016: 82). We note that interpersonal accounts of humility and arrogance struggle to include cases like these. But, while arrogance needn’t be interpersonal, we think it can be manifested in interpersonal behaviors, provided that one is already in a context of engagement. The person who is already engaging with others can manifest arrogance by, e.g., talking over them, dominating the conversation, expecting others to defer, or getting angry when they don’t. They can manifest arrogance by being haughty.Footnote ⁸ So, arrogance can, but needn’t, be manifested in interpersonal behaviors. In this way, our account can explain the arrogance of the isolated individual and the arrogance of the braggart who dominates our conversations. Further, this means that IH* and interpersonal arrogance are incompatible in contexts of engagement—one can’t simultaneously be IH* and interpersonally arrogant, given that one is already in a context of engagement.Footnote ⁹

Summing up our view thus far: unlike open-mindedness, intellectual humility as IH* does not by itself entail or consist in (1) other-oriented engagement. The second crucial point is that IH* doesn’t by itself require or consist in (2) a willingness to revise beliefs either, or so we’ll try to show below.Footnote ¹⁰ By way of preview, Section 1.3 addresses a number of important ways that we can respond to extremist views with intellectual humility. For instance, it reminds us that we can be IH* in responding to extremist views by owning our tendencies to jump to conclusions about extremists, and owning the limitations in our powers to persuade them. But it argues that since these are cases of being humble about extremists and about our own skills, rather than about extremist views themselves, it isn’t surprising that these humble responses can co-exist with refusals to revise our beliefs about extremist views. To show that IH* doesn’t entail a willingness to revise our beliefs about extremist views, we narrow the target further. We argue that we can be IH* about our beliefs about extremist views by owning the limitations in our knowledge that they are false while being unwilling to revise our beliefs that they are false.

1.3. Intellectual humility without open-mindedness to extremist views

There are several ways in which we can be intellectually humble in responding to extremist views without being open-minded to those views and, specifically, without being willing to revise our beliefs that those views are false. This section focuses on five such ways.

(I) Beginning with a suggestion from Whitcomb et al. (Reference Whitcomb, Battaly, Baehr, Howard-Snyder, Alfano, Lynch and Tanesini2021), we can be intellectually humble by owning our tendencies to jump to conclusions about advocates of extremist views, and we can do so without being open-minded to their views. Since even rank-and-file Democrats and Republicans tend to jump to unsupported conclusions about one another (Pew Research Center 2022), it would be surprising if we didn’t also draw unsupported conclusions about advocates of extremist views. For instance, following Whitcomb et al. (Reference Whitcomb, Battaly, Baehr, Howard-Snyder, Alfano, Lynch and Tanesini2021), suppose we are inclined to infer that White supremacists are irredeemable and inhuman. That tendency would be an intellectual limitation since the conclusion that all White supremacists are beyond salvage outstrips our evidence. (By comparison, the conclusions that White supremacy is false, that White supremacists are dangerous racists whose biases produce their false beliefs, and that some of them won’t change are supported by ample evidence.) Accordingly, one way to be intellectually humble is to own our tendency to jump to conclusions about extremists.

To illustrate this way of being intellectually humble, consider Daryl Davis, a Black musician who has convinced approximately two dozen Klan members to change their minds and leave the Ku Klux Klan. As documented in the film Accidental Courtesy (Ornstein Reference Ornstein2016), Davis engages with individual Klan members and their beliefs by using counter-evidence to argue against their views. He does this because he sees the Klan members he approaches as deeply biased but redeemable people who are capable of listening to arguments, learning from them, and changing their minds. (He doesn’t approach Klan members he thinks won’t change.) Now, it is possible that Davis was never inclined to jump to unsupported conclusions about Klan members to begin with. But, the more likely story is that he owned such inclinations—he responded with intellectual humility—and as a result, learned to prevent himself from regarding Klan members as irredeemable (Whitcomb et al. Reference Whitcomb, Battaly, Baehr, Howard-Snyder, Alfano, Lynch and Tanesini2021). Davis’ case is controversial, not because we question whether he is intellectually humble in his response to White supremacy, but because we question whether his intellectual humility is excessive (servile) rather than virtuous.

Whether excessive or virtuous, our present point is that his response is intellectually humble without being open-minded to extremist views. Davis owns the aforementioned limitation while remaining unwilling to revise his belief that White supremacy is false. As Davis attests in the film, he categorically rejects the “separatist ideology” of the Klan and their belief in White supremacy. Accordingly, he does not respond to their extremist views open-mindedly, since doing so would require both engaging with others who offer relevant competitors to his beliefs and being willing to revise his beliefs. Though Davis engages seriously with the arguments of Klan members, he is not willing to revise his belief that White supremacy is false. Davis isn’t trying to figure out what he should believe about White supremacy; instead, he’s trying to convince Klan members to change their minds.

(II) Taking a second cue from Whitcomb et al. (Reference Whitcomb, Battaly, Baehr, Howard-Snyder, Alfano, Lynch and Tanesini2021), agents can own the limitations in their powers to persuade extremists, without being open-minded about extremist views. What does it look like to own limitations in one’s powers of persuasion? To give some examples, an agent might recognize that he isn’t quick-enough-on-his-feet and isn’t skilled enough to rebut the arguments of extremists in real time or prevent the conversation from getting hijacked. One way to own such limitations is to resolve against directly engaging. Alternatively, an agent might admit to himself that he is bad at anticipating the objections extremists raise, and resolve to do more research before engaging.

Importantly, agents can do this without being open-minded to extremist views. To illustrate, consider Allison Gornik, who helped to convince Derek Black (“godson” of Klan spokesman David Duke) to change his mind about White supremacy (Saslow Reference Saslow2018). When Gornik and Black met in college, Black was a radio host for the Klan and the creator of Stormfront’s online content for children. Black’s subsequent renunciation of White supremacy and his departure from the Klan have been widely attributed to his debates with Gornik and other college friends (Saslow Reference Saslow2018). Gornik’s case is illustrative of owning limitations in one’s powers of persuasion without being open-minded to extremist views. Below, she reports not knowing how to argue against Black’s views in their initial discussions:

Gornik owned these limitations in her powers to persuade Black by admitting them and trying to resolve them. When she encountered arguments of Black’s that she couldn’t rebut, she looked for rebutting evidence, which she subsequently shared with Black. Throughout, Gornik remained stalwartly opposed to White supremacy—she wasn’t willing to revise her own beliefs on the matter; she engaged with Black in an effort to change his mind. Accordingly, her response wasn’t open-minded, since open-mindedness requires a willingness to revise one’s beliefs in addition to engagement.

Granted Gornik, like Davis, did engage; moreover, the ways that she owned her limitations presupposed that she would continue to engage. Rather than owning her limitations by resolving to step back from discussions with Black—viz., by resolving to disengage—she conducted extensive research on her own time in order to refute his arguments during their engagements. As we argue in Section 2, there are some conditions in which we should not engage directly with advocates of extremist views, even if there are others in which we should. Whether or not Gornik was right to own her limitations in the ways she did (which is an open question), our present point is that she owned them without being open-minded about White supremacy.

While these are both important ways of being intellectually humble in responding to extremist views, it shouldn’t be surprising that they can co-exist with refusals to revise our beliefs about extremist views. This is because they are cases of being humble about extremists and about our own skills, rather than about extremist views themselves. To explain, for all we know, people like Davis, who humbly prevent themselves from concluding that “advocates of extremist views are inhuman,” may be willing to revise their beliefs about extremists (Fantl Reference Fantl2018: 162). And people like Gornik, who humbly own the limitations in their powers of persuasion, may be willing to revise their beliefs about their own argumentative powers. To show that IH* doesn’t entail a willingness to revise our beliefs about extremist views, we must narrow our target to cases where we humbly own the limitations in our beliefs about extremist views while being unwilling to revise them. Below, we suggest three ways of doing this, all of which are ways of owning the limitations in our knowledge that extremist views are false.

(III) We have many arguments and masses of evidence against extremist views like White supremacy and election-denial. Our evidence is both broad and deep. Suppose we are assessing the comparative strength of our arguments against these views. In so doing, we can recognize and admit that some of our arguments aren’t quite as strong as others (Hannon and Kidd Reference Hannon and Kidd2022). And, we can care about these limitations and take them seriously by (e.g.) trying to avoid relying on these comparatively weaker arguments, and by trying to proportionately reduce confidence in our beliefs that White supremacy and election-denial are false. In doing these things, we humbly own the limitations in our evidence for our knowledge that White supremacy and election-denial are false. Now arguably, we can do these things without being willing to revise our beliefs that White supremacy and election-denial are false. First, let’s assume that we would need to reduce our confidence in these beliefs in order to be willing to revise them. Importantly, we don’t have direct control over reducing our confidence in these (or any other) beliefs. Our view of owning accommodates this: though owning requires caring about our limitations, taking them seriously, and trying to change or mitigate them, it doesn’t require success in changing or mitigating them. Accordingly, we can own these limitations by caring about and trying to proportionately reduce our confidence in our beliefs that White supremacy and election-denial are false, even if we don’t succeed in proportionately reducing our confidence in these beliefs—even if we don’t reduce our confidence at all. Consequently, we can own the limitations in our evidence for our beliefs that White supremacy and election-denial are false without being willing to revise these beliefs. Second, let’s further assume that in order to be willing to revise well-supported beliefs like these, we would need to significantly reduce our confidence in them—minor reductions in confidence wouldn’t be enough. Accordingly, even if owning did require success in changing or mitigating one’s limitations, and thus did require a proportionate reduction of confidence in one’s beliefs that White supremacy and election-denial are false, that proportionate reduction would be minor and insignificant. It wouldn’t be significant enough to trigger a willingness to revise these beliefs, since there would still be so very many good reasons for us to endorse them.

(IV) Relatedly, our knowledge that White supremacy is false and Trump didn’t win the 2020 election are the products of reliable belief-forming processes, including various forms of induction, deduction, perception, and testimony. In assessing these processes, we can recognize and admit that they are fallible—they don’t always produce true beliefs. And, we can care about and try to mitigate that limitation by trying to proportionately reduce our confidence in these target beliefs. In other words, we can humbly own the limitations in the processes that led to our knowledge that White supremacy and election-denial are false. And, mutatis mutandis, we can do this without being willing to revise our beliefs that White supremacy and election-denial are false. First, we can own these limitations in our processes without succeeding in proportionately reducing our confidence in the target beliefs. Second, even if owning required a proportionate reduction, it wouldn’t be significant enough to trigger a willingness to revise these beliefs since the processes that generated them are still reliable.

(V) Finally, we can presume that our degree of confidence in our beliefs that White supremacy and election-denial are false is quite high. Still, we can recognize and admit that we may hold these beliefs with less than 100% confidence (if we do). In Leary’s words, we can recognize that “a particular personal belief may be fallible” (Reference Leary, Diebels, Davisson, Jongman-Sereno, Isherwood, Raimi, Deffler and Hoyle2017: 340). We can likewise care about and try to mitigate this limitation; i.e., we can humbly own this limitation in our degree of confidence in our target beliefs. But, mutatis mutandis, we can do this without reducing confidence in these beliefs and without being willing to revise them.

We propose (III)–(V) as cases of intellectual humility without open-mindedness.Footnote ¹¹ These are cases in which we humbly own the limitations in our knowledge that various extremist views are false, without being willing to revise our beliefs that those extremist views are false. Granted, we haven’t fully analyzed what it means to be willing to revise beliefs. We note that epistemologists disagree about whether one would need to (significantly) lower one’s confidence in strongly held beliefs in order to be willing to revise them: e.g., Fantl (Reference Fantl2018) seems to endorse this requirement, whereas Pritchard (Reference Pritchard and Da Silva Filho2019, Reference Pritchard2021) denies it. While we have followed what we take to be Fantl’s view, we won’t be adjudicating this dispute here.Footnote ¹² Our claim is that if being willing to revise strongly held beliefs requires (significantly) lowering one’s confidence in them, then one can be intellectually humble about strongly held beliefs without being willing to revise them. We take (III)–(V) to demonstrate this because, in our view, owning doesn’t entail success in lowering confidence in one’s beliefs, and even if it did, our reliability, evidence for, and confidence in our beliefs that these extremist views are false are all so strong that we could own their minor shortcomings (e.g., their fallibility) without significantly lowering our confidence.Footnote ¹³

If we are correct that owning doesn’t entail success in mitigating limitations, then any minor adjustments we might make to our confidence wouldn’t be manifestations of humility, but causal consequences of it. Importantly, this also applies when agents are both intellectually humble and open-minded about beliefs they do not know, e.g., that X will win the next election. In our view, their willingness to revise such beliefs doesn’t manifest their humility, though it can be a causal consequence of it, e.g., when owning their limitations leads them to sufficiently reduce confidence in their beliefs, triggering a willingness to revise them. Notably, such humility might not lead agents to be willing to revise their beliefs, e.g., when one humbly acknowledges and cares about the weakness of one’s reasons for a belief, but has an “emotional, psychological, or some other “block” that prevents them from being open” to revising it (Spiegel Reference Spiegel2012: 35)—or, as we would put it, prevents them from sufficiently reducing confidence in their beliefs. This phenomenon indicates that humbly owning a belief by owning “limitations in the evidentiary basis of that belief” (Leary et al. Reference Leary, Diebels, Davisson, Jongman-Sereno, Isherwood, Raimi, Deffler and Hoyle2017: 340) is one thing, while being willing to revise it is another (see Krumrei-Mancuso and Worthington Reference Krumrei-Mancuso, Worthington, Ottati and Stern2023; and Whitcomb et al. Reference Whitcomb, Battaly, Baehr and Howard-Snyder2017). Agents can do the former without doing the latter.

Crucially, we have focused on extremist views that we know are false because they make the hunting easy. If we are going to find cases of IH* without OM* anywhere at all, we are going to find them in examples like (III)–(V), in which the beliefs in question are already known, and either amply supported by evidence, generated by reliable processes, or held with a high degree of confidence. To find additional cases of IH* without OM*, we advise starting with other amply-supported, reliably-produced beliefs of ours that we know, and then expanding the search to include hinge beliefs about which one might have a “block” that prevents one from being willing to revise them. If our arguments prove viable, then intellectual humility and open-mindedness can come apart, even if the cases in which they do come apart tend to be rarefied, and they usually travel together.

Throughout, we have tried to describe the traits of open-mindedness and intellectual humility in normatively neutral terms, on the assumption that they needn’t be intellectual virtues. Following the pluralist trend in virtue epistemology, we acknowledge that there is more than one way for a trait to be an intellectual virtue. Roughly, a trait will be a Reliabilist effects-virtue insofar as it tends to produce a preponderance of good epistemic effects, e.g., true beliefs (see, e.g., Sosa Reference Sosa2007). Importantly, Reliabilist effects-virtues aren’t exhausted by hard-wired faculties and acquired skills. Character traits like open-mindedness and intellectual humility will be Reliabilist effects-virtues if they are reliable—if they tend to produce a preponderance of true beliefs or other good epistemic effects (see, e.g., Baehr Reference Baehr2011b: 52–54, 60–62; Sosa Reference Sosa2015: 48). Whereas, for character traits like open-mindedness and intellectual humility to be Responsibilist character-virtues, they must be partly constituted by good epistemic motives (e.g., for truth), and good judgment (phronesis) in epistemic contexts (see, e.g., Baehr Reference Baehr2011b; Zagzebski Reference Zagzebski1996). While the traits of open-mindedness and intellectual humility are often virtues of both sorts, they can fail to be virtues of either sort. They fail to be character-virtues when the agent is badly motivated (e.g., caring only about “looking smart”) or exercises bad judgment by being “so open-minded that their brains fall out” or so humble that they (e.g.) obsessively focus on their limitations. And, they fail to be effects-virtues when they produce a preponderance of bad epistemic effects.

Below we begin to argue that being open-minded in response to extremist views might produce a preponderance of bad epistemic effects, whereas, being intellectually humble might avoid that fate.Footnote ¹⁴ Put differently, we begin to argue that intellectually virtuous people (here, people with effects-virtues) might respond humbly but not open-mindedly to extremist views they know are false, in which case we should do the same. With this in mind, Section 2 amasses some support for the claim that responding open-mindedly to extremist views can produce bad epistemic effects, whereas responding humbly can avoid those bad effects. In our view, the effects of open-minded and intellectually humble responses can diverge because only the former require us (1) to engage with others and (2) to be willing to revise our beliefs. We do not pretend to canvass all of the epistemic effects of such responses. Nor do we attempt to draw definitive conclusions about whether responding open-mindedly would indeed produce a preponderance of bad epistemic effects or whether responding humbly would not (as this is an empirical matter). Our goal is to provide initial reasons for taking such a view seriously since it is often overlooked in the literature.

We noted above that the trait of IH* does not require success in changing or mitigating one’s limitations. Accordingly, one might wonder whether IH* is even a candidate for effects-virtue, given the latter’s focus on successful effects.Footnote ¹⁵ In our view, IH* is a candidate for effects-virtue, since a quality needn’t require the production of good effects in order to produce good effects (see Baehr Reference Baehr2011b: 52–54). IH* will be an effects-virtue just as long as it produces a preponderance of good effects.Footnote ¹⁶ So, while it is true that successfully changing or mitigating one’s limitations is not required for owning them via IH*, one will sometimes—perhaps often—be able to change or mitigate one’s limitations, increasing one’s strengths and knowledge and thereby producing good epistemic effects. And, even in cases where one is unable to change or mitigate one’s limitations, one might still produce the good epistemic effect of knowing what one’s limitations are.

2.1. Responding open-mindedly to extremist views

We already know that (e.g.) Trump didn’t win the 2020 election, and that White supremacy is false. Given that we know these things, what are the possible effects on us of responding open-mindedly to views to the contrary? Do we risk losing our knowledge? Might our confidence in our convictions drop? Might we amass epistemic opportunity costs? We argue that even if we aren’t at risk for outright knowledge-loss, our confidence in our convictions might drop, and we might incur epistemic opportunity costs.

To respond open-mindedly to extremist views, we would need to do two things: (1) engage with extremist views that compete with our beliefs, and (2) be willing to revise our beliefs that the extremist views are false. And, we would need to do these things despite our knowledge that the extremist views are false. Let’s grant that unless we are in a hostile epistemic environment, responding open-mindedly to extremist views isn’t likely to result in the outright loss of our beliefs or in changing our minds. Nor is it likely to result in the outright loss of justification or knowledge.Footnote ¹⁷ Even so, responding open-mindedly to extremist views in non-hostile environments might distort our confidence (and, perhaps, our degree of justification) in our beliefs; it might lower our confidence (and justification) when it shouldn’t be lowered.Footnote ¹⁸ Why might it do this? The short answer is that even if OM*’s “revision” requirement wouldn’t put us at risk for lowering our confidence, its “engagement” requirement would. Arguably, OM*’s revision requirement does put us at risk for lowering our confidence—we would need to lower our extremely high degree of confidence in our belief that (e.g.) White supremacy is false in order to consider revising it (see Section 1.3). But, even if we set that point aside, the repetition and fluency effects of engaging with extremist views might still weaken our conviction that they are false. Empirical research (Begg et al. Reference Begg, Anas and Farinnaci1992) suggests that frequent exposure to a claim can increase both fluency in processing it and confidence in the truth of the claim while decreasing confidence in claims that conflict with it. Accordingly, regular engagement with extremist views might weaken confidence in our beliefs that they are false. Repeated exposure to (e.g.) racist views like White supremacy might also prime us to have higher levels of confidence than we should in false claims about Black persons (e.g., that most mass shootings in the US have been committed by Black men).Footnote ¹⁹ In short, repeated engagement with extremist views might saddle us with distorted confidence (and justification) in two ways: too little confidence in our beliefs that they are false, and too much confidence in claims that are consistent with them.

Adding insult to injury, we might also waste epistemic resources by engaging with extremist views. By spending our time engaging with extremist views that we already know are false, we can incur epistemic opportunity costs. We might have been spending that time on more valuable intellectual endeavors, and on our own projects and questions, rather than questions that have already been asked and answered at length.

Accordingly, responding open-mindedly to extremist views can have bad epistemic effects on us as knowledge-possessing agents. We may retain our knowledge that extremist views are false, but at the cost of lowered confidence in our beliefs, missed epistemic opportunities, and an increased susceptibility to believing false claims consistent with extremist views.

Importantly, responding open-mindedly to extremist views may also have bad effects on the epistemic character of advocates of extremist views. To explain, responding open-mindedly can signal to advocates that they are not just epistemic agents, which they are, but epistemic agents whose level of credibility on the topic merits a willingness on our part to revise our beliefs, which they are not. Put differently, responding open-mindedly can grant advocates a credibility excess, wrongly signaling to them that their credibility warrants re-opening our own inquiry into the topic. As Medina (Reference Medina2013: 60) has argued, credibility excess can facilitate vices like intellectual arrogance in those on whom it is bestowed.

Finally, responding open-mindedly to extremist views can also produce bad effects on the broader epistemic environment. Here, we focus primarily on responding online. Responding open-mindedly to extremist views online can disseminate falsehoods to third-party viewers, saddle them with repetition and priming effects, and signal to them that advocates of extremist views are credible sources on the topic, though they are not.

First, open-mindedly engaging with advocates of extremist views online can facilitate the spread of misinformation to third-parties, and increase the availability of misinformation in the environment, making it easier to encounter. Depending on how influential the knowledge-possessing agent is (e.g., how many followers they have), the sheer act of engaging, instead of ignoring, advocates of extremist views can algorithmically boost their views, leading to the spread of falsehoods and the pollution of the epistemic environment. Second, open-mindedly engaging with advocates of extremist views can also platform them. That is, it can generate the impression in third-party observers that extremist views are worth taking seriously, and their advocates are credible sources. As Levy (Reference Levy2019) argues, platforming can create higher-order evidence in favor of the claims of the person platformed and their credibility as a source. Put differently, responding open-mindedly could spread credibility excess and begin to cement it in the environment. Third, all of this might spread negative priming and repetition effects to third-parties, leading to the acquisition of false beliefs in at least some, and, reduced confidence in the falsity of extremist views in others.

Clearly, we haven’t canvassed all the potential effects of responding open-mindedly to extremist views and more work remains. Since we haven’t explored whether responding open-mindedly might also produce some good epistemic effects, we aren’t in a position to draw conclusions about the preponderance of effects it might produce. Still, we hope to have amassed initial inductive support for the view that responding open-mindedly can produce some bad epistemic effects, a point that is often overlooked in the literature. Relatedly, we might wonder whether Daryl Davis’s engagement with Klan members, even though it wasn’t open-minded, might have produced bad epistemic effects for Davis, his interlocutors, and the epistemic environment (Ferkany Reference Ferkany2019: 410).

2.2. Responding to extremist views with intellectual humility but not open-mindedness

Let’s turn to responding to extremist views with intellectual humility but not open-mindedness. One might worry that responding humbly to extremist views would be just as bad for us, for advocates of extremist views, and for the epistemic environment as responding open-mindedly, since it would also risk producing the bad effects above. But we think intellectually humble responses can avoid these bad epistemic effects because they require neither (1) engagement with extremist views and their advocates, nor (2) any willingness to revise our beliefs.

For starters, intellectually humble responses may not put us at risk for repetition and priming effects, for incurring epistemic opportunity costs, or for lowering our confidence in our beliefs that extremist views are false. To explain, because IH* is self-oriented and directed toward our own limitations, we can respond humbly to extremist views without regularly engaging with them or their advocates. An agent needn’t regularly engage with White supremacy, or White supremacists, to realize that she is likely to cast them as irredeemable monsters or that she is largely ignorant about the argumentative moves they make, or that her processes and beliefs are fallible, and some of her evidence is comparatively weaker. She might come to these realizations through reflection on the fallibility of human reason, or on her own poor reasoning and ignorance, after only limited exposure to White supremacy. Nor need she repeatedly engage with White supremacy or White supremacists to own these cognitive limitations. She can acknowledge and care about her limitations by admitting them to herself and developing plans to try to address them, none of which requires such engagement. Importantly, owning her limited abilities in persuasion might even tell against direct engagement with advocates of extremism, especially in cases where her limitations in this regard are numerous and the risks of platforming are high.

Since humble responses don’t require repeatedly engaging with extremist views or their advocates, they can avoid the negative repetition and priming effects addressed above. They can likewise avoid any reductions of confidence (and justification) in the falsity of extremist views that might result from repetition effects (see Hannon and Kidd Reference Hannon and Kidd2022). Humble responses can also avoid the epistemic opportunity costs of repeated engagements, freeing up resources for more worthwhile epistemic projects. Humble responses that succeed in avoiding these negative effects may even produce some positive effects for us, including self-knowledge about our own limitations, and knowledge about the status of advocates as human epistemic agents (not monsters).

But importantly, even if intellectually humble responses can avoid these negative effects, not all humble responses will. Only virtuously humble responses will avoid them; excessively humble (servile) responses won’t. With that in mind, let’s revisit the case of Allison Gornik, who owned the limitations in her powers to persuade Derek Black by conducting extensive research on White supremacy. Gornik engaged frequently and extensively with Black, and with other sources of White supremacy and conducted research on her own time. Accordingly, she is a candidate for repetition and priming effects, and for having less confidence than she should in the falsity of White supremacy. She is also a candidate for epistemic opportunity costs, given the immense time and effort she devoted to preparing arguments and engaging with Black. Does that mean she was excessively, rather than virtuously, humble in the way she owned her limitations?

That is a live question, whose answer depends on the preponderance of effects her response produced, including effects for her, Black, and the environment. Though we won’t be drawing any conclusions about the preponderance of effects of humble responses, it is worth pointing out that Gornik’s case would be hard to judge. Even if the epistemic effects for Gornik were predominantly negative, given the magnitude of her burden, the overall preponderance of epistemic effects might still have been positive. Black did change his mind (Saslow Reference Saslow2018) and is now an advocate against White supremacy and racism, so the epistemic effects for him and the broader environment might have been positive. But, Gornik’s way of owning limitations also prioritized engagement with Black, which platformed him, putting fellow students at risk for negative repetition and priming effects. While it is unclear whether Gornik’s humble response produced a preponderance of good or bad effects, and unclear whether it was virtuous or excessive, we can say this much. Though there will be some cases in which we should own our limitations while continuing to engage directly with advocates of extremist views—e.g., when our limitations are few and we are likely to persuade them to change their minds with minimal negative impact overall—there will be many other cases in which we should own our limitations while opting out of direct engagement—e.g., when our limitations are numerous, we are unlikely to succeed, and the risks of negatively impacting ourselves and the environment are high. Does opting out of direct engagement mean we should write-off advocates of extremist views, leaving them to their own devices?

With that question in mind, let’s explore some potential effects of humble responses on advocates of extremist views. As illustrated above, intellectual humility can help prevent us from jumping to the conclusion that they lack the capacity to reason and learn. It can help prevent us from regarding them as irredeemable while allowing us to categorically reject their views. It can help us block the inferential leap from denying what they say to denying their epistemic agency (Gunn Reference Gunn, Cassam and Samarzija2023). In helping to prevent us from drawing such conclusions and helping us see advocates of extremist views as epistemic agents, responding humbly can support virtues like intellectual beneficence. Roughly, this virtue directs our attention outward toward helping other epistemic agents gain epistemic goods. While the virtue of intellectual beneficence won’t always advise us to help other epistemic agents, much less advocates of extremist views (since the risks of bad effects will sometimes be too high), it will sometimes advise us to help advocates, in cases where the costs of doing nothing to counteract extremist views would be even higher. More specifically, intellectual beneficence will sometimes advise us to help advocates through direct (typically private) engagement, but due to its risks, will more often advise us to help advocates indirectly through improvements to our shared epistemic environment. These can take the form of (e.g.) structural changes to content algorithms that reduce echo chambers and the spread of falsehoods (Rini Reference Rini2017), and increased access to education and critical thinking (McIntyre Reference McIntyre2018). In short, humble responses can support the virtue of intellectual beneficence by helping us acknowledge the epistemic agency of advocates of extremist views. Together, intellectual humility and beneficence can help prevent us from simply writing them off, even in cases where direct engagement would be inadvisable.

What about cases where beneficence advises direct engagement? Here, we might worry that responding humbly but not open-mindedly would only make things worse for advocates of extremist views—causing their views and epistemic vices to become even more entrenched. As Fantl (Reference Fantl2018: 169) notes: “It is rational to want to be unreceptive to those whose minds are closed against you. If your interlocutor and audience were fully aware of your closed-minded attitudes toward their arguments, it would make sense for them to want to be unreceptive to you.” That is, it would make sense for them to respond in kind by doubling-down on their beliefs rather than being willing to change their minds. In reply, responding humbly might help us avoid these effects because it can help us see and engage with advocates as epistemic agents who have the capacity to learn. It can help us acknowledge their epistemic agency, which in turn might give advocates a reason to acknowledge us as agents and eventually become willing to change their minds.Footnote ²⁰ Together with beneficence, it can help us demonstrate that we care both about maintaining knowledge ourselves and about helping other epistemic agents gain knowledge, which explains why we aren’t willing to change our minds and why we are engaging with advocates. Megan Phelps-Roper Reference Phelps-Roper2017, who attributes her departure from the Westboro Baptist Church to her conversations with friends online, captures this point in her 2017 TED talk: “My friends on Twitter didn’t abandon their beliefs or their principles, only their scorn. They channeled their infinitely justifiable offense and came to me with pointed questions tempered with kindness and humor…. They approached me as a human being and that was more transformative than two full decades of outrage.” Phelps-Roper’s friends weren’t open-minded, but they were (presumably) intellectually humble in coming to see her as an epistemic agent and in owning their tendency to conclude otherwise. Phelps-Roper responded by recognizing their agency, and (eventually) by being willing to change her mind.

Relatedly, we might worry that responding humbly in direct engagements with advocates of extremist views would produce the same negative effects for advocates as responding open-mindedly. In reply, humble responses that are transparently closed-minded can avoid these negative effects because they needn’t assign advocates a credibility excess. Because they aren’t open-minded, humble responses enable us to regard advocates of extremist views as epistemic agents without regarding them as credible sources about the matters at hand. They can thus avoid assigning credibility excesses to advocates and can avoid the negative effects on advocates that credibility excesses might cause.

Finally, let’s turn to some potential effects of humble responses on the epistemic environment. Here, as above, the differences between humble and open-minded responses matter. Because humble responses don’t require engaging with advocates of extremist views, they can avoid platforming them and spreading their views, avoiding repetition and priming effects on third parties. Owning the limitations in our powers of persuasion can even help us identify cases where we shouldn’t engage—e.g., where our limitations are numerous, and we are unlikely to prevent the conversation from being hijacked.

But, what about cases where beneficence advises direct engagement, even online? Could humble responses still avoid the negative effects of platforming? Perhaps, if the humble agent is able to “responsibly platform” the advocate; i.e. if they have the requisite argumentative skills to head off the negative environmental effects associated with platforming. Of course, exactly which cases these are would be difficult to judge and one’s argumentative limitations would need to be quite minor. But the main reason it is possible for humble, but not open-minded, responses to “responsibly platform” stems from the fact that humility does not entail a willingness to revise one’s beliefs. Provided that the humble agent clearly and transparently communicates her unwillingness to revise her own beliefs and her aim to change the minds of advocates, she can avoid giving third-party observers the impression that advocates are credible sources. In bringing her argumentative powers to bear on the engagement, she also has the potential to refute the extremist views in question while minimizing their dissemination and repetition and priming effects on third parties.

We have suggested that responding open-mindedly to extremist views that we know are false can produce bad epistemic effects and that these bad effects can be avoided by responding humbly. Clearly, more work would need to be done to show that people with effects-virtues would respond humbly but not open-mindedly to extremist views that they know are false. But if that claim could be established, and if the behavior of people with effects-virtues is normative for our own behavior, then we, too, should respond to extremist views with intellectual humility but not open-mindedness.Footnote ²¹