1. What is Closed-mindedness?

Paul is unwilling to engage seriously with alternatives to his belief that ‘once a criminal, always a criminal’. We can use Paul's case, which is a paradigm of both closed-mindedness and dogmatism, to identify the key features of each of these two dispositions. Dismissing relevant alternatives to a belief – as Paul does – is one way to be both closed-minded and dogmatic. But, it isn't the only way to be closed-minded, or even the only way to be dogmatic. To see why, let's consider four features of Paul's case, none of which are necessary for closed-mindedness, and two of which – (2) and (3) – are also unnecessary for dogmatism.Footnote ³

(1) Having beliefs about the topic. Paul already believes that ‘once a criminal, always a criminal’, and in dismissing relevant alternatives to this belief, he is being closed-minded. But, closed-mindedness does not require already having beliefs about a given topic. Consider Pauline who has no beliefs about this topic and is being confronted with evidence for the very first time. Pauline can still arrive at an initial belief by conducting a closed-minded inquiry. For instance, she may ignore or be oblivious to evidence that supports the success of reentry programs. Accordingly, closed-mindedness doesn't require having extant beliefs about the given topic. But dogmatism does; it requires a belief about which the agent is dogmatic. Paul is both closed-minded and dogmatic; Pauline is closed-minded but not dogmatic.

(2) The locus of ideas and evidence. Paul is closed-minded with respect to ideas and evidence that compete with a belief he already holds. Pauline is closed-minded in the way that she handles ideas and evidence in the process of arriving at a belief – she ignores relevant evidence. But, one can also be closed-minded in the ways that one conducts inquiries more generally. One can be closed-minded with respect to which questions one asks, which sources one consults, and which methods one uses. Accordingly, the locus of closed-mindedness isn't restricted to ideas and evidence. Its locus includes ideas and evidence, but also includes other intellectual options, like relevant questions, sources, and methods. The locus of dogmatism isn't restricted to ideas and evidence either. The dogmatic agent who willfully ‘hides her head in the sand’ closes herself off to any further evidence, but also to any further sources, questions, and methods that might be relevant.

(3) Engaging with intellectual options. In dismissing intellectual options that cross his path, Paul engages with those options, at least insofar as he recognizes and rejects them, but he doesn't engage seriously with them – he doesn't evaluate the merits of competing ideas or the arguments for them. His engagement is entirely superficial. But, closed-mindedness and dogmatism don't even require this much, since one can be closed-minded or dogmatic by failing (in various ways) to engage with intellectual options. For instance, (i) one could ignore (rather than dismiss) intellectual options that one recognizes. Or, (ii) one could be oblivious to intellectual options – e.g., one could fail to recognize relevant defeaters or sources in the first place. One way to be oblivious to relevant sources is to be testimonially unjust.Footnote ⁴ Alternatively, (iii) one could fail to seek out or generate intellectual options. For instance, an agent could fail to look beyond his own echo chamber, or only look for evidence that confirms his belief that ‘once a criminal, always a criminal’.

(4) Unwillingness. Paul is unwilling to engage seriously with relevant intellectual options. Unwillingness is arguably required for dogmatism, but it isn't required for the broader category of closed-mindedness. One can be closed-minded by being unable, albeit willing, to engage seriously with intellectual options. Consider Oblivia, who shares Paul's belief that the incarcerated are irredeemable. Unlike Paul, Oblivia is perfectly willing to engage seriously with intellectual options to this belief; but she is also systematically oblivious to relevant options and thus fails to engage with them.Footnote ⁵ She may have passively inherited this impaired perception from her surrounding society,Footnote ⁶ or it may have been actively indoctrinated in her. Whatever its cause, Oblivia's impaired perception makes her closed-minded. Importantly, our environments can make us closed-minded even when we don't want to be closed-minded. This has two consequences. First, people who are unable to engage with intellectual options because of bad luck in their environments or their constitutions are closed-minded, but aren't blameworthy (in the standard sense) for coming to possess closed-mindedness. Second, as an inability, closed-mindedness can be an environmentally produced impairment, and can even result from a hard-wired impairment – it need not be a character trait.Footnote ⁷

The upshot of all this is that closed-mindedness (CM) is an unwillingness or inability to engage seriously with relevant intellectual options. Dogmatism is a sub-set of closed-mindedness: (DG) it is an unwillingness to engage seriously with relevant alternatives to a belief one already holds. There are three addenda.

First, though the examples above feature beliefs that are false, one can also be closed-minded and dogmatic with respect to beliefs that are true. This is because closed-mindedness and dogmatism do not pertain to the contents of beliefs. They pertain to the agent's unwillingness or inability to engage seriously with relevant options to her beliefs, whatever their content. Accordingly, an agent can be dogmatic with respect to her religious beliefs, whether those beliefs are true or false. She can be dogmatic with respect to her political beliefs, whether they are true or false. And, she can even be dogmatic with respect to true and relatively unimportant beliefs; e.g., that her pet is well behaved.

Second, an agent's closed-mindedness might be restricted to a particular domain. An agent might only be closed-minded about her pets (e.g., whether her pets are well behaved), or about her pets and her children (e.g., whether her pets and her children are well behaved), and not about anything else. She thus has a domain-specific disposition to be closed-minded, but lacks a general disposition to be closed-minded. It is also possible for an agent to perform a closed-minded action as a one-off – to do what a closed-minded person would do – while lacking either a domain-specific disposition or a general disposition to be closed-minded. To illustrate, on a particular occasion our agent might dismiss evidence that her dog was behaving badly, even though she does not usually dismiss such evidence.

Third, the analyses above define closed-mindedness and dogmatism in terms of relevant intellectual options. To see why this restriction is needed, consider Priscilla, a police detective who is investigating a standard break-in. In ignoring the possibility that David Bowie's ghost (or Ronald Reagan's ghost, etc.) did it, Priscilla is not being closed-minded. She isn't being closed-minded because these options aren't relevant to her inquiry. To be closed-minded, she would need to ignore an option that is relevant (e.g., the teenager responsible for the break-in next door also conducted this one). This much should be clear: closed-mindedness and dogmatism require relevancy restrictions on intellectual options. What is not clear, and what warrants exploration, are the precise conditions on relevancy. To provide a complete picture of closed-mindedness and dogmatism, we would need to answer the question: which options are, and which are not, relevant for an agent in an inquiry, and why? Alas, I won't be defending any single answer to this question, though I take the viable candidates for relevancy conditions to be familiar enough. Epistemic externalists will define relevant/irrelevant options for an agent in an inquiry to be: those options that are objectively likely/unlikely to be true (or likely/unlikely to be helpful in reaching the truth). Externalists may also include options that the agent's community reliably believes to be likely/unlikely to be true (or to be likely/unlikely to be helpful in reaching the truth). Whereas, internalists will take relevant/irrelevant options to be those that the agent, or community, has good reason to believe are likely/unlikely to be true. I won't be plumping for either of these over the other. But, the good news is that whether we are internalists or externalists, intellectual options like ‘2 + 2 = 5’, ‘The Holocaust never happened’, and ‘The earth is flat’ will be irrelevant in ordinary environments – since these options are in fact false, and we (and our communities) believe reliably, and with good reason, that they are false. This means that in ordinary environments we aren't closed-minded in ignoring these options, because they aren't relevant.Footnote ⁸

Let's now explore a different candidate condition on relevancy, one that will play a role in the discussion below. Call it the ‘pervasiveness’ condition. It claims that the widespread presence or absence of an option in an environment suffices to make that option relevant or irrelevant for the agent. It is a live question as to whether we should endorse the pervasiveness condition. On the one hand, we have reason to think that the absence of an option in the agent's environment is not enough to make it irrelevant to her inquiry. Consider George Orwell's Oceania, in which the Ministry of Truth re-writes options, disseminating fabrications that support the party line, and destroying facts that do not.Footnote ⁹ Arguably, the facts that it destroys are still relevant options for agents conducting inquiries. For instance, we see Winston Smith struggling to reconstruct the events of his childhood without the benefit of any external record of them. To put the point differently, the epistemic environment in Oceania makes its subjects more closed-minded, not less. It doesn't decrease the number of relevant options; it makes those relevant options harder to access. This counts against the pervasiveness condition.

On the other hand, we have reason to think that the ubiquity of an option in an environment might make it relevant to the agent's inquiry. This counts in favour of the pervasiveness condition. Return to Oceania – an epistemic environment that is thoroughly polluted with ‘alternative facts’. Arguably, the ubiquitous presence of such options makes them relevant to the agent's inquiry in the same way that the widespread presence of fake barns makes that option relevant. The option that a barn was fake wasn't relevant until one stumbled into fake barn country, where it became relevant and stayed relevant, even though the agent's beliefs – ‘That's a barn’ – are true.Footnote ¹⁰ (The agent happens to look at the one real barn in the area.) Likewise, the option that ‘ignorance is strength’ wasn't relevant until one woke up in Oceania, where it became relevant and stayed relevant, even though the agent's beliefs – ‘Ignorance is not strength’ – are true. For our purposes below, the key issue will be this: in ordinary environments, options like ‘The earth is flat’, ‘2 + 2 = 5’, and ‘The Holocaust never happened’ are not relevant, and thus we aren't closed-minded in ignoring them. But, if the pervasiveness condition proves viable, then in epistemically hostile environments, options like ‘2 + 2 = 5’ will be relevant, and we will be closed-minded in ignoring them. Is such closed-mindedness intellectually vicious? Or, might it even be intellectually virtuous?

2. Closed-mindedness as an Intellectual Vice

Thus far, I have adopted working analyses of the dispositions of closed-mindedness and dogmatism. These analyses do not presuppose that closed-mindedness and dogmatism are intellectual vices. To count as intellectual vices, they will need to meet further conditions. Which conditions? That depends on our analysis of intellectual vice. Arguably, there is more than one kind of intellectual vice: (a) effects-vice, (b) responsibilist-vice, and (c) personalist-vice. Roughly, closed-mindedness and dogmatism will be (a) effects-vices whenever they produce a preponderance of bad epistemic effects (or fail to produce a preponderance of good epistemic effects). They will be (b) responsibilist-vices whenever they are bad epistemic character traits for whose possession the agent is blameworthy (accountable). Finally, they will be (c) personalist-vices whenever they are bad epistemic character traits for whose possession the agent is not blameworthy (accountable). The distinction between effects-vice and responsibilist-vice should be familiar to virtue epistemologists, since it corresponds to that between reliabilist- and responsibilist-virtue. We need the additional category of personalist-vice to capture the bad epistemic character traits of the indoctrinated (e.g., graduates of Hitler's Jugend.) My view is that in standard cases, like Paul's, the dispositions of closed-mindedness and dogmatism are intellectual vices. At a minimum, Paul's closed-mindedness meets the conditions for an effects-vice. Since effects-vices can be epistemic character traits, Paul's closed-mindedness may also meet the conditions for a responsibilist-vice or a personalist-vice (depending on the details of its acquisition). Here, I focus on effects-vice.

Briefly, effects-vices are stable cognitive dispositions that either consistently produce a preponderance of bad epistemic effects, or consistently fail to produce a preponderance of good epistemic effects. If we understand vices to be the contraries of virtues – such that one could simultaneously fail to have either – then effects-vices will produce a preponderance of bad epistemic effects. Whereas, if we understand vices to be the contradictories of virtues – such that whenever one fails to have a virtue, one thereby has a vice – then effects-vices will fail to produce a preponderance of good epistemic effects.Footnote ¹¹ Any stable cognitive disposition – be it a character trait, an environmentally produced impairment, or a hard-wired impairment – that meets these conditions will be an effects-vice. Accordingly, readers can take their pick: in the below, readers are (e.g.) welcome to operate on the assumption that closed-mindedness is a character trait.

Why think that in standard cases, like Paul's, the disposition of closed-mindedness will be an effects-vice? For starters, closed-mindedness can produce a heap of bad epistemic effects for the closed-minded agent, for other agents, and for the epistemic environment.Footnote ¹²

(1) Bad epistemic effects for the closed-minded agent. Closed-mindedness enables the agent who possesses it: (i) to sustain false beliefs that he already has. Paul's closed-mindedness enables him to sustain his false belief that ‘once a criminal, always a criminal.’ Similarly, in failing to look for sources outside our own epistemic bubbles, we may be sustaining false beliefs that we already have. Indeed, repeated agreement among the ‘friends’ in our bubbles may even lead us to mistakenly strengthen our confidence in our beliefs.Footnote ¹³ Closed-mindedness can also: (ii) prevent the agent from acquiring true beliefs and knowledge. Paul's closed-mindedness prevents him from acquiring true beliefs about whether incarcerated people can change. Moreover, closed-mindedness can: (iii) compound and expand an agent's extant system of false beliefs. It can lead agents to doxastically double down. We see this in the case of Samuel A. Cartwright, a nineteenth century white American doctor who believed that slaves lacked agency. When confronted with contrary evidence – their attempts to escape slavery – Cartwright doxastically doubled-down: he judged escape attempts to be manifestations of a mental disorder peculiar to slaves, which he invented out of whole cloth. Relatedly, closed-mindedness can lead agents: (iv) to pursue irrelevant questions, projects, and inquiries, and thus to waste their epistemic resources.

(2) Bad epistemic effects for other agents. In being closed-minded with respect to sources, one might: (i) fail to see another agent as a source of knowledge when she is, assigning her a ‘credibility deficit’.Footnote ¹⁴ As a one-off, the harm done by a single instance of such closed-mindedness may be relatively ephemeral. But, closed-mindedness with respect to sources can take the form of testimonial injustice, whereby the closed-minded agent systematically overlooks the credibility of women and people of color, wronging them as ‘giver[s] of knowledge’.Footnote ¹⁵ Relatedly, such closed-mindedness can: (ii) impede the development of intellectual virtues in women and people of color, and facilitate their development of intellectual vices. Additionally, it can: (iii) result in their exclusion from educational institutions, obstructing their acquisition of knowledge. The closed-minded agent can also: (iv) assign too much credibility to sources within his epistemic bubble. He may see them as sources of knowledge when they are not, assigning them a credibility excess. This, too, can cause agents to develop intellectual vices.Footnote ¹⁶ Similarly: (v) the closed-mindedness of agents who set the intellectual agenda for others – school boards, journalists – can be ‘epistemically corrupting’; i.e. it can ‘encourage the development and exercise of epistemic vices’ in others.Footnote ¹⁷

(3) Bad epistemic effects for the environment. Closed-mindedness can also lead to: (i) the intentional or unintentional pollution of the epistemic environment. Closed-minded agents who pursue irrelevant inquiries and do so sincerely, e.g., ‘true believers’ like Cartwright (above), can unintentionally disseminate falsehoods in their environments. Closed-minded agents who re-post the claims of their ‘friends’ without seeking independent corroboration, can likewise populate their feeds with unwitting falsehoods. By inadvertently polluting their epistemic environments with false claims, these closed-minded agents may also be: (ii) obfuscating truths and knowledge, making them harder to find. Of course, pollution will be a matter of degree: the wider the dissemination of falsehoods, the more polluted the environment; at extreme levels of saturation, polluted environments will be hostile. Arguably, closed-mindedness can also lead to intentional pollution and obfuscation. Let's grant that intentionally polluting the environment with claims one knows to be false, and intentionally concealing claims one knows to be true, involve dishonesty and deceit rather than closed-mindedness. Even so, dogmatic ‘true believers’ can still intentionally hide or erase competing ideas (that they incorrectly believe to be false) in an effort to prevent others from believing them. We see this repeatedly in book-burnings, and also in the EPA's 2017 decision to delete or move information about anthropogenic climate change from its main webpage into its archive.Footnote ¹⁸ Dogmatic ‘true believers’ can also intentionally populate the environment with options that will divert agents who would otherwise endorse competing views. ‘True believers’ may do this in a sincere effort to ‘control the message’, or to disseminate what they think are truths, or to manufacture doubt. For instance, they may publicize studies on the role of natural variability in climate change.Footnote ¹⁹

In short, the disposition of closed-mindedness can produce a ream of bad epistemic effects. Now, to be an effects-vice, it must produce a preponderance of bad epistemic effects, or fail to produce a preponderance of good epistemic effects. Does it meet these conditions? Below, I argue that the disposition of closed-mindedness will minimize the production of bad effects in hostile epistemic environments. But, in ordinary epistemic environments like ours,Footnote ²⁰ it is still reasonable to think that the disposition of closed-mindedness (usually) meets the conditions of an effects-vice. As a failure to seek out sources beyond our bubbles, it leads to misplaced confidence and credibility excess. As testimonial injustice, it obstructs the intellectual virtues of other agents, facilitating intellectual vice. As dogmatism about a belief that is false, it results in the maintenance, strengthening, and compounding of false beliefs. And, in all of these forms, it obstructs the acquisition of knowledge. That is a plethora of bad epistemic effects. Even at our most conservative, we can conclude that closed-mindedness sometimes, perhaps often, fails to produce a preponderance of good epistemic effects. That's enough to make it an effects-vice.

The disposition of closed-mindedness also meets the conditions for an effects-vice when the belief one is being closed-minded about is true.Footnote ²¹ Suppose I believe that my pet is well behaved, and that this belief is true but doesn't constitute knowledge (I am unjustified in believing it). In ignoring relevant options – I refuse to engage with relevant evidence to the contrary or with relevant sources who criticize my pet's behavior – I am ignoring options that are indeed false and unreliable. My closed-mindedness enables me to sustain a true belief. But, in ignoring these relevant options, I am not engaging with them seriously – I am not evaluating them on their merits. Accordingly, I may be sustaining a true belief that my pet is well behaved while blocking my ability to gain related epistemic goods, like knowledge that my pet is well behaved, or an understanding of what makes her well behaved. Here, too, closed-mindedness may obstruct the acquisition of knowledge. And, of course, it may do this while simultaneously producing many of the other bad effects mentioned above, including misplaced confidence. Here, too, we can at least conclude that closed-mindedness sometimes, perhaps often, fails to produce a preponderance of epistemic goods.

3. Closed-mindedness as an Intellectual Virtue: One-off Instances of Closed-minded Action in Ordinary Environments

I have argued that the disposition of closed-mindedness is an intellectual vice in standard cases like Paul's, and more generally, that it is an intellectual vice in ordinary epistemic environments. But, is it always an intellectual vice? In epistemically hostile environments, might the disposition of closed-mindedness even be an intellectual-virtue, albeit a ‘burdened’ virtue of some sort?Footnote ²² I explore that question in the concluding section. Here, I address a worry about my analysis of closed-mindedness as it applies to ordinary environments. My analysis of closed-mindedness (CM) is broad. It encompasses closed-minded actions in addition to dispositions, and entails that closed-mindedness will be more common than we might have thought. It even entails that we act in closed-minded ways with some frequency. The worry is that (CM) is so broad that instances of closed-minded action will outstrip instances of intellectually vicious action. Hence, closed-mindedness won't always be intellectually vicious. I embrace this worry and its implications. The below argues that there are instances of closed-minded action that are intellectually virtuous (though these may not be quite as common as the objector thinks). Indeed, I think there are advantages to approaching the analysis of closed-mindedness, and its status as an intellectual vice, as distinct questions. Namely, this approach can help us home in on what makes closed-mindedness a vice when it is one, and on what makes it a virtue when it is one.

Here, as above, let's restrict the discussion to effects-virtues and vices. Accordingly, our question is: are there any one-off instances of closed-minded action that produce more good epistemic effects than bad ones in ordinary environments? Clearly, there are instances of ignoring options – e.g., that the earth is flat, that 2 + 2 = 5, that the Holocaust never happened – that produce a preponderance of good epistemic effects. But, at least in ordinary environments, cases like these are beside the point because in ignoring these options one isn't being closed-minded. Closed-mindedness requires ignoring relevant options, and the options above are not relevant in ordinary environments. Whatever relevancy conditions we adopt – internalist, externalist, pervasiveness – these options will fall short. What we need are instances of ignoring relevant options that produce more good epistemic effects than bad ones.

There are at least three sorts of candidate cases to address. First, consider being closed-minded with respect to knowledge you already possess. In his ‘Dogmatism Paradox’, Saul Kripke asks whether possession of knowledge that p could justify ignoring future evidence against p.Footnote ²³ His answer is that it sometimes can.Footnote ²⁴ Ernest Sosa describes Kripke's reasoning as follows:

Might the closed-minded actions described here enable the agent to sustain her knowledge, and prevent her from devoting epistemic resources to relevant but ultimately misguided options? Do these instances of closed-mindedness produce an overall preponderance of good epistemic effects? In my view, the jury is still out. Ignoring options that are relevant but misguided will prevent the agent from devoting resources to those options, thus freeing up those resources for more promising epistemic pursuits. It will also preserve her true belief that p. The question is whether it causes her to lose her knowledge that p. Quassim Cassam thinks it does, Jeremy Fantl thinks it doesn't, and Ernest Sosa takes the road between. Cassam argues that when an agent is confronted with relevant evidence against p, which she can't refute and closed-mindedly dismisses, she loses her justification for p and thus loses her knowledge that p.Footnote ²⁶ In direct contrast, Fantl contends that the agent can sometimes retain her knowledge that p when she dismisses a relevant counterargument that she can't refute.Footnote ²⁷ Whereas, Sosa thinks the agent retains her animal knowledge that p, but is prevented from having reflective knowledge that p. For Sosa, the closed-minded agent is still ‘apt’, but not ‘fully apt’:

We can say at least this much: closed-mindedness with respect to knowledge you already possess produces some good epistemic effects (it preserves truth). But, the jury is still out on exactly which bad epistemic effects it produces. Accordingly, it is an open question as to whether it produces a preponderance of good epistemic effects. We can also conclude that if knowledge is lost rather than preserved, then closed-mindedness with respect to knowledge won't be a clear advance on closed-mindedness with respect to unjustified true belief (e.g., that my pet is well-behaved). And, so, if closed-mindedness with respect to unjustified true belief fails to produce a preponderance of good epistemic effects (as argued above), so will closed-mindedness with respect to knowledge.

Second, it is worth considering whether the closed-minded behavior of a group-member might contribute to the production of epistemic goods by the group as a whole. In this vein, Christopher Hookway suggests that ‘a research team may benefit from having some members who are dogmatic, and unwilling to take on board new possibilities, while others are much more ready to take seriously seemingly wild speculations’.Footnote ²⁹ The suggestion is that the closed-minded behavior of a group-member might help a group of diverse agents produce a preponderance of epistemic goods. Relatedly, it is worth considering whether a group composed entirely of closed-minded agents might produce a preponderance of epistemic goods. Adam Morton is optimistic about this possibility. In his words:

Relatedly, Miranda Fricker employs the example of a debating club, all of whose members are prejudiced, but whose prejudices cancel one another out, rendering the group as a whole neutral.Footnote ³¹ Though it is unclear whether the closed-minded behavior of one or more group-members would help a group produce a preponderance of epistemic goods, this question is worthy of exploration.

Even if the jury is still out on the two considerations above, I think the third consideration is decisive. Think about the most recent article you wrote, or about what it took to actually start writing your dissertation. At some point, you likely ignored relevant options, in order to focus on developing your own answer – not because it was your own, but because it was the answer! Because you thought it was true. You stopped reading alternative views – you tuned them out. You knew that another article had just been published on your topic, but you ignored it, in an effort to make progress on the solution you thought was correct. According to (CM), this behavior is closed-minded. Moreover, there will be some cases where it produces a preponderance of good epistemic effects; e.g., where researchers on the verge of a big discovery (e.g., the cure for a disease) ignore relevant (but different) work that has just been published in order to push forward and successfully complete their own line of inquiry. They may even (causally) need to ignore that work in order to make their discovery. Note that the goods here produced are epistemic – the researchers produce knowledge. The method of production is also epistemic – they push forward in their own inquiry. Their motives are likewise epistemic – they are motivated to attain knowledge. So, the values in play are epistemic and not, or not merely, pragmatic.Footnote ³²

Granted, some cases of ignoring relevant options will produce a preponderance of merely pragmatic goods, or moral goods, but not epistemic goods. As when closed-mindedly avoiding inquiries into one's own health, and sustaining the false belief that one is fine, actually helps one heal. This category may also include cases of closing off inquiry after one has gained just enough knowledge, though these cases can be tricky. For starters, consider the graduating senior who waits until the last minute to write a term paper. He addresses only two sources on the topic, ignoring other sources and closing off his inquiry in order to meet the deadline. Arguably, his closed-minded behavior helps him produce pragmatic goods – shutting down his inquiry helps him produce, in this case, a barely passing paper, which is a necessary condition for graduating. But, it should be fairly clear that shutting down his inquiry doesn't help him gain any epistemic goods. Unlike the researchers above, the graduating senior doesn't continue his inquiry – he turns the paper in and walks away from the subject. For him, ignoring relevant options isn't a necessary step in the production of epistemic goods; it is an impediment to the production of further or better epistemic goods. Epistemically, he would be much better off if he kept his inquiry open. Consider a more complicated case. Suppose I have promised a friend that I will make pavlova for her party tomorrow. Having never made it before, I read the recipes of twenty celebrity bakers. I ignore the remaining 10 million recipes turned up by Google and close off my inquiry, in order to keep my promise and get the pavlova done. Here, too, closed-minded action arguably helps me produce a pragmatic good – closing off my inquiry helps me produce, in this case, a good pavlova. Closed-minded action also helps me produce a moral good – keeping a promise. Does it help me produce any epistemic goods? Perhaps, not – ignoring other recipes may be an impediment to the production of further or better epistemic goods. It may be the twenty-fifth recipe that gets me special insight into baking meringue. On the other hand, perhaps I have hit the point of diminishing epistemic returns – after having read twenty recipes (which seems like a lot!), keeping my inquiry open may not produce further or better epistemic goods. Accordingly, ignoring other recipes may prevent me from devoting further epistemic resources and amassing epistemic opportunity costs.

In sum, however we end up classifying these tough cases, there will be some instances of closed-minded action – e.g., of the researchers above – that are epistemically virtuous. The other candidate cases above also warrant further exploration.

4. Closed-mindedness as an Intellectual Virtue: The Disposition of Closed-mindedness in Epistemically Hostile Environments

What about the disposition of closed-mindedness? Is it always an intellectual vice, or could it be an effects-virtue in epistemically hostile environments? Could it be a ‘burdened’ virtue of some sort – or at least a disposition that is only useful for surviving in environments that are hostile or oppressive?Footnote ³³ Recall the pervasiveness condition on relevant options – let's assume, for the sake of argument, that it holds. Accordingly, the widespread presence of an intellectual option in an environment will suffice to make that option relevant. Further, let's suppose that an epistemically hostile environment is not minimally or moderately polluted, but extremely polluted – it is utterly saturated with intellectual options that are false, unreliable, or aimed at misdirection.Footnote ³⁴ Some of these options will be explicit statements (e.g., ‘ignorance is strength’), some will be unreliable sources (e.g., the dimwits in the film Idiocracy), and some will be implicit norms (e.g., that discredit women and people of color as sources of knowledge). The pervasiveness condition renders these options relevant in epistemically hostile environments.

So, what is a knowledge-possessing agent to do when she finds herself in an epistemically hostile environment? My proposal is that there are epistemic reasons for her to be closed-minded – to be unwilling to engage seriously with relevant intellectual options that conflict with what she already knows. That is, if she knows that, e.g., ‘2 + 2 = 4’, ‘ignorance is not strength’, and ‘the earth is round’, then there are externalist epistemic reasons for her to refuse to engage with the options that, e.g., ‘2 + 2 = 5’, ‘ignorance is strength,’ and ‘the earth is flat.’ Why should she be closed-minded? Because, in an epistemically hostile environment, closed-mindedness is an effects-virtue. When a knowledge-possessing agent is stuck in an epistemically hostile environment, surrounded by falsehoods, incompetent sources, and diversions, closed-mindedness about options that conflict with what she knows will minimize the production of bad epistemic effects for her.

To explicate, suppose the knowledge-possessing agent wakes up in Mike Judge's Idiocracy Footnote ³⁵, which is flooded with incompetent sources, or in Orwell's Oceania, which is flooded with lies. Closed-mindedness about options that conflict with what she knows will avert at least one bad epistemic outcome and produce at least two epistemic goods for this agent. First, it will enable her to sustain the true beliefs she already has. That is one good, and it is not insignificant – there is considerable risk of her coming to believe what the Ministry of Truth wants her to believe (its control is totalizing, its slogans appear on every screen, etc.) Second, closed-mindedness will prevent her from devoting epistemic resources to options that are relevant (due to pervasiveness) but misguided and from amassing epistemic opportunity costs. It averts that bad outcome. It thus frees up those resources for more promising epistemic pursuits; it enables her to continue to pursue her own intellectual projects and options (e.g., Winston Smith keeps a journal). That is a second good. Should she closed-mindedly forego opportunities to understand why the ‘idiots’ in the Idiocracy believe what they do? I think she can, though the answer will partly depend on whether such understanding could effect change in the hostile environment. Suppose it couldn't, and suppose she would be amassing epistemic opportunity costs in pursuing ‘idiot-diagnosis’ instead of other projects. Closed-mindedness would avert those epistemic opportunity costs.

Some agents – members of non-dominant groups – don't have to imagine being in an epistemically hostile environment. They already live in one. Our current epistemic environment routinely discredits women and people of color as sources of knowledge. It does this even though it isn't entirely over-run with misinformation – despite Kellyanne Conway's ‘alternative facts’ and Facebook's dissemination of Russian propaganda, our current epistemic environment is not saturated with falsehoods about every topic. On this score, it falls several magnitudes shy of Orwell's Oceania. But, it is thoroughly saturated with norms that discredit women and people of color – these norms run deep and systematically track agents across domains. Hostility can take different forms, and our environment is hostile for these agents. Accordingly, closed-mindedness might be an important resource for members of non-dominant groups. Here, too, it will enable knowledge-possessing agents to hold onto their true beliefs, avert epistemic opportunity costs, and pursue their own intellectual projects. Importantly and ironically, it might also help them ward off the vice of intellectual servility – closing themselves off from denials of their credibility might prevent them from losing confidence in their intellectual strengths and over-attributing limitations to themselves.Footnote ³⁶

One important unanswered question is whether closed-minded agents in epistemically hostile environments retain their knowledge. Hostile environments compound this problem because they are unsafe; i.e. they are environments in which any agent – closed-minded or not – could easily go wrong when revisiting her belief.Footnote ³⁷ In short, hostile environments may themselves rob an agent of knowledge, whether or not she is closed-minded; they are hostile, after all. Suppose an agent knows that countries X and Y are at war, but then wakes up in Oceania, in which the Ministry of Truth has replaced all references to the war with a sanitized history. When the agent revisits his belief, he can now easily go wrong. Is he thereby robbed of knowledge? If so, can he somehow inoculate himself against this unsafe environment by being closed-minded – by ignoring the falsehoods spouted by the Ministry of Truth? Alternatively, suppose we were to claim that the agent retains his knowledge despite his unsafe environment (knowledge doesn't require safety). Does his closed-mindedness then cause him to lose his knowledge (in a manner similar to that described by Cassam above)? Whatever conclusions we draw about knowledge, I submit that in epistemically hostile environments, the disposition of closed-mindedness still succeeds in minimizing bad epistemic effects for the agent himself, even if it doesn't produce an outright preponderance of good epistemic effects for the agent. In such environments, we may have to sacrifice knowledge in order to avoid even worse epistemic effects. Such is the power of hostile environments.

Does the closed-mindedness of the knowledge-possessing agent minimize bad epistemic effects for other agents in the hostile environment? One might worry that it does not. Indeed, an objector might argue that in order to minimize bad epistemic effects for deluded or incompetent others in Oceania and the Idiocracy, the knowledge-possessing agent should be somewhat open-minded. Even if she is ultimately unwilling to revise her own beliefs (2 + 2 = 4), she should still engage seriously with deluded or incompetent others, and the options they endorse, in an effort to change their minds and practices.Footnote ³⁸

I want to suggest an avenue of reply. Deluded or incompetent others in hostile environments like Oceania and the Idiocracy are unlikely to change their minds and practices as a result of one-on-one engagements with a knowledge-possessing agent. Such engagements are unlikely to produce the desired epistemic goods. They are also likely to produce bad epistemic outcomes for the knowledge-possessing agent herself (see above). Accordingly, the combined outcome of such engagements – likely epistemic bads for the agent herself and unlikely epistemic goods for others – also favors closed-mindedness.Footnote ³⁹

By way of further reply, it is worth considering whether serious engagement with deluded or incompetent others in a hostile environment might do them or the environment an epistemic disservice. When the knowledge-possessing agent engages seriously with deluded or incompetent others, might she be inflating their epistemic credibility, or at least signaling to them that they are credible enough to be taken seriously (though they aren't)? José Medina argues that credibility excess facilitates the development of intellectual vices, and can begin to do so over the course of a single conversation.Footnote ⁴⁰Accordingly, we can ask whether engaging with deluded or incompetent others facilitates, or sustains, intellectual vices via credibility excess. If it does, this would be a further strike against engagement. Relatedly, in engaging seriously with the Ministry of Truth and its fabrications, might our knowledge-possessing agent make the epistemic environment worse, rather than better? One might argue that our agent would need to choose her engagements strategically, in an effort to avoid the Ministry's smear campaigns (and avoid being executed). But, in a hostile environment like Oceania, that might not be feasible. Our agent's engagements might unavoidably be fodder for the Ministry's manipulations; in which case, they would inadvertently contribute to making the environment more polluted, not less. This would also count against engagement.

To sum up, I have argued that in an epistemically hostile environment, closed-mindedness on the part of the knowledge-possessing agent would minimize bad epistemic effects for the agent herself. This is enough to make her closed-mindedness an effects-virtue, since her engagement with other agents is unlikely to benefit them, and might even do them and the epistemic environment a disservice. To put the point differently, this is enough to make her closed-mindedness a ‘burdened’ effects-virtue in a hostile epistemic environment.

Now, for some caveats. First, I am not arguing that the knowledge-possessing agent in a hostile environment should be closed-minded in every domain or possess the general disposition of closed-mindedness. Rather, I am arguing that she should be closed-minded about relevant intellectual options that conflict with what she already knows. That will cover many domains, since the environment is hostile. Still, she can be largely open-minded in the way she conducts her own intellectual projects or her projects with epistemically reliable allies – here, she should still brainstorm relevant options. Nor must she give up open-mindedness as a valued goal for herself or the environment. Even if she can't make her hostile environment more open-minded by being an exemplar of open-mindedness herself – by engaging seriously with deluded or incompetent others and the fabrications they endorse – she may try to facilitate open-mindedness in other ways. (Perhaps, populating the environment with truths, to compete with the received falsehoods, would be a start.)

Second, I am not arguing that the knowledge-possessing agent should abandon her motivation for truth or develop epistemic motivations that are intrinsically bad. I am not advocating that she develop the responsibilist vice of closed-mindedness. I am only arguing that in hostile environments, her disposition of closed-mindedness will minimize the production of bad epistemic effects.

Nor, third, am I arguing that morally, pragmatically, or politically, the knowledge-possessing agent should be closed-minded. Rather, I am merely arguing that she has epistemic reasons to be closed-minded. Closed-mindedness on the part of the knowledge-possessing agent minimizes bad epistemic effects for the agent herself, and may also minimize bad epistemic effects for other agents, and the environment. If closed-mindedness produces bad moral, pragmatic, and political effects in hostile environments (or our current environment), those will need to be weighed against, and may trump, its epistemic effects.

Fourth, importantly, I advise caution in any attempts to apply the claims above to our current epistemic environment. Our current epistemic environment is not ‘hostile’ (for members of dominant groups), as I am using that term. It may be moderately polluted, but it isn't extremely polluted – it isn't utterly saturated with false and unreliable options in the way that Orwell's Oceania and Judge's Idiocracy are. This has two repercussions for attempts to apply the claims above. First, options like ‘the earth is flat’ may not be pervasive enough in our current environment to count as epistemically relevant. If they don't count as epistemically relevant, then in ignoring them, we aren't being closed-minded. Accordingly, the claims above would fail to apply: since we wouldn't be closed-minded in ignoring flat-earthers, the issue of whether we should be never gets off the ground. But, second, even if some misguided options (‘Donald Trump is a good President’) are pervasive enough in our current environment to count as relevant, and even if we are closed-minded in ignoring them, our closed-mindedness might produce different epistemic effects than it did above. We will need to ask whether, in our current environment, closed-mindedness on the part of the knowledge-possessing agent will minimize bad epistemic effects for other agents. It may not. After all, a proportion of Trump-voters can be convinced otherwise – engaging with them may ultimately produce a preponderance of good epistemic effects for them. Likewise, we will need to ask whether closed-mindedness on the part of knowledge-possessing agents will minimize bad epistemic effects for our current environment. Again, it may not; if it is corrupting – if it facilitates closed-mindedness – it may not minimize bad epistemic effects.

Fifth, one might worry that the reasoning above could be used by far-right conservatives to justify their closed-mindedness. Consider, for instance, religious fundamentalists who deny the rights of women. Such conservatives, believing that they are in a hostile environment and that they have knowledge, could defend their closed-mindedness on those grounds. In reply, their beliefs would be false. They are not in a hostile environment, nor are they knowledge-possessing agents – their beliefs about women are false. Nor would their closed-mindedness minimize the production of bad epistemic effects. Quite the contrary! It would be akin to Paul's closed-mindedness, and to epistemic injustice, which are effects-vices, not effects-virtues. Recall that effects-virtues and effects-vices are defined externally (along reliabilist lines). So, even if these conservatives believed that their closed-mindedness was an effects-virtue, they would be wrong. This is a strength of epistemic externalism. Granted, those of us on the progressive end of the political spectrum (who are in dominant groups) aren't in a hostile environment either (see point four). This means we also need to exercise caution (even when we have knowledge) – we, too, should avoid jumping to the conclusion that we are in a hostile environment that justifies our closed-mindedness.

This brings us to a final set of open questions, which are well worth exploring. How does one know whether one is in a hostile environment? And, relatedly, how does one know whether one's closed-mindedness would be intellectually virtuous or vicious – how does one know when one should be closed-minded? The virtue epistemologist's answer is that one will need to do what an open-minded person would do – engage with relevant options – in order to know whether one is in a hostile environment. Likewise, for knowing the epistemic effects of one's closed-mindedness (where this is an admitted weakness of epistemic externalism and of consequentialist views more broadly.) This doesn't mean that such knowledge requires possessing the general disposition of open-mindedness. But, it does mean that those who already possess the general disposition of closed-mindedness will have a tough time gaining such knowledge. Breaking that cycle might require finding a cure for closed-mindedness.

In sum, I have argued that in standard cases, including Paul's, the disposition of closed-mindedness is an intellectual vice. But, I have also argued that closed-mindedness can be intellectually virtuous. In ordinary environments, some one-off instances of closed-minded action will produce a preponderance of good epistemic effects. Moreover, in epistemically hostile environments, the disposition of closed-mindedness will be an effects-virtue, albeit a ‘burdened’ one.Footnote ⁴¹