Philosophers often make exotic-sounding modal claims, such as: “A timeless world is impossible”, “The laws of physics could have been different from what they are”, “There could have been an additional phenomenal colour”. Otherwise popular empiricist modal epistemologies in the contemporary literature cannot account for whatever epistemic justification we might have for making such modal claims. Those who do not, as a result of this, endorse scepticism with respect to their epistemic status typically suggest that they can be justified but have yet to develop some distinct, workable theory of how. That is, they endorse a form of non-uniformism about the epistemology of modality, according to which claims about philosophically interesting modal matters need to be justified differently from e.g. everyday or scientific modal claims, but they fail to provide any more detail. This article aims to fill this gap by outlining how such a non-uniformist view could be spelled out and what story about philosophically interesting modal justification it could contain.
The philosophy of modality concerns possibility, necessity, and related notions such as counterfactuals, dispositions, or essences. The epistemology of modality concerns the human cognitive situation with respect to modality – typically alethic (as opposed to e.g. epistemic) modality. Alethic modal matters range from the mundane and everyday, like “It is possible for this table to break” and “David Lewis could have been a prosecutor rather than a philosopher”, to the exotic that tend to interest philosophers, such as “It is possible for there to be a philosophical zombie”, “The laws of nature could have been different from what they are”, “A timeless world is impossible”, “There could be such a thing as a utility monster”, “Personal fission is not possible”. The latter bunch are examples of modal matters apparently very distant or different from what we know to actually be the case. I will refer to claims about such matters as philosophically interesting modal claims.
In contemporary modal epistemology, many philosophers hold that we make a lot of justified modal claims, and also that we have good accounts of how they are justified. But when it comes to philosophically interesting modal claims, things are different, and these good accounts do not apply. In response to this, some modal epistemologists embrace partial modal scepticism1 and claim that we cannot make justified philosophically interesting modal claims. Others, however, suggest that we have yet to develop some distinct, workable theory of their justification. That is, they are committed to some form of non-uniformism about the epistemology of modality. Non-uniformism is the view that there is more than one basic route to modal justification, i.e. it is a form of pluralism about modal justification. Self-proclaimed non-uniformists often hold that different forms of justification are tied to different kinds of modal claims. In this case, the view would be that claims about philosophically interesting modal matters need to be justified differently from e.g. everyday modal claims. While several authors have gestured in this direction, no details have so far been provided. This article is concerned with how such a non-uniformist view could be spelled out and what story about philosophically interesting modal justification it could contain.
The non-uniformist view I outline has two main characteristics. First, it appeals to a more general pluralism, namely epistemic value pluralism. Second, it subsumes the epistemology of philosophically interesting modality under a more general view of the epistemology of philosophy, by making one’s justification for such modal claims a function of one’s justification for philosophical accounts of relevant phenomena. In doing so, I will be adapting an existing account of modal justification – Fischer’s (2017) theory-based modal epistemology – and fitting it into a value pluralist framework. The suggested view will only be one of many possible views, of course, and I anticipate it will not be to everyone’s immediate liking. Others are free to offer their own preferred story, resting on other assumptions than mine, but I hope to make a convincing case that the present alternative is worthy of serious consideration.
The plan for the article is as follows. Section 2 describes the background against which philosophically interesting modal claims seem problematic. Section 3 briefly explains why questions of epistemic value are relevant to questions of justification. Section 4 outlines how pluralism about epistemic value can underwrite a form of non-uniformism about modal epistemology. Section 5 develops the axiologically motivated non-uniformism I have in mind and the epistemology of philosophy (and thereby, philosophically interesting modal claims) that goes with it. Section 6 makes some clarifications in response to anticipated objections, and section 7 concludes.
2 What Is the Problem with Philosophically Interesting Modal Claims?
At issue in this article are philosophically interesting modal claims. This term is not ideal, yet I think it captures what I have in mind better than alternatives. Let me say a few words about how I use it. Vaidya (2015) talks of “extraordinary” (as opposed to “ordinary”) modal judgements as “typically (…) about some special philosophical concept”. This is clearly in the right direction, but might be too narrow. For instance, ‘law of nature’ is certainly of philosophical interest, but is it a “special philosophical concept”? Moreover, philosophically interesting modal claims are not always directly “about” some such concept, but might state e.g. an exotic possibility that conflicts with some theory or other of a philosophically interesting concept or phenomenon. This highlights that philosophically interesting modal claims can be interesting either in themselves or in virtue of what they imply. Others (van Inwagen 1998, Roca-Royes 2017) talk of “remote” modal facts, as the contrast class of the “everyday” facts that are uncontroversially knowable in the sense that it would be strongly revisionary to deny that we often have knowledge of them. What I call “philosophically interesting” modal matters share this “remoteness” feature: they are not uncontroversially knowable. This means that some modal matters, presumably of interest to philosophers, are not going to count as philosophically interesting in my sense here, because they are not remote. For instance, it is both of interest to many epistemologists and uncontroversially knowable that it is possible for there to be a life-size papier-mâché mock-up of a barn. However, many remote – i.e. not uncontroversially knowable – modal matters will be subject to scientific rather than philosophical investigation, and they are not my focus here.
The class of philosophically interesting modal claims, as I use the term, is thus a subclass of claims about remote modal matters, namely those that typically and primarily occur in contexts of philosophical research and discussion. Finally, I recognise that the modal claims at issue are clearly more common in some philosophical fields (such as metaphysics broadly construed, including, e.g., parts of metaethics and philosophy of mind) than in others, but I don’t think it is possible to say exactly where they do and do not occur. It is hard to say much more of what, exactly, unifies the modal matters that fall in this supposedly problematic class. For the sake of this article, I assume these remarks, together with the examples in the introduction, give enough of an idea.
Why do philosophically interesting modal claims present a special problem in contemporary modal epistemology? Traditional modal epistemologies – e.g. conceivability theories (Yablo 1993, Chalmers 2002), modal intuitionism (Bealer 2002), and certain essence-based modal epistemologies (Lowe 2012, Hale 2013) – were rationalist in nature. They took modal justification to have an a priori source. They also tended to target all modal claims in more or less the same way. But under the widespread assumption (among modal epistemologists, at least) of modal realism (i.e. that modal truths are about mind-independent facts), modal rationalists appear to have a hard time explaining how these a priori methods can be reliable means of finding out about modal facts. To cut a long story2 very short, this has turned modal epistemologists of recent years to non-rationalist alternatives. But these new, non-rationalist modal epistemologies do not seem applicable to philosophically interesting modal claims. In a nutshell, the problem is that certain methods (e.g. intuition, conceivability) appear to be unreliable with respect to modal truth, and other methods (e.g. induction, perception, various scientific methods) that are otherwise thought to be reliable guides to truth nevertheless appear to be unreliable with respect to philosophically interesting modal truths.
For instance, according to empiricist modal epistemologies (Bueno and Shalkowski 2014, Strohminger 2015, Leon 2017, Roca-Royes 2017), modal justification has its source in experience and/or experiential knowledge plus certain extrapolation from that knowledge. They are attractive because they make modal justification a matter of epistemic resources – e.g. perception, induction – we generally regard as reliably delivering true beliefs. But it is highly controversial to assume that this reliability extends to philosophically interesting modal matters.
For another example, according to Williamson (2007), one is justified in claiming that p is possible if one does not arrive at a contradiction when counterfactually developing the supposition that p in imagination. This account is attractive because it makes justification of possibility claims a product of a cognitive capacity that we likely possess, namely the ability to reliably evaluate counterfactuals, and there is an evolutionary story to tell of why we should have acquired such a capacity. But first, there are doubts about whether that story covers also reliability with respect to philosophically interesting modal matters; and second, the search for contradiction must be sufficiently thorough in order to deliver justification of a possibility claim, and as Strohminger and Yli-Vakkuri (2018) note, the more different p is from the actuality that we have experience of, the more difficult it likely is for us to competently develop the supposition that p in imagination: we will presumably be more likely to fail to detect relevant contradictions. This arguably goes for our counterfactually developing philosophically interesting ps.
3 Interlude: Some Observations about Epistemic Value
To say that a belief is epistemically justified – whatever one means by that – is to say that the belief is in positive epistemic standing, that it is good, or valuable, from an epistemic point of view. Many other states and properties are also epistemically valuable: knowledge, true belief, rationality, understanding, curiosity, etc. But arguably some of these things have merely derivative epistemic value, i.e. their epistemic value can be fully explained in terms of their relation to something distinct that is of fundamental epistemic value. Something has fundamental epistemic value just in case its epistemic value cannot be fully explained in terms of the epistemic value of distinct thing(s) (Sosa 2007, Sylvan 2018).
One might hold that justification – on some understanding or other – has fundamental epistemic value, but a more common view is that justification has derivative epistemic value. That is, justification is valuable because it promotes something of fundamental epistemic value. Many epistemologists are monists with respect to fundamental epistemic value: they think there is just one fundamental epistemic good. The hands-down most popular monist position is veritism, i.e. the view that believing truly (and avoiding false belief) is the only fundamental epistemic good (Goldman 1986, David 2013, Pritchard 2014, Engel 2017, Sylvan 2018). Veritists have to account for everything that is epistemically valuable, including justification (however it is understood), in terms of its relation to the fundamental good of believing truly. Note that this is in principle compatible with allowing, with e.g. Alston (1993), that all the properties in terms of which philosophers have attempted to spell out ‘justification’ – e.g. coherence, reliability, intellectual virtue, evidence, rationality – are epistemic good-making properties.3 All that is required is that their value can be accounted for in terms of their promoting the epistemic good of believing truly. For a veritist then, a method, practice, or property can bestow justification on a state just in case the method, practice, or property in question tends to promote the fundamental epistemic good of believing truly.
The question of epistemic value has not been much discussed in modal epistemology, but it is reasonable to assume that veritism is the default view there too.4 The way in which the debate is conducted reflects that modal epistemologists tend to assume that a method, practice, or property can justify a modal belief just in case it tends to promote the acquisition of true beliefs about modal matters. As we saw in the previous section, the perceived problem with philosophically interesting modal claims is that suggested routes to modal knowledge are apparently not reliable with respect to the philosophically interesting modal matters. That is, being a reliable means to true modal belief is seen as a necessary condition on any method which is to provide modal justification. Why? Presumably because, as veritists would have it, the epistemic value of justification is taken to lie in its being a way to promote acquisition of the fundamental epistemic good, i.e. true belief.
While popular, veritism is not the only available view of epistemic value. Another option is epistemic value pluralism, according to which there is more than one fundamental epistemic good, beside that of believing truly.5 A pluralist has more resources than a veritist when it comes to explaining instances of derivative epistemic value: a state or method x may be epistemically valuable in virtue of its relation to either of the fundamental epistemic goods.6 This opens up for a pluralism about particular epistemic good-making properties too. Justification may be realized in different ways, so that a state may be epistemically justified in virtue of promoting either of a plurality of fundamental epistemic goods. Alternatively, one might talk about two (or more) different kinds of justification, each understood as the promotion of one out of several epistemic goods.7 In what follows, I will appeal to a form of epistemic value pluralism to make headway with respect to the problem of philosophically interesting modal claims. In particular, I will exploit the way in which epistemic value pluralism opens up for pluralism with respect to modal justification too.
4 Non-uniformism about the Epistemology of Modality
Some philosophers endorse scepticism about the modal matters, including the philosophically interesting ones, that lie beyond the reach of the otherwise reliable methods (van Inwagen 1998, Leon 2017, Machery 2017). Others retreat to the claim that philosophically interesting modal claims can (perhaps) be justified through a distinct method, but it remains to be spelled out how (Bueno and Shalkowski 2014, 679; Strohminger 2015, 369; Roca-Royes 2017). The latter, agnostic alternative hints towards non-uniformism, i.e. pluralism with respect to modal justification.
Typically, the idea is that different types of modal claims require different kinds of justification. How might such a view be motivated? As I have argued elsewhere (Sjölin Wirling 2020), the key to defending a non-uniformist view about the epistemology of modality is to explain why we should expect different modal epistemologies for the relevant subclasses of modal beliefs. What explanation is suitable will depend on what heterogeneity, i.e. what subclasses, among modal beliefs one is appealing to. For instance, someone wishing to defend non-uniformism with respect to modal beliefs about abstract and concrete entities respectively (e.g. Roca-Royes 2017, 2018) can claim first, that there are modal claims about both abstract and concrete entities (i.e. a certain heterogeneity), and second, that we have general reasons (independent of particular claims in modal epistemology) to think that we gain knowledge of abstract and concrete entities in different ways. Hence, we should expect that justified modal claims about entities of the respective kinds need to be accounted for by different theories. The background assumption here is that the nature of the objects of knowledge matter to how we can come by the relevant knowledge. Call that a metaphysically motivated non-uniformism. It is compatible with veritism – indeed fuelled by it, in a sense: the idea is that different methods are reliable with respect to truths about different kinds of things.
Of course, the abstract/concrete distinction won’t give us non-uniformism with respect to everyday versus philosophically interesting modal matters since there are both everyday and philosophically interesting claims about the same entities, whether abstract or concrete (e.g. “Table t can possibly break” and “Table t could possibly have been made from a different chunk of wood”). One could, in principle, motivate it metaphysically if some other epistemically relevant metaphysical difference between such claims could be found. But a non-uniformist could also appeal to some other, non-metaphysical heterogeneity among modal claims. The important thing, as far as the argument for non-uniformism is concerned, is that the heterogeneity one is pointing to is plausibly relevant to the question of what the correct account of modal justification is. It should be clear from the previous section that facts about fundamental epistemic value are relevant to what accounts of justification – including modal justification – are licensed. A monist can only allow that a method or practice is justificatory if it is appropriately related to the fundamental epistemic good of believing truly. A pluralist, on the other hand, can as already noted allow that a method or practice is justificatory also in virtue of promoting some other fundamental epistemic good, independently of whether and how it relates to believing truly.
If one accepts a plurality about fundamental epistemic goods, it is plausible that a plurality of fundamental epistemic goods might be relevant to the epistemic evaluation of modal claims too. One way in which types of modal claims could be relevantly different is thus with respect to the fundamental epistemic good in relation to which they are to be epistemically evaluated. In particular, modal claims could be heterogeneous in the sense that different epistemic goods might be relevant to the epistemic evaluation of different types of modal claims. If so, we should expect different modal epistemologies for these different types of modal claims. Call this axiologically motivated non-uniformism.
4.1 Axiologically Motivated Non-uniformism
Epistemic value pluralism can be cashed out in many different ways, and the issue is seriously underexplored in the literature.8 Here I will just state the form I have in mind as underwriting an axiological non-uniformism according to which philosophically interesting modal claims are justified in a different way from e.g. claims about everyday modal matters, because philosophically interesting modal claims are to be epistemically evaluated in relation to a different fundamental epistemic good than everyday modal claims.
First, as indicated towards the end of section 3, I understand epistemic value pluralism as disjunctive in the sense that I take each fundamental epistemic good as capable of conferring epistemic value independently. That is, if x and y are both fundamental epistemic goods, it is enough for some z to relate appropriately to either x or y in order for z to be derivatively epistemically valuable.9 Second, I will assume that believing truly is a fundamental epistemic good, and pluralism is a matter of postulating one or more additional fundamental epistemic goods. In particular, I will assume – with most others in the modal epistemology debate – that many modal claims outside of the philosophically interesting class are typically to be epistemically evaluated in relation to the fundamental good of believing truly. I will argue that philosophically interesting modal claims are not to be epistemically evaluated in relation to that fundamental epistemic good(s), but to a distinct one.
This raises the question of how it is decided which epistemic good a particular claim, belief, method, or practice is to be evaluated in relation to. On the view I am currently outlining, that depends on the context in which it is made, as different fundamental epistemic goods are salient in different contexts of inquiry. This is one way of cashing out the fact that different epistemic contexts are governed by different norms, or infused with different epistemic obligations. As e.g. Jennifer Nado (2017, 2019) has convincingly argued, professional contexts of inquiry – e.g. science, philosophy, journalism – are governed by different epistemic norms than contexts of everyday inquiry. Scientists are under the obligation to perform a lot of epistemic actions – double-blind studies, checking for biases, etc. – that ordinary epistemic agents are not. Philosophy is also permeated with more demanding epistemic norms than ordinary inquiry, although they are typically less clearly articulated. One attractive way of explaining why different epistemic norms govern different contexts of inquiry is to say, as Nado does, that the epistemic activities in the different contexts are aimed at realising different epistemic states. This need not imply that different fundamental epistemic goods are relevant to evaluating these practices – but depending on what the states are, their epistemic value may well be best explained in terms of distinct epistemic goods.
Given a plurality of epistemic goods and a plurality of contexts of inquiry, one might wonder what decides which good is salient in a given context. This is a hugely complex question, but my general view is that what epistemic good primarily governs a certain context of inquiry is partly up to us, in the sense of being determined by our interests, and how epistemic value(s) connect with other kinds of value. I’ll come back to this briefly in section 6.
Now, the basic claim of the axiologically motivated non-uniformism I am presenting is that different kinds of modal claims need to be justified in different ways because they generally occur in contexts of inquiry governed by distinct fundamental epistemic goods. In particular, while many modal claims occur in contexts primarily governed by the aim of attaining true beliefs, contexts of inquiry where philosophically interesting modal claims are typically made – contexts of philosophical inquiry, that is – are governed by the aim of attaining some other fundamental epistemic good(s). Since what promotes another good might well be different from what promotes true belief, the door is open for non-uniformism.
The way I have characterised philosophically interesting modal matters – the subclass of remote modal matters that typically and primarily are of interest in contexts of philosophical research and discussion – makes it eminently plausible that epistemic states or attitudes pertaining to them mainly occur and are epistemically evaluated in contexts of philosophical inquiry. However, they can sometimes occur in other contexts. A couple of friends can debate the metaphysical possibility of personal fission over beers in the pub, or a pensive teenager can form the belief that the laws of nature could possibly have been different. Assuming that believing truly is the salient epistemic good in those contexts, axiological non-uniformism implies that one may be justified in claiming e.g. that personal fission is possible in one context but not in another, on basis of the same method or reasons. Differently put, a modal claim may be epistemically justified in one sense but not in another.10 Here, I am only concerned with the epistemic status of philosophically interesting modal claims as they occur in contexts of philosophical inquiry, and with the sense of justification relevant in such contexts.
5 The Nature and Value of P-justification
What then, is the sense of justification relevant to philosophically interesting modal claims when they occur, as they normally do, in the context of philosophical inquiry? Basically, my view is that Bob Fischer (2017) is right when he suggests that claims about the modal propositions that our amplified experiential knowledge does not directly bear on need to be justified via our theories that bear on the relevant phenomena. That is, I will adopt the general framework of his theory-based epistemology of modality (TEM). According to TEM, one is justified in making a philosophically interesting modal claim “It is possible that p” just in case one justifiably accepts a philosophical theory T according to which it is possible that p, and one’s claim “It is possible that p” is based on T. For TEM to deliver justification for modal claims, three things need to be in place. First, it needs to be the case that theories have modal content, or at least have modal implications; second, we must have an account of theory justification; and finally, we must make sense of what it is to believe or claim something on the basis of a theory that one accepts.
Like Fischer, I take most of the interesting action here to lie with the second condition. TEM is designed to take care of remote modal claims, and it plausibly delivers justification of many scientifically interesting modal claims, because it is widely held that many scientific theories are supported by reliable scientific methods, and hence that we are justified in accepting them. However, Fischer suggests that TEM delivers modal scepticism about philosophically interesting modal claims because it is implausible that the same goes for theories that would imply the truth of philosophically interesting modal propositions.
Unsurprisingly, I do not share Fischer’s pessimism about the justificatory status of philosophical theories. Or rather, I do not share the tacit veritist assumption that underwrites the pessimism, i.e. that all epistemic justification requires truth-conduciveness. Instead, I suggest that there is a distinct type of justification relevant to the epistemic merit of philosophical theories, the epistemic value of which should not be understood in terms of the epistemic goodness of believing truly. Call this p-justification. I previously suggested that philosophically interesting modal claims also are justified in a distinct sense. What they require is p-justification, simply because they depend for their justification on the justification of philosophical theories that imply them, and those theories require p-justification. This is the sense in which my proposal, as flagged in the introduction, subsumes the epistemology of philosophically interesting modal claims under a more general epistemology of philosophy. The key claim is thus that the relevant sense of epistemic support in philosophical inquiry more generally is p-justification, which is to be understood as the promotion of a fundamental epistemic good distinct from that of believing truly.
What, then, is p-justification? I think it is clear that the epistemic practices in which philosophers engage do provide support for the views that they put forward: many philosophical theories are supported, and some are better supported than others. It is exactly this support that I intend to capture with ‘p-justification’. P-justification is provided by, more or less, philosophical business-as-usual. Philosophers go to great epistemic lengths when engaging in philosophical inquiry. There are a number of norms and standards regulating this inquiry, that they are expected, indeed epistemically obliged, to conform to. The epistemic actions that philosophers are required to and often do perform in the course of defending their views, prescribed by these norms and standards, do provide p-justification for these views and the claims that constitute them. Of course, p-justification is a matter of degree, and certainly not all philosophical inquiry is in line with these epistemic norms. But presumably we all think that the more it is in line with them, the better it is qua philosophical inquiry.
It is my view that p-justification is epistemically valuable. In the following two subsections, I will first argue that this epistemic value should not be understood as promotion of true beliefs, and then propose two different ways in which an epistemic value pluralist might account for the epistemic value of p-justification: in terms of non-factive understanding or procedural objectivity.
5.1 The Case against Truth-conduciveness
I will present three considerations that jointly present a good case for exploring the idea that believing truly is not the salient fundamental epistemic good in contexts of philosophical inquiry, in relation to which the practices, reasons, and methods that support philosophical theories are to be epistemically evaluated.
First, I said above that philosophical methodology provides p-justification for accounts of philosophically interesting matters, and so several philosophical accounts enjoy a type of justification – sometimes to a high degree. But it is also quite clear that with respect to most philosophical questions, there will not be a uniquely best p-justified account. Instead, we can often expect there to be at least two (but typically more) mutually exclusive accounts of the same phenomenon that both (or all) enjoy an equally high degree of p-justification, and philosophers will continue to disagree over which one of them is the correct account. This fact of systematic peer disagreement (Goldberg 2013) on substantive philosophical questions should be familiar to everyone working in the discipline. Apart from the fact that philosophers’ epistemic entitlement to regard their own views as true might be undercut by the presence of systematic peer disagreement, it arguably also gives us a reason to doubt that the methods philosophers use to provide p-justification are reliable guides to true belief (in domains where such disagreement is observed).
In addition to this threat from disagreement against the reliability-assumption, it also seems we lack good reasons in favour of that assumption. For instance, it has been argued that philosophers are not justified in taking explanatory virtues like simplicity, parsimony, and explanatory power, to be truth-conducive. Even if there are reasons to think that some virtues are truth-conducive in specific scientific contexts, we cannot assume without independent argument that this extends to philosophy, which is very different (Ladyman 2012, Novick 2017, Saatsi 2017). Similar critique has been directed at the use of intuitions in philosophy: while intuition may well be reliably truth-conducive in ordinary contexts of inquiry, there is little reason to think that it will remain reliable in the non-standard cases that philosophers like to consider (Machery 2017). Indeed, the criticism against intuition- and conceivability-based modal epistemologies referenced in section 2 is of a piece with this line of attack.
In response to these concerns about truth-conduciveness of philosophical methodology, Helen Beebee (2018) suggests that we reconceive of philosophy’s aim. She agrees with a number of authors (e.g. Goldberg 2009, Brennan 2010, Fumerton 2010, Kornblith 2013) who have concluded from the above misgivings that we have no reason to think that philosophical methods do or ever will afford knowledge of true answers to the questions that philosophers inquire into, such as the nature of properties, knowledge, moral responsibility, and so on. This conclusion is clearly bad news insofar as the aim of philosophical inquiry is to establish knowledge and/or justified belief about the true answer to various philosophical questions but, Beebee suggests, it is not. I agree. Concerns about reliability are worrying to a friend of veritistic justification in philosophy, but for a pluralist there is hope to vindicate p-justification as perfectly respectable when viewed as promoting some distinct fundamental epistemic good.
Second, the epistemic standards that govern philosophical inquiry are in many senses more demanding than those governing other contexts of inquiry (Daly 2017, 35–38; Nado 2017). It is not easy to say exactly what those are, but at a minimum a philosopher advancing a view is arguably under the obligation to very carefully define all central concepts and terms, she must display familiarity with the relevant literature, clearly outline the differences between hers and neighbouring views, present reasons that support her view that others will indeed recognise as such reasons, give proper consideration to counterarguments and alternative solutions, and so on. Moreover, philosophers are also expected to very carefully scrutinise other philosophers’ arguments, attempt to present counterexamples, draw out implications and underlying assumptions and scrutinise – perhaps question – them in turn, and so on. This is all part and parcel of providing p-justification. But we are under no such obligations in either everyday inquiry or – for some of these philosophy-obligations at least – in many scientific disciplines.
As Nado argues, if we want to say that ordinary folk have the knowledge we normally think they have, then fulfilling the epistemic standards of professional fields of inquiry, including philosophy, is not necessary for knowledge. This indicates that inquiry governed by these supererogatory epistemic norms aims at an epistemic state distinct from that to which everyday inquiry aspires, i.e. “knowledge (and other ‘usual suspects’ like true belief)” (2019, 133). I agree. If this is right, methods of p-justification should presumably be evaluated with respect to how well they promote that distinct aim. Although Nado is silent on what state(s) philosophy is aimed at, some of her discussion indicates states that include/imply knowledge but also go beyond it, such as knowing that one knows. However, the fact that philosophical inquiry is governed by epistemic norms that are more demanding than the norms that govern everyday inquiry is also compatible with the idea that philosophy aims at something other than knowledge – something which might well be very demanding, although it does not imply or contain knowledge or true belief. Actually, this latter option is quite attractive if we combine the observations about demandingness with the reliability-considerations that drive Beebee to suggest a reconception of philosophy’s aim. On the one hand, philosophical inquiry is governed by epistemic norms that are demanding – and arguably much actual philosophical inquiry is in line with them. On the other hand, we have reason to doubt that actual philosophical inquiry – despite being in line with demanding norms – is sufficiently truth-indicative to count as justificatory in the sense relevant to knowledge.
The third point I want to raise concerns the nature of the systematic peer disagreements in philosophy. In paradigmatic disputes over facts, the function of arguing – of giving reasons that are supposed to weigh with the other party, or the audience – is to settle which side has the true description of the fact in question. We can expect many paradigmatic factual disputes to be settled at some point, as the world tends to “push back” against false beliefs. Of course, some factual disputes are irresolvable. For instance, distinguished archaeologists Sharon and Shapur might disagree over whether Neanderthals buried their dead, but there is no archaeological evidence available that would settle the debate and no reason to think that such evidence is forthcoming. However, Sharon and Shapur – and plausibly parties in most paradigmatic factual disputes – have no problem agreeing on what kind of finding would in principle settle the issue and that they would adjust their beliefs in response to such a finding.
Philosophical disputes are similar to paradigmatic factual disputes in the sense that philosophers offer reasons that are supposed to weigh with others as reasons supporting their views. But we have little or no reason to expect that philosophical disagreements can be settled. First, there are typically no empirically accessible facts in light of which false philosophical claims can be corrected. Second, I doubt that philosophers in general can agree on what even in principle would constitute a conclusive reason to settle the issues they are debating. David Lewis (1983, x) remarks that
[t]he reader in search of knock-down arguments in favor of my theories will go away disappointed. Whether or not it would be nice to knock disagreeing philosophers down by sheer force of argument, it cannot be done. Philosophical theories are never refuted conclusively.
I think this is right, and even more clearly, philosophical theories are never conclusively shown to be true. I do not think the explanation of why it is right is that the reasons and arguments philosophers present are bad, or too weak. Rather I think it is because settling conclusively which side of a dispute has the truth is not the function of reason-giving in philosophy; “reason-giving in philosophy” is meant here as a central part of the complex set of practices that has the potential to generate p-justification.
In fact, I think philosophical disputes have a lot in common with how Elgin (2017) describes a dispute in aesthetics over how Cezanne’s Le Compotier is best interpreted. As in philosophy, reason-giving and argument is central: not just any interpretation is viable; the disputants offer considerations that bear on the work in question, and are supposed to weigh with others, in support of their respective preferred interpretation. But we rightly don’t expect a dispute like this to be resolvable – the notion of a conclusive reason to settle on one interpretation over all others does not make sense. Yet, that is not a problem. Irresolvable disagreement is not an obstacle to successful epistemic activity in cases like this. This suggests, according to Elgin, that the point of reason-giving simply is not to convince one’s opponent or audience that one view is true and the other false – it must have another function. I would add that if it has another function, then it should be epistemically evaluated primarily in relation to that function. This function of reason-giving, as not meant to establish knowledge of a truth, is not limited to disputes over non-factual matters (as we may well think that there is no fact of the matter as to e.g. what interpretation of Le Compotier is the correct one). As Elgin notes, the function that reason-giving plays in aesthetics can also be relevant to factual disputes – especially where we have little or no reason to expect that the issue will be resolved, and this includes philosophical matters.
5.2 Two Alternatives: Understanding and Procedural Objectivity
I have argued that we should not think of p-justification as the promotion of true belief (or any state which necessarily includes it). But what epistemic good does p-justification plausibly promote? I will presently outline two promising alternatives available to the axiological non-uniformist.
One option is to postulate that besides believing truly, there is another type of epistemic state that it is fundamentally epistemically valuable to be in, and this state is non-factive in the sense that it does not entail true belief. A promising candidate state here is non-factive understanding. The relevant type of understanding is not concerned with individual propositions but objectual (Kvanvig 2003), i.e. it is understanding of a subject matter (e.g. causation, moral responsibility, the Comanche dominance of the southern plains of North America between the 17th and 19th century). Roughly, one has understanding insofar as one grasps a comprehensive set of interrelated propositions about this subject matter and how they relate to each other. Kvanvig thinks that understanding is factive in the sense that (at least) all the “central” propositions in the set must be true, but others (e.g. Elgin 2007, Riggs 2009, Potochnik 2020) reject factivity. For current purposes, let us grant that some types of understanding are factive, and that the epistemic value of such understanding may well be explainable in veritistic terms (Ahlström-Vij 2013, 31–34). This is compatible with there being non-factive understanding too, which is fundamentally epistemically valuable (alongside e.g. true belief).
In order to explain the epistemic value of p-justification in terms of non-factive understanding, practices of p-justification should typically promote non-factive understanding. Is that plausible? The idea that philosophy’s epistemic success consists in some kind of non-factive understanding is not new (see e.g., Graham 2017; Shand 2017, 292; McSweeney forthcoming), although it is not always clear just what non-factive understanding is, beyond the very rough characterisation above. Here is one reason to be optimistic, however. When Beebee (2018) suggests we reconceive of philosophy’s aim, she puts forward a position she calls equilibrism about philosophy, which draws on some famous methodological remarks of David Lewis’s. According to equilibrism, the aim of philosophical inquiry is to “find out what equilibria there are that can withstand examination” (Lewis 1983, x; Beebee 2018, 16). An equilibrium is an internally cohesive set of philosophical claims and assumptions, both first-order and methodological (e.g. about how Occam’s razor is to be applied, or how important common sense is to preserve). With respect to most philosophically interesting phenomena, there will be more than one equilibrium that can withstand examination.
As we have seen, there is little reason to think that p-justification can ever adjudicate between such equilibria positions. But there is quite good reason to think that p-justification is an efficient means to the charting of equilibria with respect to philosophically interesting questions. Indeed, the co-existence of equally well-supported accounts of the same phenomenon is the very predicament that leads many veritists to despair about the epistemic value of p-justification. However, if there is relevant epistemic value in the situation of multiple equally well-supported philosophical accounts of the same phenomenon, that is not a problem. Beebee doesn’t discuss what the value of equilibria-charting might be, but recall Elgin’s discussion of aesthetics, where reason-giving and disagreement co-exist. Elgin argues that the function of reason-giving there is to increase (non-factive) understanding: publicly available and assessable reasons highlight features of the object of inquiry and make a case for their importance and relevance. The continuous and irresolvable “competition” between multiple accounts contributes to our understanding of the object: it is enhanced when we appreciate the force of, and interconnections between, the arguments, reasons, and assumptions that make up a perspective or an interpretation. Again, this function is not limited to disputes where there is no fact of the matter (Elgin 2017, 181):
the highlighting of relevant features (…) is why we find advancement in understanding in fields like palaeontology where in principle resolvable disagreements may never be resolved. We understand more about the Neanderthals and about our understanding of them when we appreciate the force of arguments for and against the claim that they buried their dead.
If this is right, then the co-existence of multiple philosophical accounts in equilibrium about causation, or moral responsibility, may well be a state of non-factive understanding. Insofar as non-factive understanding is fundamentally epistemically valuable, p-justification can derive its epistemic value from its ability to promote that type of state.
I want to briefly mention one other option for the axiological non-uniformist here, which takes a somewhat different approach to how epistemic value can be transferred. So far, we have thought of derivative epistemic value in instrumentalist terms: p-justification has (or fails to have) derivative epistemic value because it is (or fails to be) instrumental to acquiring products with fundamental epistemic value. Proceduralism about value takes things to be the other way around. Products (e.g. epistemic states) may have derivative epistemic value in virtue of being the result of a procedure or practice which is (or instantiates properties that are) fundamentally epistemically valuable. As Goldman (1999, 75) puts it, a proceduralist epistemology “fasten[s] on the intrinsic merits of intellectual practices to judge their epistemic worth or propriety”. A proceduralist explanation of the epistemic value of p-justification would hold that the philosophical procedures and practices that tend to produce p-justified accounts are themselves epistemically valuable.
According to Helen Longino’s (2002) well-known proceduralist social epistemology of science, the epistemic value of scientific inquiry lies in the objectivity of the procedures that constitute it. An objective procedure is one that fulfils the following four conditions: (1) “publicly recognised forums for the criticism of evidence, methods, and of assumptions and reasoning”; (2) mutual “uptake of criticism” in the sense that criticism must be responded to and taken seriously, but critics must also take account of responses; (3) “publicly recognised standards by reference to which theories, hypotheses and […] practices are evaluated and by appeal to which criticism is made relevant”; (4) a “tempered equality of intellectual authority” meant to guard against letting e.g. the social status of some epistemic subjects influence assessment of reasons and accounts (2002, 128–135).11 Something counts as scientific knowledge, on Longino’s picture, insofar as it is the product of a procedure characterised by something like these contraints. There are no further constraints on the outcome.
Whether or not this is an attractive picture of the epistemic value of scientific inquiry, there is some promise to understanding the value of philosophical inquiry in terms of it. As Peter (2009, 124) notes,
[w]hat Longino puts at the center of her epistemology is the demand that […] claims be scrutinized from a variety of perspectives, and in particular that it is possible to subject the background assumptions embedded in scientific practices that support these claims to critical examination of their metaphysical, empirical, and normative implications.
Longino’s procedural objectivity is meant to embody conditions where this is the case. I’m not wedded to the idea that the epistemic value of procedures that ensure continuous, meticulous scrutiny of any claim or assumption, including very basic or entrenched ones, examination of methodological principles, and draws out implications of all kinds, needs to be understood in terms of objectivity (nor that this is the best way to understand ‘objectivity’). However, I think there is some plausibility to the idea that inquiries that realize (1)–(4) instantiate something very central to epistemic life, a type of intellectual conscientiousness perhaps, that is valuable quite independently of whether it, in a given context, tends to lead to some particular epistemic state with independent epistemic value (e.g. true belief). Of course, if the epistemic value of p-justification is to be explained in terms of the value of objective procedures (or whatever we might want to call them), practices of p-justification must fulfil the conditions for procedural objectivity.
Here too there are reasons for optimism. First, if we look at the state of philosophy, it certainly seems that philosophical inquiry invites the scrutiny of any claim, including deeply entrenched background assumptions, from a variety of perspectives. As we know, little if anything is ever agreed upon as settled once and for all. This is witnessed by the fact that philosophical issues are perennial: the same questions and approaches come up, or are questioned, again and again, even when we for some time might have thought that we had settled the fate of that view or approach once and for all. I think there is a good case for thinking that philosophical inquiry, when done well, does go some way towards realizing something like (1)–(4). On the other hand, there is certainly a lot of room for improvement. For instance, Wilson (2013) argues that bias and sociological determinants impede philosophical debate, which would seem to violate the fourth criterion of procedural objectivity.
But even if philosophical inquiry is not always or fully procedurally objective, the procedural claim that philosophical inquiry is epistemically valuable and thereby able to provide genuine p-justification insofar as it is procedurally objective might be a fitting way for us to understand epistemic value here. First, we can certainly take measures to improve epistemic activities in philosophy with respect to procedural objectivity, while it is highly unclear how to do that with respect to promotion of true belief. Second, philosophical disagreements are similar in nature to other disagreements where proceduralism is considered suitable. On Longino’s account, objectivity is a bulwark against partiality, illegitimate exclusion and silencing of certain perspectives, preferences or experiences. The procedural account of objectivity is meant to recognise the possibility of deep disagreement between participants in an epistemic environment, which makes it implausible that all sides can agree on an independent condition for the correctness or acceptability of the outcomes of the procedure. Hence, it is better to define acceptability in terms of the conditions of the procedure. This ability to accommodate deep disagreement (on e.g. values, or what constitutes the “common good” even when there is agreement on the facts) has made pure proceduralism popular among political philosophers defining democratic legitimacy (Peter 2008, 2009). Deep disagreement is a fact of philosophy too, as memorably captured in the hypothetical debate between Argle and Bargle on whether or not holes are material objects (Lewis and Lewis 1970). Argle and Bargle finally recognise that they agree on what the costs and benefits of each view are, but disagree on how to weigh those together, i.e. on how important the respective costs and benefits are. This, of course, is the situation philosophers find themselves in when there are multiple equally p-justified accounts of the same phenomenon.
5.3 Summary: The Epistemology of Philosophically Interesting Modal Claims
According to TEM, one is justified in making the philosophically interesting modal claim p just in case one is justified in accepting a philosophical theory that contains or implies p. One is justified in accepting a philosophical theory insofar as it is p-justified, that is, has undergone the examination that is the bread and butter of academic philosophy as we know it (and this is obviously a matter of degree).
This is combined with an epistemic value pluralism, according to which there are other fundamental epistemic goods than true belief; and which epistemic good an epistemic practice or state is to be evaluated with respect to depends on the context of inquiry, because different epistemic goods are salient in different contexts of inquiry. Thus, the optimism about p-justification is underpinned by the postulation of a fundamental epistemic good that p-justification plausibly promotes and the claim that this epistemic good is plausibly salient in the context of philosophical inquiry. In order to fully defend this position, either of the approaches sketched in the previous subsection would be interesting to explore. That is, either p-justification is derivatively epistemically valuable because it tends to promote non-factive understanding, which, in addition to true belief, is fundamentally epistemically valuable. Or the practices of p-justification are epistemically valuable in themselves because they are procedurally objective, and the results of such a process (e.g. a plurality of well-supported philosophical theories of the same phenomenon) have derivative epistemic value.
This is a form of axiological non-uniformism about modal epistemology, because it also holds that modal claims outside the class of philosophically interesting ones need other types of modal justification. For instance, everyday modal claims might call for truth-conducive justification based on relevant experiential knowledge, and interesting modal claims in contexts of scientific inquiry might call for truth-conducive justification of the theories that imply or contain them, in line with Fischer’s original version of TEM. This would then be because in everyday and scientific inquiry, believing truly is the salient fundamental epistemic good.
To be clear, I do not pretend to have argued for epistemic value pluralism here – just as those who generally assume veritistic monism do not tend to provide arguments for their position. I think there are independent arguments in light of which pluralism is plausible, although there is no space for them here. Neither do I pretend to have argued independently for the claim that non-factive understanding, or procedural objectivity, are fundamental epistemic goods, although I think both of the discussed candidates are well worth exploring in such a role. My main concern was, in any case, with presenting a coherent and interesting way to make sense of the non-uniformist position according to which philosophically interesting modal claims need to be justified in a different way than e.g. everyday (and perhaps scientific) modal claims.
6 Clarificatory Remarks
In this final section of substance, I will make some clarificatory remarks that I hope will diffuse some potential worries about the account proposed above.
Since there can often be more than one p-justified theory, two philosophers who accept different p-justified theories can be equally justified in making or accepting opposing modal claims. As a toy example, a dualist might be justified in claiming that philosophical zombies are possible, and a physicalist might be justified in claiming that philosophical zombies are not possible. I do not think there is anything odd about this consequence in itself. In fact, similar scenarios can occur even when justification for p must indicate the truth of p, as long as we allow the very plausible assumption that p’s being justified does not guarantee that p is true. None of this implies that philosophically interesting modal claims are not truth-apt, or that their truth is relative in some controversial sense – it is perfectly compatible with there being an objective fact of the matter as to which claim or theory is true.
However, we might have to re-think some of our practices where we rely on modal claims in philosophical argumentation, or at least of what such arguments can accomplish. For instance, Chalmers (1996) intends the claim that zombies are possible to be an independent reason to think that physicalism is false. But if the only support for that claim comes from its being part of or implied by a theory one accepts, e.g. dualism, it will hardly have traction with someone who rejects dualism in favour of physicalism. This will certainly be less than what some have hoped for. One objection might be that the reason claims like the one about zombies are philosophically interesting is that they can play this role of independently justified premises in modal arguments, i.e. that they are not interesting “in themselves”. An epistemology of philosophically interesting modal claims like the current one thus undermines the interestingness. It might be true of some possibility claims (about e.g. zombies, or utility monsters) that they are philosophically interesting only insofar as they can help us decide whether we should be utilitarians or physicalists about mind. But many modal claims are philosophically interesting in themselves (e.g. timeless world, different laws of nature). Moreover, let me stress that the idea of philosophical arguments as ultimately meant to convince one’s opponent and/or an impartial audience that one’s theory is true is in need of revision anyway, given the proposed reconception of epistemic value sketched above. That the same is true for arguments with modal premises is thus unsurprising.
Another worry is that there is really no difference between partial modal scepticism and the epistemology of philosophically interesting modal claims offered here. On both views, we lack truth-conducive reasons to believe one way or the other with respect to philosophically interesting modal matters. However, I have argued for a number of other claims about our epistemic situation. In particular I have suggested that we can have epistemic support for making claims about philosophically interesting modal matters, and I have detailed how that might work. It is true that all of this is compatible with what the sceptic says. But the suggested view isn’t scepticism, because it proceeds from an assumption of epistemic value pluralism: p-justification is a form of epistemic support because it promotes a fundamental epistemic good, although a distinct one from true belief. Of course, pluralism may be rejected – I have not had the space to argue systematically for it here. Veritists too often assume rather than argue for their view. Of course, assuming veritism, my account would be a form of scepticism. But even then the above is an improvement on the “bare” partial sceptical thesis prominent in the present literature, in the sense that it offers an analysis of philosophical practice and its value (even if, given veritism, this value wouldn’t be epistemic).
Finally, I anticipate that it will be objected that many philosophers do not recognize themselves in this picture: they aim to find the truth, and they intend philosophical methodology to be truth-conducive, and in fact they think it often is. Let me stress a couple of points that will make this insistence less worrying for my proposal.
To begin, it is not part of the current proposal – whether filled out in terms of proceduralism or non-factive understanding – that anything goes and philosophical inquiry is completely free-floating of known facts about the world. There is a whole lot of things known about the world that one needs to take into account in the process of p-justification, just as there are facts about Le Compotier that serious interpreters must take into account.
Next, the view does not require that philosophical methods are not in fact truth-conducive with respect to philosophically interesting truths. Maybe some of them sometimes are. The problem is that we cannot tell, and therefore this is not a very useful standard for evaluating epistemic practices in philosophy. So while we certainly can evaluate philosophically interesting modal claims, or philosophical theories, for whether they promote true belief – indeed this is what partial modal sceptics, or sceptics about philosophy, have done – this perspective is unduly limiting if epistemic value pluralism is viable. From a veritist perspective, we cannot say why our methods seem valuable, why they seem to provide support, why we keep doing what we are doing despite being aware of all the shortcomings, and most importantly it is hard to see how we could improve. That is the main point of the proposal: there are other epistemic goods in relation to which we are in a much better position to evaluate – and also improve – philosophical practices, and there is a good case to be made for them being salient in the context of philosophical inquiry. Moreover, it is not my view that some contexts of inquiry are essentially and forever governed by one fundamental epistemic good, and that everyone who engages in the relevant inquiry immediately picks up on and accepts it. It is therefore natural that there is disagreement over this, and the current piece is a petition in such a debate.
Relatedly, I think there is a useful distinction to be made between the individual philosopher’s beliefs and personal aims – which may involve settling on a true belief – and the epistemic states, aims, and values of philosophy as a discipline, where I have focused on the latter (see also Nado 2017, 12–13; Beebee 2018, 11). Although epistemology has often focused on beliefs, other states too can be epistemically evaluated. Nevertheless, one might worry that if the epistemic support afforded by p-justification is not primarily understood as providing reason to think that a theory or claim is true, it might seem irrational for one to believe on the basis of p-justification (whether or not that is one’s aim). So what individual attitude does it licence? One option is to go with a watered-down belief substitute that requires a lesser degree of e.g. confidence, as has been suggested in the literature on philosophical disagreement (Goldberg 2013, Barnett 2019), but since that route is tied to the veritist assumption rejected above, I consider it more promising to think of it in terms of what van Fraasen calls acceptance in the context of scientific theories (see also Elgin 2017, Beebee 2018, Rosen 2020). Acceptance is voluntary, and involves taking on certain commitments – for instance to employ certain claims as premises in reasoning, to confront new relevant phenomena using the resources of that theory, to answer questions about and defend it in the face of criticism, and assume “the role of the explainer” (Van Fraassen 1980, 13).
I have offered a way to fill out the non-uniformist idea that philosophically interesting modal claims need to be justified in a different way from e.g. everyday modal claims. I cashed this out as an axiologically motivated non-uniformism, according to which different modal claims need to be justified differently because they occur in different contexts of inquiry governed by different fundamental epistemic goods. This is underwritten by an assumption of epistemic value pluralism, i.e. the view that there is a plurality of fundamental epistemic goods. In particular, philosophically interesting modal claims primarily occur in the course of philosophical inquiry, which is – unlike e.g. everyday inquiry – governed by a fundamental epistemic good distinct from that of believing truly. In line with Fischer’s (2017) TEM, I proposed that one is justified in making or accepting a philosophically interesting modal claim c when c is part of or implied by a justified philosophical theory that one accepts. However, unlike Fischer, I suggested that philosophical theories often are – or at least can be – justified in the sense relevant to philosophy, and that this sense is best understood in terms of either promotion of non-factive understanding or as resulting from a procedurally objective inquiry.
Many thanks to Frans Svensson and Helen Beebee for their careful reading and helpful comments on this article, and to Anna-Sofia Maurin and countless others for discussing the ideas in here with me over the years. This research was generously funded by the Swedish Research Council, grant no. 2019-00635.
Bealer, George 2002. “Modal Epistemology and the Rationalist Renaissance.” In: Conceivability and Possibility, edited by Tamar Szabo Gendler& John Hawthorne, Oxford: Clarendon Press, 71–125.
Beebee, Helen 2018. “Philosophical Scepticism and the Aims of Philosophy.” Proceedings of the Aristotelian Society 118 (1), 1–24.
Bueno, Otavio & Shalkowski, Scott A. 2014. “Modalism and Theoretical Virtues: Toward an Epistemology of Modality.” Philosophical Studies 172 (3), 671–689.
Chalmers, David 2002. “Does Conceivability Entail Possibility?” In: Conceivability and Possibility, edited by Tamar Sazbo Gendler & John Hawthorne, Oxford: Clarendon Press, 145–200.
David, Marian 2013. “Truth as the Primary Epistemic Goal: A Working Hypothesis.” In Contemporary Debates in Epistemology (Second Edition), edited by Matthias Steup, John Turri & Ernest Sosa, Chichester: Wiley-Blackwell, 363–377.
Engel, Pascal 2017. “A Plea For Epistemic Monism.” In: Epistemic Pluralism, edited by Annalisa Coliva & Nikolaj J. L. L. Pedersen, Cham: Palgrave Macmillan, 93–125.
Fumerton, Richard 2010. “You Can’t Trust a Philosopher.” In: Disagreement, edited by Richard Feldman & Ted Warfield, New York: Oxford University Press, 91–110.
Goldberg, Sanford C. 2013. “Defending Philosophy in the Face of Systematic Disagreement.” In: Disagreement and Skepticism, edited by Diego E. Machuca, New York: Routledge, 277–294.
Graham, Gordon 2017. “Philosophy, Knowledge, and Understanding.” In: Making Sense of the World: New Essays on the Philosophy of Understanding, edited by Stephen Grimm, New York: Oxford University Press, 98–115.
Kornblith, Hilary 2013. “Is Philosophical Knowledge Possible?” In: Disagreement and Skepticism, edited by Diego E. Machuca, New York: Routledge, 260–276.
Leon, Felipe 2017. “From Modal Skepticism to Modal Empiricism.” In: Modal Epistemology After Rationalism, edited by Bob Fischer & Felipe Leon, Cham: Springer International, 247–261.
McSweeney, Michaela M. (forthcoming). “Metaphysics as Essentially Imaginative and Aiming at Understanding.” American Philosophical Quarterly.
Nado, Jennifer 2019. “Who Wants to Know?” In: Oxford Studies in Epistemology, edited by Tamar Szabo Gendler & John Hawthorne, Oxford: Clarendon Press, 114–136.
Pedersen, Nikolaj J. L. L. 2017. “Pure Epistemic Pluralism.” In: Epistemic Pluralism, edited by Annalisa Coliva & Nikolaj J. L. L. Pedersen, Cham: Palgrave Macmillan, 47–92.
Pedersen, Nikolaj J. L. L. 2020. “Pluralist Consequentialist Anti-scepticism.” In: Epistemic Entitlement, edited by Peter J. Graham & Nikolaj J. L. L. Pedersen, Oxford: Oxford University Press, 297–326.
Pritchard, Duncan 2014. “Truth as the Fundamental Epistemic Good.” In: The Ethics of Belief, edited by Jonathan Matheson & Rico Vitz, Oxford: Oxford University Press, 112–129.
Riggs, Wayne D. 2009. “Understanding, Knowledge, and the Meno Requirement.” In: Epistemic Value, edited by Adrian Haddock, Alan Millar & Duncan Pritchard, New York: Oxford University Press, 331–338.
Roca-Royes, Sònia 2017. “Similarity and Possibility: An Epistemology of de re Possibility for Concrete Entities.” In: Modal Epistemology After Rationalism, edited by Bob Fischer & Felipe Leon, Cham: Springer International, 221–245.
Roca-Royes, Sònia 2018. “Rethinking the Epistemology of Modality for Abstracta.” In: Being Necessary: Themes of Ontology and Modality from the Work of Bob Hale, edited by Ivette Fred & Jessica Leech, Oxford: Oxford University Press, 245–265.
Rosen, Gideon 2020. “Metaphysics as a Fiction.” In: Fictionalism in Philosophy, edited by Bradley Armour-Garb & Frederick Kroon, Oxford: Oxford University Press, 28–47.
Saatsi, Juha 2017. “Explanation and Explanationism in Science and Metaphysics.” In: Metaphysics and the Philosophy of Science: New Essays, edited by Matthew Slater & Zanja Yudell, Oxford: Oxford University Press, 162–191.
Sjölin Wirling, Ylwa 2020. “Non-uniformism about the Epistemology of Modality: Strong and Weak.” Analytic Philosophy 61 (2), 152–173.
Sjölin Wirling, Ylwa 2021. “An Integrative Design? How Liberalised Modal Empiricism Fails the Integration Challenge.” Synthese 198 (6), 5655–5673.
Strohminger, Margot & Yli-Vakkuri, Juhani 2018. “Moderate Modal Skepticism.” In: Knowledge, Belief, and God: New Insights in Religious Epistemology, edited by Matthew A. Benton, John Hawthorne & Dani Rabinowitz, Oxford: Oxford University Press, 302–321.
Vaidya, Anand J. 2015. “The Epistemology of Modality.” In: The Stanford Encyclopedia of Philosophy (Summer 2015 ed.), edited by E.N. Zalta, <http://plato.stanford.edu/archives/sum2015/entries/modality-epistemology/>.
Vaidya, Anand J. & Wallner, Michael 2018. “The Epistemology of Modality and the Problem of Modal Epistemic Friction.” Synthese 198 (suppl. 8), 1909–1935.
Wilson, Jessica M. 2013. “Three Dogmas of Metaphysical Methodology.” In: Philosophical Methodology: The Armchair or the Laboratory?, edited by Matthew Haug, New York: Routledge, 145–165.
I focus on justification (broadly construed) rather than knowledge. So, with ‘scepticism’ I intend the view that we do not, and (given our current cognitive machinery) are unlikely to ever, make justified claims. Scepticism with respect to subject matter s is thus the view that we do not, and are unlikely to ever, make any justified claims about s.
For criticism of rationalist modal epistemologies, see e.g. Roca-Royes 2010 and 2011, Vaidya and Wallner 2018, Jenkins 2008 (chapter 2.5), Mizrahi and Morrow 2015, Horvath 2014, Leech 2011, Worley 2003. For an argument that puts pressure on the idea that empiricist alternatives do obviously better than rationalism on some matters, see Sjölin Wirling 2021.
For arguments to the effect that Alston and several other self-proclaimed pluralists are veritistic monists on the level of fundamental epistemic value, see Pedersen 2017 and Peels 2010.
The only explicit discussion of epistemic value in modal epistemology that I am aware of is due to Lam (2017). Lam also identifies veritism as the default background assumption behind much criticism of imagination-based modal epistemology, and then, interestingly, goes on to question the veritist assumption, just as I will. This thread has not, to the best of my knowledge, been taken up elsewhere.
One may be a pluralist and hold that believing truly is not on the list of fundamental epistemic goods. I will ignore that option here.
Note that this disjunction is not exclusive: if there are, say, two final epistemic goods v and w, a given state or practice can be epistemically valuable both in virtue of contributing to v and in virtue of contributing to w, even if contributing to only one of v and w would be sufficient for being epistemically valuable.
Pluralism about justification does not follow from epistemic value pluralism; one may hold that justification is promotion of one particular fundamental epistemic good (e.g. believing truly), even if there are several.
Exceptions include Pedersen (2017, 2020) and Axtell and Olson (2009).
One can also imagine a pluralism according to which a state or practice must contribute to all, or at least more than one, of the fundamental epistemic goods, in order to have epistemic value. I ignore that option in what follows.
Of course, a modal claim can also be justified in more than one sense, for instance if the method supporting it promotes more than one fundamental epistemic good.
For another procedural account of objectivity see Elgin 2017, ch. 7.