Epistemic Autonomy and the Shaping of Our Epistemic Lives

ABSTRACT

I present an account of epistemic autonomy as a distinctively wide-ranging epistemic virtue, one that helps us to understand a range of phenomena that might otherwise seem quite disparate – from the appropriate selection of epistemic methods, stances and topics of inquiry, to the harms of epistemic oppression, gaslighting and related phenomena. The account draws on four elements commonly incorporated into accounts of personal autonomy: (i) self-governance, (ii) authenticity, (iii) self-creation and (iv) independence. I further argue that for a distinctively epistemic virtue of autonomy; the above elements must ultimately reliably lead to valuable epistemic goods (for the agent herself and others). I then turn to the domains or ways in which epistemic autonomy so understood, can be made manifest. I suggest that epistemic autonomy is a virtue that allows us to appropriately choose (i) subject matters and areas of inquiry, (ii) methods, sources, and processes of belief formation, (iii) epistemic goals and (iv) epistemic stances or frameworks. So understood, epistemic autonomy has a role to play in shaping most every aspect of our epistemic lives.

KEYWORDS:

• Autonomy
• epistemic autonomy
• epistemic injustice
• epistemic virtue

In what follows I present an account of epistemic autonomy as a distinctively wide-ranging epistemic virtue – one that fundamentally shapes our epistemic lives. On the current account, epistemic autonomy underlies a range of phenomena that might otherwise seem quite disparate – from the selection of epistemic methods, stances and topics of inquiry, to the harms of epistemic oppression, gaslighting and related phenomena.

I begin with two short sections providing background for the current proposal. The first draws attention to varying conceptions of the scope of the epistemic – I will here assume a broad understanding of the epistemic (and will say a few words on behalf of such understandings later in the paper). The second section turns to early accounts of epistemic autonomy that treat it primarily as a matter of properly balancing an agent’s own epistemic efforts with relying on the testimony of others; I will contrast the current proposal with these more narrowly focused accounts.

With this background in hand, I develop the current account of epistemic autonomy by drawing on four elements commonly incorporated into accounts of personal autonomy: (i) self-governance, (ii) authenticity, (iii) self-creation and (iv) independence. I explore how we can incorporate these elements into the epistemic realm. I further argue that for a distinctively epistemic virtue of autonomy, the above elements must ultimately reliably lead to valuable epistemic goods (for the agent herself and others) – a fifth element. I then turn to the domains or ways in which epistemic autonomy, so understood, can be made manifest. I suggest that epistemic autonomy is a virtue that allows us to appropriately choose (i) subject matters and areas of inquiry, (ii) methods, sources and processes of belief formation, (iii) epistemic goals and (iv) epistemic stances or frameworks. As such, epistemic autonomy can be seen as having a role to play in most every aspect of our epistemic lives as reflective epistemic agents.

Throughout the paper I attempt to draw attention to the ways in which epistemic autonomy can be undermined or restricted by external agents as well as ways in which an agent might lack or fail to manifest such autonomy due to factors internal to the agent herself. On the other hand, I also attempt to illustrate the many benefits that accrue to individuals and their communities when epistemic autonomy is broadly supported and embraced, and when unjustified restrictions on such autonomy are minimized.

Broad versus Narrow Epistemology

There are some common understandings of the epistemic (and epistemology) whereby epistemic autonomy, as it will be treated here, would fail to be an epistemic virtue. For example, Richard Feldman’s account of evidentialism posits a single epistemic ‘ought’:

For any person S, time t, and proposition p, if S has any doxastic attitude at all towards p at t and S’s evidence at t supports p, then S epistemically ought to have the attitude toward p supported by S’s evidence at t. (Feldman 2000, 679)

Feldman rejects any genuinely epistemic duties to inquire or seek out additional evidence and so on; his construal of the epistemic is particularly narrow. Any recommendations or requirements concerning processes of inquiry will end up being moral, pragmatic or grounded in some other domain.

On the other hand, others embrace more expansive, broad understandings of epistemology, even if there is a distinctive core pertaining to knowledge and its analysis. For example, Ernest Sosa has suggested that

we do well to distinguish between two parts of epistemology: (a) theory of knowledge, and (b) intellectual ethics. The latter concerns evaluation and norms pertinent to intellectual matters generally, with sensitivity to the full span of intellectual values. It is therefore a much broader discipline than a theory of knowledge focused on the nature, conditions, and extent of human knowledge. (Sosa 2007, 89)

Epistemic autonomy, as treated here, would qualify as an epistemic virtue given this broad understanding of epistemology and the epistemic.

Finally, epistemic autonomy clearly falls within the realm of the epistemic if we embrace a broader social epistemology; indeed, it could be a particularly important virtue in this domain:

social epistemology is concerned with how people can best pursue the truth with the help of, or sometimes in the face of, other people or relevant social practices and institutions. It is also concerned with the pursuit of truth by groups, or collective agents. […] social epistemology proceeds on the idea that we often rely on others in our pursuit of truth. Accordingly, social epistemology’s core questions revolve around the nature, scope, and epistemic significance of this reliance: what are the ways we rely on others when seeking information, and how does our relying on others in these ways bear on the epistemic goodness of our resulting beliefs? (O’Connell, Goldman and Goldberg 2023)

While it is not the primary goal of this paper, I will on occasion address potential concerns that could be raised by those, like Feldman, who embrace narrow construals of the epistemic, and argue on behalf of a more liberal, wide-ranging understanding of the domain.

Epistemic Autonomy and Relying on Oneself vs Others

Much early work on epistemic autonomy (within contemporary discussion) focuses on the issue of how an epistemic agent ought to balance using her own faculties to obtain and assess relevant evidence when forming a belief, versus relying on the testimony of others, and perhaps not fully grasping all of the reasons for a given belief herself. And much of this early work treats epistemic autonomy as involving an extreme self-reliance. Consider, for example:

[The epistemically autonomous thinker] takes no one else’s word for anything, but accepts only what she has found out for herself, relying on her own cognitive faculties. (Fricker 2006, 225)

An epistemically autonomous subject is one who judges and decides for herself, where her judgments and decisions are reached on the basis of reasons which she has in her possession […] (Goldberg 2013, 169)

[Intellectual autonomy is] the good of being epistemically self-reliant […] (Pritchard 2016, 38)

It is widely agreed that such a pure epistemic autonomy – understood here as complete epistemic self-reliance – is hardly a desirable epistemic state. As Sosa notes,

[w]e are ourselves among our main sources of information. […] Without implicit trust in the testimony of our neighbours or our own memories, we would be greatly reduced epistemically, well below the level of an isolated Robinson Crusoe reliant on his memory at every turn. Indeed it is hard to see how any human could live once so radically reduced epistemically. (Sosa 2015, 186–7)

More recent work on epistemic autonomy tends to explicitly address the need for an appropriate balance between relying on ourselves and relying on others, depending on the circumstances; such work reflects an effort to characterize a virtuous epistemic autonomy:

[Intellectual autonomy] requires thinking for ourselves while relying on others appropriately - neither too much nor too little. (King 2021, 94–5)

[Epistemic autonomy requires the disposition to make] good judgement about how, and when to rely on your own thinking, as well as, how, and when to rely on the thinking of others […] (Matheson 2022a, 182–3)

Indeed, there is now widespread acknowledgement of the importance of others to the ongoing development and expression of an agent’s autonomy.¹

In what follows I focus on further aspects of epistemic autonomy, going beyond this balancing between self and other-reliance. In so doing, I attempt to build on the work of several authors who have already taken important steps in this direction.² In particular, I hope to contribute to current discussions by further elaborating on plausible constitutive elements of epistemic autonomy and further exploring ways in which epistemic autonomy can be made manifest – or restricted or undermined.

Five Elements of Epistemic Autonomy

There are, of course, myriad accounts of personal autonomy.³ In this section I focus on a range of elements found in many such accounts and explore how they might be combined and applied to an agent’s epistemic life. These include (i) self-governance, (ii) authenticity, (iii), self-creation and (iv) self-reliance. In addition, I will incorporate a fifth element, (v) epistemic virtuousness, that will ensure that our account captures a distinctively epistemic form of autonomy.

Self-Governance

Self-governance is the core element of personal autonomy. Broadly, it consists in giving oneself rules, commitments, values and plans and abiding by them. In the case of epistemic autonomy, it will involve pursuing an active epistemic life governed by one’s values, goals and commitments. It will involve choosing questions for inquiry, methods and processes by which to pursue these questions, in addition to choosing epistemic goals in various domains, as well as broader epistemic frameworks or stances (I will elaborate on each of these in section IV).

Still, bare self-governance in this sense does not yet capture a full, genuine autonomy. Such self-governance might rule out extreme forms of brainwashing or indoctrination such that an agent effectively has goals and commitments imposed upon her by another. It would also rule out involuntary compulsions outside of the agent’s own control. But an agent’s autonomy can be undermined or weakened in less extreme cases. For example, a woman might choose not to pursue a degree in an unwelcoming, male-dominated field or a particularly toxic department. This choice again is an instance of self-governance, but it seems clear that her autonomy is being undermined; she is not becoming the epistemic agent she wishes to be (with the skills, understanding and knowledge that would have come with her preferred degree). More generally, there will be a wide range of instances where individuals settle on (and pursue) various goals and commitments that do not resonate with their deepest commitments and values but are expedient given their social circumstances or environment. There is self-governance here, but it is constrained in potentially problematic ways.

Authenticity

Autonomy is a matter of degree – an individual can be more or less free of external pressures and internal compulsions in their selection and pursuit of various goals and commitments. Somogy Varga and Charles Guignon suggest that authenticity goes beyond basic autonomy in that it is ‘not just about being involved in the authorship of such a [self-governing] law, but about how this law fits with the wholeness of a person’s life, and how or whether it expresses who the person is’ (Varga and Charles 2023, section 1.2). Invoking the notion of authenticity may seem controversial here – after all, it might seem to overly complicate an account of autonomy, and the term is not widely used in discussing autonomy as such.

But the concept of authenticity can help us to capture what is lacking in cases of self-governance where an agent compromises and defers to external forces – both in the extreme cases of brainwashing or indoctrination and in more common cases of social or other pressures. Those presenting accounts of personal autonomy generally agree that mere self-governance is not yet enough for full autonomy and provide further conditions to capture what is missing. Thus, some appeal to an endorsement of an agent’s first-order desires and motives by their higher-order preferences or a coherence between an agent’s actions and long-term plans or commitments (see Bratman 2007; Frankfurt 1971), others look to the history of how a commitment or preference came to be embraced (Christman 1991) and so on. All of these various efforts attempt to capture the sense that a given commitment (or value or action) needs to be suitably grounded in the individual’s self in order to be autonomous; ‘authenticity’ seems a reasonable term for what is being sought even if not commonly used in this context.

In drawing attention to authenticity of this kind, we are not claiming that an agent must exhibit full authenticity in all of her epistemic projects in order to qualify as being virtuously epistemically autonomous. Indeed, much of what we do in our epistemic lives (as in other parts of our lives) is mildly indifferent or hardly speaks to our deepest values and interests. We need to watch for passing cars as we try to cross the street, calculate various figures to complete our taxes, etc. These can all exemplify virtuous epistemic autonomy, even if not the highest levels of authenticity – it’s not that finding out when the next train is due needs to speak deeply to our soul. But we will have a fuller, more complete epistemic autonomy to the extent that more of our inquiries are ones that genuinely capture our imagination and curiosity, ones that engage us deeply and that we deliberately seek out.

As noted above, an appeal to authenticity can help us to make sense of cases where we are self-governed but feel distanced from our chosen path. Maura Priest draws attention to the significant external pressures that can constrain the epistemic autonomy of philosophers in the academy (Priest 2022). She appeals to a conception of epistemic autonomy grounded solely in self-governance, though she is open to other conceptions or adding further elements. Priest provides the following example:

There might be pressure to make changes to an argument to please a referee, even if the author does not believe that these changes are epistemically best. […] I have done this myself: I responded to a referee comment, changing my paper in a way to please the referee even though I believed that these changes made the paper worse off. In this instance, I handed over some of my intellectual governance for the sake of professional expediency. Said differently, myself and other philosophers might calculate that inclusion, even inclusion that comes at the cost of sincerity, is better than exclusion. Hence, fear of exclusion from the philosophical community can motivate epistemic insincerity. (Priest 2022, 82–3)

I do not believe that we best capture the problem here by focusing on self-governance. Priest argues that our intellectual or epistemic self-governance is undermined or abandoned in such cases as we sacrifice what we believe is epistemically best for pragmatic or other reasons. But it’s not clear that there is a lack of epistemic self-governance. Instead, there is self-governance – it’s just that Priest has chosen to act on pragmatic or other concerns. The decision is still hers, and she still governs herself, even if she doesn’t follow what she thinks is epistemically best. Compare: a medical researcher might reject pursuing a certain study on human subjects due to grave moral concerns, even though it would be very powerful epistemically. The mere fact that she chooses to act for non-epistemic reasons does not entail that she thereby loses or even partially gives up her intellectual or epistemic self-governance; she simply acts on different reasons.

Priest comes closer to the current proposal when she notes that there is a lack of sincerity at stake. She doesn’t elaborate on what she means by ‘epistemic insincerity’, but I believe she intends something like authenticity, being the epistemic agent one wishes to be and acting on one’s deep interests and commitments. In choosing to modify a paper to satisfy a referee, the paper may become convoluted and inaccessible, reducing the paper’s reach or potential audience. The author may be forced to cut substantial material – even if this is the material that they think is most provocative and potentially impactful. They might end up sharing arguments and claims that they see as largely unimportant for their community while only sharing a small part of what had been most exciting and original. In such cases we begin to lose our voice as an epistemic agent, as one contributing to our community – there is a lack of authenticity in making these changes, and to that extent our epistemic autonomy is diminished in this instance.

Or consider a First Nations student forced to attend a residential school. They will presumably gain various kinds of knowledge and epistemic methods from their instructors. But to the extent that they still identify primarily as members of their First Nations community, they may feel (i) alienated or distanced from the methods and knowledge they do acquire and (ii) frustration and a sense of loss in being forbidden from using their own language (and all that comes with it) and in not gaining the knowledge and methods of their community. They could still be, to varying degrees, self-governing in acquiring the knowledge and methods taught in these settler schools. But their epistemic autonomy is clearly undermined, and they may well feel alienated from what they learn to the extent that comes at the cost of the epistemic methods and bodies of knowledge and understanding with which they identify.⁴

Why might such authentic autonomy be relevant from an epistemic point of view? After all, we can pursue lines of inquiry, have appropriate doxastic attitudes relative to our epistemic position and so on, even if we are coerced (or merely pressured) into doing so. There can be clear moral concerns, of course, as in the case of residential schools. But what are the epistemic impacts?

First, we can, ceteris paribus, expect better epistemic performances by agents when their inquiries are authentic in this way. As Priest writes,

all things considered equal, we can reasonably believe that work produced by dedicated and engaged agents will be better than work produced by agents that are neither dedicated nor engaged. (Priest 2022, 84)

When agents are engaged in epistemic projects that reflect their deep interests and spark their curiosity, we can typically expect them to put forth greater effort in coming to develop hypotheses and in acquiring relevant data, to demonstrate greater resilience in the face of setbacks and so on. As such, we can typically expect better epistemic results in such cases.⁵

Jonathan Matheson argues that in pursuing inquiry, even if unsuccessfully, agents can develop perseverance, humility and other epistemic virtues. They can also gain a richer understanding of a given topic, even if they do not arrive at a firm conclusion or answer. Similarly, they can acquire valuable knowledge or understanding of related issues and topics (Matheson 2022b, 12–14). In the case of authentically autonomous inquiry or reflection, we again have reason to expect that agents will be more likely to achieve these additional epistemic goods insofar as they are more likely to persevere and exert greater effort in pursuing such inquiries (ceteris paribus).

Such authenticity can epistemically benefit an agent’s community or communities, while its absence can harm them. Consider again the case of an indigenous student forbidden from cultivating their epistemic agency and projects in the ways they wish. Their indigenous community effectively loses an individual who could have contributed to the preservation, sharing and further development of their systems of knowledge and ways of knowing; this is especially important in cases where a community’s population is small and their epistemic works at risk. The broader community as whole also risks the loss of these knowledge systems and methods over time – as well as the potentially different questions, framings and heuristic resources that could have shaped future inquiry.⁶ Of course, if such students were instead able to pursue the epistemic paths of their indigenous communities, we could expect benefits to accrue to these communities (and others) through the presence of additional inquirers.

More generally, there could be significant benefits – epistemic but also moral and practical – to the extent that members of underrepresented or oppressed groups are able to engage in authentic epistemic projects that reflect their genuine interests – to inquire into issues that speak to them but that might be overlooked by others, to cultivate additional heuristic resources and epistemic processes or methods and so on.

Self-Creation

We form our epistemic selves through our choices – through the belief-forming methods we cultivate, the inquiries we choose to focus on, the bodies of knowledge and understanding we acquire and so on. As C.A.J. Coady writes,

we can decide what intellectual priorities we will pursue. We can ‘author’ our intellectual life and hence structure to some degree our intellectual world by determining what masteries we will seek to achieve, what importance we will give to different facts and intellectual projects and hence what place in our intellectual economy different truths will have. (Coady 2002, 369)

Of course, as we’ve noted above, this is an idealized picture – we often face a variety of social pressures and other factors that can greatly restrict or undermine our ability to shape our epistemic lives or selves.⁷ Still, for many, there is at least some scope for self-creation and determination.

Such self-creation is closely related to the self-governance and authenticity discussed above. But appealing to the concept of self-creation helps to preempt any assumption of an antecedent, unchanging, ‘true’ self to which we must be authentic. Rather, the process of creating an epistemic self involves ongoing feedback between the self we currently are, and the methods, projects and commitments (etc.) that we pursue. Over time these efforts and their results will shape our self – which in turn will shape what commitments, projects and goals (etc.) can be seen as authentic and reflecting our evolving epistemic interests and commitments.

Importantly – and for better or worse – an agent’s projects, methods, values and commitments can become epistemically autonomous over time, even if they are profoundly heteronomous when first taken on. Consider a student who is pressured into pursuing a pre-med program by her parents. She loathes organic chemistry and many other courses she is required to take. She feels alienated from her work, lacks any real interest in what she is studying and contemplates dropping out. She is simply going through the motions in her chosen program. Still, an effective and inspiring instructor might spark her interest in some of her courses; she becomes curious and desires to learn more. She might seek out opportunities to work in a faculty lab over the summer, etc. Over time, her epistemic self evolves; her inquiries become authentic and autonomous.

Other cases of such changes in what is authentic and autonomous for an agent may be more complicated. An indigenous student removed from his community may, over time, gradually become assimilated into his new surroundings and community. His interests change; he gradually comes to think of the epistemic methods of his (former) indigenous community as backwards and unscientific. Over time, he may come to sincerely and autonomously identify with the settler community and its epistemic practices, despite the alienation he may have experienced for many years. Perhaps if his life had been different – if he had not been taken away from his community – he would now be fully immersed in its epistemic norms and resources. But as things are now, that would be a very different life and a very different self from the one that has emerged.⁸ There may be no going back.

Self-Reliance/Independence/Self-Sufficiency

Independence is perhaps the fundamental notion here – in cultivating and creating a self that is self-governing and authentic, we are establishing ourselves as a distinct, independent agent, with a range of projects, values and commitments of our own. Taken to an extreme, we arrive at a conception of autonomy whereby an agent’s autonomy requires complete self-reliance, with no or minimal vulnerability or dependence on others, and a ‘freedom’ from external influences. Of course, we’ve already seen above (in section III) that such a conception of epistemic autonomy would unduly limit our ability to learn and gain understanding; we are deeply dependent on others across multiple aspects of our epistemic lives.

A more plausible approach would instead encourage agents to develop their epistemic skills and talents and to develop a broad range of epistemic interests, but with no aspiration to pure self-reliance or independence. By cultivating such skills, talents and interests we can benefit ourselves and our communities. As Heidi Grasswick suggests,

in contexts of oppression, autonomy-related virtues such as independent thinking, confidence and security in one’s own thought, and self-trust will be especially important for those in subjugated positions to develop if they are to have the ability to resist many epistemic dimensions of oppression. (Grasswick 2018, 205)

Grasswick spells this out further, drawing on elements of Coady’s (2002) account of epistemic autonomy:

[i]ndependence manifests itself by facing putative information with a degree of skepticism, ensuring that an agent is not dominated by the ideas of others. This will be an especially important virtue for the oppressed; the ideas of the oppressors tend to dominate social discourse. Intellectual self-creation involves the capacity to assess what is important and what intellectual priorities an agent ought to pursue. These priorities might be specific to an agent, making it important that she be able to identify these as her own, and not be pushed into following the epistemic priorities of others […] (Grasswick 2018, 205)

In becoming an independent, autonomous epistemic agent with a range of epistemic skills and abilities, an individual can contribute to her community in multiple ways. She might become an expert in a field or become sufficiently knowledgeable and well-equipped to identify genuine experts in various fields (even if she is not one herself). She might also provide valuable insight and thoughtful questions for her community in the absence of available experts. Independence and authenticity in her epistemic life can also help to ensure a broad, diverse range of voices contributing to (and shaping the progress of) her community’s epistemic activities. Finally, such independence and authenticity help to ensure that her own epistemic life is rich, with a focus on questions of importance or interest to her, and with the skills and abilities to successfully pursue such work.

Virtuous Epistemic Autonomy

An agent is virtuously epistemically autonomous to the extent that her chosen methods, sources and processes do in fact reliably lead to valuable knowledge, understanding, true belief and other epistemic goods – for herself, but also for others. Thus, an agent who authentically, autonomously becomes a person who attempts to use tarot card readings to divine truths about the number of grains of sand on their local beach would not possess a virtuous epistemic autonomy.

The claim that an epistemically virtuous autonomous agent acquires valuable knowledge (or other epistemic goods) is perhaps controversial. After all, knowledge might be valuable for moral, practical or myriad other reasons that go beyond the narrowly epistemic. Requiring that epistemically virtuous, autonomous agents acquire valuable epistemic goods may seem to unacceptably expand the scope of the epistemic.

One line of response to this concern would be to note that even on a narrow construal of the epistemic, some epistemic goods are distinctively valuable insofar as they are epistemically powerful. Knowledge that modus ponens is a valid form of inference is more valuable than knowledge that there are four leftover grains of rice on a given plate, even from a narrow epistemic point of view. So, we could restrict the scope of valuable knowledge (and other epistemic goods) to narrow epistemically valuable instances of these goods.

But I believe there is good reason to instead opt for a broader conception of the epistemic and the demands of epistemic virtue. Anne Baril nicely captures one of the main reasons for doing so:

If our aim is to articulate plausible epistemic norms—norms about how to deliberate, inquire and so forth—then it seems we must find a way to recognize the apparent fact of significance, the difference in status between trivial and non-trivial beliefs. If we do not, then (given that trivial true beliefs can be acquired more quickly, and with more safety, than true beliefs of a deep or important nature) it is unclear why we ought not recommend that agents engage in such menial epistemic projects as phonebook memorization and blimp counting, rather than in (intuitively) more epistemically valuable projects—learning about human culture or the laws of physics, for example. (Baril 2010, 216–7)⁹

Taken to an extreme, on a narrow understanding of the epistemic (such as Feldman’s), a government could restrict its citizens to memorizing phone books with no epistemic loss; after all, it would be easy for each to have an appropriate doxastic attitude in response to reading each number in the book.

More extreme still, it seems that even remembering is irrelevant to a narrow understanding of epistemic success. We could have a community of epistemic agents who all learn the very same phone number, immediately forget it, then learn it and forget it once again, over and over. Each time they respond appropriately to their evidence.¹⁰ We could even imagine that they only get to see the number from a distance and cannot bother putting on their glasses, so they can only form weak beliefs or middling credences about the phone number (‘it kind of looks like it might be 555–3093’), which they then forget and repeat over and over again. According to a narrow construal of epistemology and epistemic success, these agents are epistemically flawless – they form appropriately middling credences over and over, in accordance with their epistemic position at any given moment.¹¹ Indeed, they could be seen as aspirational epistemic role models … Our worry, then, is that such narrow construals of epistemology fail to capture important epistemic desiderata. This is starkly illustrated by the fact that an individual could be a flawless, ideal epistemic agent on such an account yet clearly fall short of being any sort of plausible epistemic exemplar.¹²

Manifestations of Epistemic Autonomy

With these core elements of a virtuous epistemic autonomy now in place, we can examine a range of domains where such autonomy can be made manifest: (i) selection of areas of inquiry, (ii) selection of methods and processes, (iii) selection of epistemic goals and (iv) selection of epistemic stances or frameworks. For each, I will briefly explain the domain and provide examples of how epistemic autonomy could be undermined or restricted in the given area.

Selection of Subject Matters/Areas of Inquiry

Virtuous epistemic autonomy can be exhibited in appropriately choosing areas of inquiry – an agent may choose to study calculus, the Pāli Canon of Buddhist texts, a technical trade and so on. An agent can govern herself in choosing to focus her attention on certain subject matters and in cultivating skills and techniques to succeed in her inquiries.

Such autonomy is restricted when, for example, a religious community forbids its members to encounter texts written by outsiders (thereby closing off paths of inquiry) or a government prevents an indigenous community from sharing its language and culture with its children. Of course, in some cases such restrictions may be morally (or epistemically) justified – governments restricting access to information on creating improvised explosive devices or protecting sensitive personal information of its citizens.

An agent’s epistemic autonomy could similarly be restricted by schools or families excluding options for areas of inquiry, such that the agent (particularly as a child or youth) simply never has an opportunity to cultivate an interest in a given area. Consider a community that simply does not provide education to girls past the age of ten – it’s not that further education is forbidden in this case (we might stipulate), but simply that it is no longer provided. Even so, the epistemic autonomy of the affected girls is thereby clearly restricted.

Finally, an agent’s autonomy with respect to areas of inquiry could be undermined if others coercively impose certain subject matters on the agent. It is one thing to prevent inquiry into certain areas; it’s another to instead impose subject matters on an individual. There can indoctrination that involves imposing certain views on what are worthy topics for inquiry or that pressures agents to focus their attention on certain fields or subjects. Once again, this might be justified in some cases – ensuring that a child gains a basic introduction to a range of fundamental topics even if he is frustrated at the time.

One could argue that such undermining or restrictions being placed on inquiry is morally or all things considered problematic but not an epistemic issue as such. After all, even if we cannot investigate certain topics, there are other topics available to us. We can still, at any given point, appropriately respond to our evidential position at that time; this would remain true even if we are pressured into studying only certain topics. We can still be excellent epistemic agents, regardless of subject matter.

In response, note that this objection relies on a narrow construal of the epistemic, and we’ve already provided some initial concerns for such construals (particularly in failing to capture plausible epistemic exemplars). But beyond this, and to focus on the current objection, note that the subject matters and individual propositions known and investigated in the community will differ when restrictions are placed on areas of inquiry. Future lines of investigation and inquiry, methods developed and understandings acquired would all also be affected. Imagine a community that restricts all inquiry to the same religious text. Clearly this would limit the introduction of new knowledge or understanding to the community, restrict further lines of inquiry and so forth. Plausibly, if we have a wide range of virtuously epistemically autonomous agents who face only minimal, well-justified restrictions on this autonomy, a community will gain more diverse and valuable knowledge and understanding.

Kathleen Okruhlik writes:

Once we recognize that the content of science is affected by the social arrangements that govern its practice and production, then those social arrangements acquire epistemic significance as do the affirmative action programs and other interventions undertaken to alter those social arrangements. Any adequate philosophy of science will have to take this into account. (Okruhlik 1994, 39)

Different individuals and communities within the broader overall community will likely draw attention to different areas of interest for inquiry and to different questions even within the same broad subject matters.

To this point we’ve considered external restrictions on epistemic autonomy with respect to areas of inquiry.¹³ But problems can arise from the agent herself. For example, an agent would fail to demonstrate virtuous epistemic autonomy in this respect if, for example, they have no interest in inquiry and lack curiosity – they fail to choose any subject matters to investigate. They might simply form beliefs about whatever happens to be available to them at a given moment, with little or no interest in the world around them. In influential recent work, Jane Friedman introduces the following zetetic principle:

ZIP If one wants to figure out Q^?, then one ought to take the necessary means to figuring out Q^?. (Friedman 2020, 503)

Other things being equal, it seems problematic if an inquirer fails to follow ZIP. But we might have related worries about an individual who never (or only rarely) engages in inquiry, someone who does not care about coming to understand anything at all. Such a person lacks epistemic goals and interests, and they would fail to manifest epistemic self-governance to the extent that they fail to engage in inquiry at all. They are highly unlikely to contribute much, if anything, to the stock of valuable knowledge and understanding in their community.

A lack of epistemic autonomy could also be exhibited if an agent finds herself forming beliefs with no apparent control or finds herself compelled to investigate certain questions. For example, an individual with an obsessive-compulsive disorder might find herself compelled to repeatedly check whether she’s locked a door, well past the point of any epistemic or practical purpose. To the extent that an agent feels alienated from her own belief-forming processes, she lacks epistemic autonomy.

Methods/Sources/Processes

Epistemic autonomy can be manifested in appropriately choosing methods of inquiry. An agent might choose which experts to consult, which texts to read, which tests to run, what equipment to use and so on. Here, an agent’s autonomy might be undermined or restricted if they are forbidden access to needed equipment, if key documents are censored and so on.¹⁴ Once again, in some cases such restrictions may well be justified – we can forbid morally appalling medical research. An engineering professor may rightfully refuse to accept a student’s work based on astrology. Not all methodologies are reliable (epistemically), and many could be morally suspect, regardless of their epistemic efficacy.

This is the aspect under which many, perhaps even most, discussions of epistemic autonomy could be placed – those that address what a virtuous epistemic autonomy would entail in cases where one might pursue inquiry oneself or instead rely on the testimony of others. But we should remember that this is simply one instance of a broader range of concerns. That is, an agent could also manifest epistemic autonomy in appropriately choosing methods for a given domain and for given purposes. A social science researcher might choose mixed or qualitative methods given a particular topic; quantitative methods might be more appropriate for others. A rough and ready test to categorize certain objects might be the best option if fine-grained accuracy is not needed in a given case, rather than a much more accurate but time-consuming and demanding test.

The appropriate choice of methods/sources/processes is not simply a matter of what might be most immediately reliable in a given case. For example, as Matheson (2022b) notes, there can be good reasons for a novice to continue to investigate an issue for herself even if she could instead simply defer to an expert.¹⁵ The appropriate choice can depend on the agent’s broader epistemic goals and interests. For example, it could be that I should study a topic myself, using methods with which I am only mildly familiar, if my goal is to become an expert in the topic or to develop expertise with these methods. I may wish to develop my epistemic skills and methods in a given domain – even if I could far more easily simply consult with an expert from the start. Of course, it will often be wise to later verify our results against those of the expert(s) and to evaluate our use of various methods, etc. Overall, the key point is that an appropriate, virtuously autonomous choice of methods and processes will depend on an agent’s epistemic goals (long-term and immediate), their interests and values, their potential for developing various epistemic skills or methods and so on.

Virtuously manifesting epistemic autonomy in the choice of methods and processes again relies on a background of a supportive social environment. Heidi Grasswick suggests that with respect to epistemic autonomy,

fostering nurturing epistemic environments where we can safely trust others to gain baseline knowledge through testimony, as well as test out and develop skills of how to sort through evidence critically, will be very important, especially for our children. Family and community contexts will be important environments for this work, and educational contexts will also serve as an important source of epistemic skills for agents who need to learn how to sort through evidence and negotiate different points of view. Inadequate access to education could threaten the level of epistemic autonomy that some may be able to achieve. (Grasswick 2018, 201)

In other words, to develop valuable epistemic skills, to acquire reliable methods, to learn to compare and understand their relative strengths and so on, we will be dependent upon a well-functioning supportive epistemic community.

The damage that can be caused to an epistemic agent’s autonomy when faced with a hostile epistemic environment can be significant. Consider Kate Abramson on gaslighting:

the phenomenon that’s come to be picked out with that term is a form of emotional manipulation in which the gaslighter tries (consciously or not) to induce in someone the sense that her reactions, perceptions, memories and/or beliefs are not just mistaken, but utterly without grounds—paradigmatically, so unfounded as to qualify as crazy. Gaslighting is, even at this level, quite unlike merely dismissing someone, for dismissal simply fails to take another seriously as an interlocutor, whereas gaslighting is aimed at getting another not to take herself seriously as an interlocutor. (Abramson 2014, 2; quoted in Leydon-Hardy 2021, 142)

Such gaslighting undermines the victim’s sense of self-trust and their ability to effectively assess their own reliability (in their various belief-forming methods) and the reliability of others (particularly the gaslighter). Similarly, Lauren Leydon-Hardy draws attention to epistemic harms of grooming relationships:

• distorting norms: what they are, should be, how they apply and whether one is tracking them accurately;

• constraining evidence: inculcating false beliefs to third parties, rendering them defective sources of information/consultation;

• inculcating false beliefs to target individuals and blocking access to other sources of testimony through shame, guilt, requests or promises of secrecy;

• eroding self-trust: denying through testimony or otherwise that one’s judgments are veridical/can be trusted (including expressions of dismay and disappointment, as well as outright denial);

• impeding reasoning: undermining a target individual’s ability to draw certain inferences, to reason according to the available evidence, to acquire the necessary evidence to reason healthfully, to possess self-trust, or to confidently hold or express certain beliefs as a result of the epistemic fog arising from all of the above. (Leydon-Hardy 2021, 135)

Here again, the groomer can radically undermine the epistemic autonomy of the targeted individual in multiple aspects, from self-trust to their epistemic norms and goals.

There are many other ways in which our epistemic autonomy can be restricted in this regard (of appropriately choosing methods or processes). An agent might not be given an opportunity to join a research group or not be allowed access to crucial equipment. Members of her community (including experts) might refuse to provide testimony on a topic the agent wishes to investigate. Here again, such restrictions might be justified in some instances (an eight-year old is presumably rightly denied a spot in a NASA research group…). But if such restrictions and exclusions are systematic and widespread, additional harms can accrue, whereby an agent begins to question himself and his competence as an epistemic agent. His self-trust could be undermined, and he may eventually fail to trust his use of faculties or processes where he is in fact highly skilled and reliable. He may fall victim to imposter syndrome or change his areas of inquiry as he worries – in light of the exclusions and rejection from his community – that he is not competent in the areas he genuinely wishes to study. Or he might change fields while still trusting himself. He might recognize that despite his competence, his inquiries will fail given the lack of support and cooperation from his community (or the aspirational community he wishes to join).

Epistemic Goals

An agent can exhibit epistemic autonomy in appropriately choosing her epistemic goals in various domains and at different times. Thus, for example, an individual might seek a thorough understanding of some fields in which she is particularly interested, while being content with mere true belief or middling credences in others (perhaps in gathering sports trivia for a pub quiz). An agent might seek near certainty when it comes to beliefs concerning issues related to her tax or other legal obligations, while being much less risk-averse when forming beliefs or credences involving trivial office gossip.

To the extent that an agent has a strong interest in a certain question or subject matter, we might see them as blameworthy to the extent that they fail to engage in appropriate inquiry, other things being equal. That is, if I seek to gain an understanding of a certain field or issue, yet I fail to take obvious, easy and effective steps to do so (again, all else being equal), this seems to be a shortcoming on my part qua inquirer.¹⁶ On the other hand, I would not typically be blameworthy as an inquirer if I fail to pursue other potential knowledge or understanding if I lack any goal of learning about these matters. In some domains I can rightly be happy with weakly-held beliefs, middling credences, agnosticism or simply no doxastic attitudes at all. There is no reflection on me as a poor epistemic agent in such cases.

A possible exception to this latter claim might be if there were knowledge or understanding so fundamental that all epistemic agents (or all human epistemic agents) ought to pursue it. It’s not clear that there is such. Another possible exception would be if I were to fail, more generally, to engage in inquiry into at least some important or valuable forms of knowledge or understanding. Compare our earlier discussion of autonomy in choosing areas of inquiry. As finite beings, we do not need to pursue all available valuable knowledge or understanding, and it is also typically permissible to pursue some relatively trivial knowledge. But it can be problematic if an agent fails to engage in inquiry at all or fails to engage sufficiently with potentially valuable knowledge or understanding.¹⁷

An agent can also fail to manifest a virtuous epistemic autonomy with respect to her epistemic goals if she chooses radically inappropriate goals in a given domain. Seeking absolute certainty with respect to the presence of life on some particular planet several galaxies over is unattainable – to actively pursue such certainty is futile. Such pursuit might be autonomous in the sense of a mere trait, as a goal that authentically comes from the agent. But it does not manifest a virtuous epistemic autonomy. Timeframes can also be relevant here – it is one thing to wish to gain a thorough understanding of population ecology, another to have a goal of achieving this in a day.

We can also manifest virtuous epistemic autonomy in our goals for sharing epistemic goods with our community. In some cases, we might engage in inquiry simply for our own sake. But more often we wish to share our knowledge and understanding – or at least have it available to others. Recall our discussion of Priest’s concerns about the ways in which academic philosophy may impede the epistemic autonomy of those working within the discipline. We might also want to share more speculative ideas with the hope that others will also inquire into their potential. We might want to raise certain questions – to throw certain accepted ideas into doubt or to encourage the course of future inquiry in our community. In all of these efforts we can demonstrate epistemic autonomy, particularly if we think in terms of a broader social epistemology and the pursuit of shared epistemic goals. We are shaping the knowledge and other goods available across our community and influencing future inquiry.

In turn, these latter considerations point to further ways in which others can restrict or undermine our epistemic autonomy. Familiar examples of testimonial injustice can illustrate this point.¹⁸ When the testimony of individuals of marginalized groups is systematically ignored or downplayed, these individuals are prevented from making potentially significant impacts upon shared bodies of knowledge and practices of inquiry (in addition to the possible harms of diminished self-trust, etc. discussed in connection with methods and processes above). Similarly, epistemic autonomy is clearly undermined and restricted by what Kristie Dotson refers to as ‘epistemic oppression’:

[p]ersuasively sharing epistemic resources can take on any number of forms, from one’s ability to utilize those resources for effective and accurate communication; to being able to use a given set of shared resources to make sense of one’s experiences; to the ability to rely upon the existence of fair and accurate standards within shared epistemic resources. Taken together, epistemic oppression refers to a persistent and unwarranted infringement on the ability to utilize persuasively shared epistemic resources that hinder one’s contributions to knowledge production. (Dotson 2014, 116)

Individuals facing epistemic oppression are unable to be the voice that they wish to be, to make the contributions they wish to make. They are, in effect, silenced or ignored within their community. At best, they might be forced to make inauthentic contributions – ones that the oppressed know will be accepted or approved.

Epistemic Frameworks/Stances

As noted above, we might (for example) seek understanding in one field, but mere true belief in another. But our epistemic autonomy can be made manifest more broadly in our embrace of epistemic frameworks or stances. Laura Callahan characterizes epistemic frameworks as follows:

[v]ery broadly, an agent’s framework will amount to her deepest commitments as to epistemically good ways of interpreting and evaluating evidence. Some think of frameworks as belief-forming dispositions (in idealized circumstances), weightings of the epistemic values across contexts, or credence functions. (Callahan 2021, 540)

On her preferred view, what she refers to as ‘epistemic existentialism’,

it’s not just that what’s rational to believe on the basis of evidence can vary according to agents’ frameworks, understood as passive aspects of individuals’ psychologies. Rather, what’s rational to believe on the basis of evidence is sensitive to agents’ choices and active commitments (as are frameworks themselves). What’s rational for us to believe on the basis of evidence is, at least in part, up to us. It can vary not only across individuals but for a single individual, overtime, as she herself makes differing epistemic commitments. (Callahan 2021, 540)

Such epistemic existentialism can be understood as a manifestation of epistemic autonomy. Thus, for example, you might place greater weight on the possibility of acquiring true belief in some particular domain at a given time, while I’m more cautious in this domain, and treat avoiding falsehoods as carrying greater weight. In turn, what is an epistemically rational response to evidence in this domain might vary for us – it might be rational for you to believe a given proposition, while I should suspend or withhold given our different weightings.

In the recent philosophy of science literature there has been discussion of a similar notion: epistemic stances.¹⁹ Anjan Chakravartty characterizes an epistemic stance as follows:

a stance is a strategy, or a combination of strategies, for generating factual beliefs. A stance makes no claim about reality, at least directly. It is rather a sort of epistemic ‘policy’ concerning which methodologies should be adopted in the generation of factual beliefs. […] stances are not themselves propositional; they are guidelines for ways of acting. One does not believe a stance in the way one believes a fact. Rather one commits to a stance, or adopts it. (Chakravartty 2004, 175)

Much as Callahan recognizes a certain voluntarism concerning frameworks, Chakravartty embraces a similar position concerning stances:

[a] voluntarist about stances is one who believes that rival stances—i.e. ones that cannot be adopted simultaneously without engendering some sort of inconsistency or incoherence—are rationally permissible, and that there can be no rational compulsion to adopt any one rational stance as opposed to a rival, so long as it too is rational. That is far from saying that all stances are rational, of course […] (Chakravartty 2011, 39)

Here again it seems that we could use the example of different comparative weightings of our twin goals of acquiring true beliefs and avoiding false beliefs as an instance where stances could rationally differ. As a further example, Van Fraassen (2002) suggests that scientific realists tend to embrace a metaphysical stance that prioritizes the epistemic value of explanation, while anti-realist empiricists embrace a stance that stresses following the methods of the sciences. Dalila Serebrinsky elaborates on this latter case:

the differences between realists and anti-realists go beyond a difference in the propositions each part accepts/rejects (as we have seen, these differences also exist within realism and empiricism). Realists and anti-realists (say, empiricists) differ in the way they do things, epistemic things. And the way they do such things depends on the values and commitments they have. That will determine, at least to some extent, the doxastic attitudes they adopt towards certain propositions, but those doxastic attitudes are just the result of those epistemic practices prescribed by each stance. As stances are not sets of propositions, but sets of practices, they are not true or false, but they can be rational or irrational. (Serebrinsky 2023, 3)

Regardless of how we ultimately understand the relationship between epistemic stances and frameworks, both concepts allow us a space for the manifestation of epistemic autonomy in the embrace of these epistemic strategies and values that shape our epistemic behavior (and in turn, what qualifies as justified or rational for us) more broadly.

Callahan suggests that

the epistemic existentialist has a natural way of explaining how we may be responsible for what it’s rational for us to believe, on the basis of our evidence. For according to the epistemic existentialist, we have some agency in shaping our frameworks. […] But if we are non-epistemic existentialist subjectivists, we may find it difficult to explain how people are accountable for what they believe. For if being rational entails believing in accordance with the framework one happens to have, over which one has no control, then people may be doomed to either problematic beliefs or irrationality. (Callahan 2021, 553)

Callahan can be understood as appealing to an autonomous exercise of agency in embracing and shaping our epistemic framework(s). Chakravartty points to similar considerations with respect to epistemic stances:

[i]n just the same way that one may be acculturated with certain beliefs, solely as a consequence of immersion in a local culture, one may find oneself holding a particular stance. […] Thinking about acculturation is important, of course, because the ways in which our beliefs and stances may be formed in response to ambient social, political, economic, and epistemological cultures is important. But it is also philosophically unreflective, in the sense that while such factors may well determine how beliefs and stances are passively absorbed, they do not determine—not necessarily, at any rate—the beliefs and stances we choose upon reflection. Let us distinguish, then, between merely taking a stance, which can be essentially passive, and choosing a stance, which is the outcome of reflection. (Chakravartty 2011, 40–1)

Here again, there is a turn to what we can plausibly consider epistemic autonomy in our reflective choice of an epistemic stance. An epistemic community with members who autonomously embrace a range of rational epistemic frameworks or stances will benefit from a range of perspectives, emphases and approaches. For example, a community will likely benefit from having a mix of members, some of whom are more cautious (and thereby more sceptical) in a given domain, while others are more concerned with gaining true beliefs.

How might one fail to embrace a virtuous, autonomous epistemic framework or stance? Most straightforwardly, one might embrace a framework or stance that is inconsistent and unreliable – one that leads to poor epistemic results. But we can also imagine cases where a community more or less indoctrinates its members with certain attitudes and assumptions with respect to epistemic matters, as Chakravartty notes above. This can be more or less coercive and lead to more or less reliable frameworks. But to the extent that the same framework is imposed within a community, it will miss out on the benefits of having diverse epistemic strategies at work.²⁰

Conclusion

On the proposal that has been sketched here, appropriately balancing investigating for ourselves and relying on the testimony of others is only a small aspect of epistemic autonomy, even if often an important one. Instead, we have characterized epistemic autonomy as an epistemic virtue that guides and shapes our epistemic lives as a whole – both reflecting and developing our individual interests, abilities and circumstances, all while constrained by a need to successfully achieve valuable epistemic goods for ourselves and others.

A final worry might be that this understanding of epistemic autonomy is thereby too broad and all-encompassing. After all, it shapes every aspect of our epistemic lives, from our choice of topics to investigate to our methods and epistemic stances. It is invoked in explaining the harms caused by gaslighting, epistemic oppression and related phenomena. It may thus seem implausibly wide-ranging.

But this is precisely what we should expect of an epistemic virtue that captures the nature and exercise of our autonomy and agency within the epistemic realm. Consider what Miranda Fricker suggests concerning the fundamental harm of testimonial injustice. According to Fricker, such injustice

involves insult to someone in respect of a capacity essential to human value […] When someone suffers a testimonial injustice, they are degraded qua knower, and they are symbolically degraded qua human. (Fricker 2007, 44)

Notice how the current proposal complements Fricker’s suggestion – in instances of gaslighting, testimonial injustice and related harms, those who endure them are degraded precisely as autonomous agents and not simply as knowers. Our approach here parallels the relationship between oppression and autonomy more generally – oppression restricts and undermines autonomy.²¹ And on the other hand, strong epistemic communities that support well-founded trust, collaboration, free and open inquiry and that encourage the autonomous contributions of diverse individuals and communities will enhance and support the cultivation of virtuous epistemic autonomy.²²

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

Support for open-access publishing was provided by the Colgate University Research Council and the Carl Benton Straub ’58 Endowed Chair in Culture and the Environment.

Notes on contributors

Jason Kawall

Jason Kawall is Carl Benton Straub ‘58 Endowed Chair in Culture and the Environment at Colgate University. His research focuses on issues in ethics (both normative and metaethics), environmental ethics and epistemology, with an emphasis on virtue-based approaches to each.

Notes

1. Grasswick (2018) has been particularly important in this regard, with its careful attention to relational understandings of autonomy.

2. See, for example, Battaly (2022), Coady (2002), Grasswick (2018), Matheson (2022a, 2022b) and Priest (2022).

3. For excellent overviews of such work, see Buss and Westlund (2018) and Dryden (n.d.)

4. See Woolford (2015) for an excellent, if sometimes harrowing, overview of such schools and their impacts on students and indigenous communities in Canada and the United States.

5. Robert Simpson makes a similar claim: ‘more epistemic goods are likely to be realized if, at any given time, people are striking a balance […] that is responsive to their own temperamental leanings’ (Simpson 2022, 102). Of course, there can be exceptions – you might be an especially dedicated inquirer in solving a series of riddles set by a supervillain who has kidnapped a loved one … you are coerced but likely to be highly committed.

6. Note that this is not to assume that the broader non-indigenous community has a right to such knowledge and other resources; indeed, in many cases such communities (and many others) will have reason to be skeptical of ‘sharing’ their knowledge, at least in current circumstances. But the loss of such knowledge systems precludes any such future possibilities.

7. Coady also acknowledges these pressures and limitations (Coady 2002, 365–6). The importance of social factors and influences upon our epistemic autonomy will be a recurring theme in section IV.

8. Similar considerations could apply in the case of brainwashing or indoctrination. The agent’s self can be so changed over time that eventually their prior life is simply a distant memory and no longer who they are. This need not happen, but it does seem possible. This is who they have become … Compare the reverse scenario – a person raised in a cult but who escapes. Later in life they could be a completely different person – their early beliefs, commitments and attitudes could seem entirely alien and with little or no bearing on their autonomy now.

9. For related claims, see also Coady 2002, 365–9.

10. I address this sort of case, and the importance of memory and enduring beliefs for autonomy more generally, in Kawall (2010).

11. We won’t here address the relationship between beliefs and credences. For a clear overview of the relevant issues, see Jackson (2020).

12. A proponent of a narrow construal of the epistemic might hold that the scope of the epistemic does not extend to guiding inquiry. See Flores and Woodard (2023) for an overview of arguments for such a claim and for a compelling case for extending epistemic norms to evidence-gathering and inquiry.

13. It should be stressed that there are also external aids to autonomy. A teacher might provide a student with a book she suspects will spark her interest in a topic; a parent might encourage their child to ask questions about the world (and model effective inquiry when helping them to discover answers). Education as a whole plays a key role in enabling students to cultivate a virtuous epistemic autonomy. More generally, being able to draw upon the epistemic work of others allows us to achieve far more than we could have on our own and provides us with far greater options with respect to topics, methods and frameworks than if we had to discover these ourselves. We have far greater scope for virtuous epistemic autonomy when situated in strong, supportive epistemic communities.

14. The line between restrictions on sources and restrictions on subject matters is blurry. If certain texts are the only source on a given topic, restrictions on access to these texts limits both our choice of subject matter and the sources we can use. Similarly, if there’s only a small range of viable methods to study some area, in being forbidden to use such methods, we are effectively forbidden from studying this area.

15. See also Battaly (2022), 159–60.

16. Recall Friedman’s (2020) zetetic principle, ZIP, cited earlier.

17. What would determine sufficiency? We might look to the assessments of virtuous agents with phronesis – their assessments of what qualifies as an adequate engagement with important questions and valuable potential knowledge or understanding in light of an overall goal of human flourishing.

18. See, for example, Fricker (2007).

19. Much of this literature traces back to Bas Van Fraassen (2002).

20. A possible worry with embracing diverse epistemic frameworks or stances is that this will result in persistent, potentially irresolvable deep disagreements. If there is insufficient overlap in such elements as hinge propositions, fundamental epistemic principles or broader normative commitments across different frameworks or stances, it becomes unclear whether such deep disagreements could ever be rationally resolved and (more generally) how individuals and communities ought to respond to such disagreements. For more on deep disagreement see, for example, Ranalli and Lagewaard (2022), Matheson (2018) and Lynch (2013).

21. See again Grasswick (2018), particularly 204–6.

22. For helpful discussion of an early version of this paper, I am grateful to the participants in the Philosophy of Epistemic Autonomy Workshop (UNF, October 2021) and to Catin LeChapeau. I am especially grateful to Jon Matheson for his encouraging comments and advice on a more recent draft of the paper and for his remarkable patience in bringing this special issue to fruition.