Chris Ranalli (VU Amsterdam), "Is Radical Doubt Morally Wrong?"

Erkenntnis, 2024

By Chris Ranalli

In contemporary epistemology, pragmatic responses to skepticism have been consistently unfashionable. The key thought is that they’re not wrong but irrelevant. The problem of radical skepticism, properly understood, is that there are compelling reasons to think that, while we are justified in believing everyday things like ‘my coffee is getting cold’, we aren’t justified in denying radical skeptical hypotheses, such as that we’re brains-in-vats, but that if we were justified in believing everyday things, we would be justified in denying radical skeptical hypotheses as well. Each claim enjoys intuitive support, but they are mutually inconsistent. We have to give up something we initially found compelling.

Put like this, it’s easy to see why pragmatic responses look irrelevant. Suppose I am pragmatically justified in believing certain everyday things. The radical skeptic, who resolves the skeptical paradox above by denying that that we are justified in believing everyday things, will say that he doesn’t (or doesn’t need to) deny that I have pragmatic reason to believe what I do. His claim is that I’m not justified in believing what I do in a way that matters for having knowledge. I lack epistemic justification to believe what I do. Thus, the point about pragmatic justification just seems irrelevant, since his point is that I lack epistemic justification. What do pragmatic reasons got to do with it?

This strikes me as a mistake. Or, at least, I think there is an interesting dialectic here that gets overlooked. There’s a good reason for why it gets overlooked, however. The heyday of ordinary language philosophy relegated radical skepticism to the dustbin of bad ideas. Radical skepticism poses no serious problem, the thought went, because the argument for radical skepticism rests on pulling language out of its natural home. Once we attend to how we actually use terms like ‘knowledge’, ‘justification’, and ‘reasonable’ in ordinary life, there’s no question about whether they’re ever satisfied.

This orthodoxy was challenged in the 1980’s by the ‘new’ skeptics. Peter Unger tried to meet ordinary language philosophers on their own terms (there’s a kind of pun here, but I swear it is accidental!) by arguing that ‘knowledge’, like ‘flat’, is an absolute term. Practically speaking, it’s OK to use it as we do, but strictly speaking when say something like “the driveway is flat”, we speak falsely. Similarly with “I know that my coffee is getting cold” and all the rest.

Barry Stroud and Thomas Nagel were more and less radical than Unger. Unlike Unger, they didn’t argue outright for radical skepticism, but they tried to show that it is philosophically significant, representing a serious problem arising out of everyday epistemic practices and our natural intellectual limitations. Stroud represented the problem as a deeply human paradox. For him, any solution involves giving up “platitudes we all accept”, which he thought was not much better than embracing skepticism. It was a lose-lose game. This way of thinking about skepticism went on to influence the way a number of household names in contemporary epistemology approach the problem of radical skepticism to this day.

As a kind of armchair sociology, my suspicion is that it looks to many of them like a major philosophical setback to return to thinking about radical skepticism pragmatically, as a failure to engage with the problem as it ‘meant to be understood’, underappreciating the advances made since the 1980’s.

I understand that impulse, but I think there’s much to be gained by thinking about radical skepticism pragmatically. Or, as I argue in my recent Is Radical Skepticism Morally Wrong?, there’s much to be gained by thinking about radical skepticism ethically, through the lens of moral philosophy and the ethics of belief. One reason is that, with so much excellent work in ethics, meta-ethics, and epistemology since then—especially with the ‘value turn’ in epistemology—it would be a missed opportunity not to look at the pragmatic costs of skepticism again, armed with fresh insights from these areas. I’m thinking here of the work on normative reasons and oughts, moral encroachment, doxastic wronging, and epistemic value. The same can be said for the ‘social turn’ in epistemology. This is what motivated me to undertake the project.

The basic ideas occurred to me when I was thinking about certain connections between skepticism and conspiracy theory. I was preparing an introductory lecture for high schoolers interested in studying philosophy. I thought it would be interesting to see how conspiracy theorists sometimes use skeptical hypotheses (even quite radical ones) to initiate doubt. They want people to weaken their beliefs, to no longer believe that mainstream sources are trustworthy. There is a kind of norm here, that one should believe what the mainstream says only if one has eliminated the possibility that there’s no conspiracy to coverup the truth. Otherwise, one should not believe what the mainstream says. As with radical skepticism, it’s hard to meet that demand, because to counter claims like “9/11 was an inside job”, at some point we need to rely on evidence that qualifies as ‘mainstream’ by the conspiracy theorist’s lights, and so begs-the-question.

What’s interesting here is that the motivations for doubt are themselves practical. When conspiracy theories are deployed by alternative media personalities (and politicians) to initiate doubts about certain mainstream views or facts, the goal is to instigate doubt in the service of some practical end, like driving up internet traffic, promoting certain policies, or securing personal financial benefits. ‘Doubt is the product’. As we know, this sort of doubt-mongering has bad real-world effects.

Similar things can be said about online trolls and gas-lighters. Their goal is to get us to doubt, but the state of doubt, and the transition from belief to lacking the kind of commitment belief embodies, can sometimes be harmful, hurtful, or even wrong. I think that radical skepticism has similar costs and that, to the extent that we complied with norms that radical skeptics are, I think, committed to, we’d be doing something bad to ourselves and something wrongful to others. This is the main thesis of the paper. One motivation for it is the parity between the harms and wrongs done to people when doubt is weaponized ‘in the wild’ and when it is the upshot of radical skepticism.

The first step in the argument was to show that this upshot exists. Skeptics say that we’re not justified in believing everyday things. From the fact that one is unjustified in believing that P, it’s natural to think that one ought not believe that P. Taken together, skepticism seems to imply that we ought not believe the everyday things that we believe. However, this follows only if the relevant ‘ought’ is indexed to norms that have sufficient normative authority over our thinking. After all, from the fact that, by the rules one enforces in D&D as dungeon master, you aren’t justified in thinking that New York is a U.S. city, it doesn’t follow that you (really) ought not believe that New York is a U.S. city. So, ideally, I’d need to defend the view that the normative relationship between the skeptic’s justification claim and the doxastic ought is different from that, that it has authority over our thinking.

A defense of that claim is a paper in its own right. My paper is already very long (15,000 words!), but the more basic idea forms part of a follow up project I’m currently working on. So, my strategy in the paper was to go about this indirectly, to show that if the skeptic denied the normative oomph of their justification claim, it would render the problem of radical skepticism unworrying. Yes, skepticism would formally remain a paradox, but it would be more like a stubborn logic puzzle, something that doesn’t raise serious normative or existential concerns. It would not be like the problems of freewill or nihilism.

Moreover, it would be unclear what its philosophical significance is. If the skeptic’s claim does not even purport to be one that bears on what we should believe, why not shrug our shoulders at the potential ‘loss’ of justification? Most of our beliefs already lack countless properties, even good- or better-making epistemic properties, like being incorrigible, deducible from the laws of logic, or knowable by reflection alone. Finding out that our external world beliefs lack these properties is no big deal and needn’t lead us to belief revision, so what’s one more property to the list? The point is that if epistemic justification is such that its presence or absence doesn’t bear on what we ought to believe, it’s difficult to see why finding out that our beliefs are unjustified is really problematic. Skeptics should thus be ‘normativists’ about epistemic justification.

The second step of the argument is that many of our most intimate and important beliefs make claims about what the external world is like. They are minimally objective, in reaching beyond oneself and how things merely seem to one. For example, my belief that my son loves me involves commitment to my son’s existence, and to there being a certain relationship between us, the relation of love.

Combining these two steps, we get the sub-conclusion that we ought not believe many of our most intimate, interpersonal beliefs. I ought not believe that my son loves me. I ought not believe that my family cares about me. I ought not even believe that my friends exist, and so forth.

If skepticism were correct, then, we ought to revise these personal, meaning-making beliefs. This has significant interpersonal costs. What I mean is that if we were perfectly compliant with radical skepticism, complying with the norm one ought not have external world beliefs, we should give up and no longer form external world beliefs, including these more personal, meaning-making beliefs. We should rather suspend judgment. But what kind of person would I be if I became undecided about whether my friends exist, or that my son loves me, despite no changes in my experiences?

The third step in the argument is that, complying with the norm, we’d need to roll back on those sorts of beliefs, but it feels like a harm to myself and a wrong (in my case) to my son and others to do that. In the paper, I defend this claim on three fronts. One is just that it strikes me as intuitively plausible. In conversation, I tend to get two reactions to this claim. First, that it’s obviously correct. It just jives with the way many people think and feel. Many have a gut reaction of uncomfortableness to giving up such beliefs. The other is that it’s clearly wrong. What matters is not that we have certain beliefs (or are not suspending judgment), but that we have certain emotions and perform certain actions.

I think this reaction is mistaken. Everything else being equal, we prefer the person who does not systematically suspend judgment on certain propositions about us. Stronger still, we want them to believe certain propositions about us. Consider two versions of your partner, spouse, or similar, where they act exactly the same way, say the same things, and your experiences remain the same. Imagine one of them believes that you love them, that you’re amazing, and a good person. They also act like this and have the fitting emotions. Now imagine the other version, who suspends judgment about whether you love them (they suspend judgment about whether you even exist!), whether you’re amazing, whether you’re a good person, and so forth. Are we really indifferent about whether we would rather be with the first person or the second person? More generally, isn’t there something preferable about the person whose doxastic life aligns with their affective and practical commitments and fits with shared experiences?

The other arguments are motivated by reflections about authenticity and respect. Imagine your suspensive partner again. They would be inauthentic for leading a life just as if he/she/they believed but didn’t really believe. They instantiate an objectionable sort of inauthenticity. It is not a one off or irregular occurrence. Rather, the person who complied with the skeptical norm, foregoing external world beliefs, would systematically lead an inauthentic life, acting like a believer without harboring the relevant commitments that belief embodies. They are deeply inauthentic. One could repair their situation by either believing again or by giving up on their actions and as-if behavior. But the latter would be a serious deprivation, reflecting an extreme ascetic life. This argument from authenticity suggests, then, that someone compliant with skepticism would need to be either deeply inauthentic or extremely ascetic. The former is objectionable, the latter is tragic.

The third argument is born out of reflecting on doxastic attitudes and respect. I think that suspending judgment in certain cases is a directed attitudinal wrong, but in the paper I didn’t (or at least I didn’t think I needed to) defend that claim outright (but I do elsewhere!). I appealed instead to the idea that when one lacks belief in certain cases, it is disrespectful. You can test this out by telling someone you mutually love that you do not believe they love you. Try this out on your best friend, child, or partner... Actually, don’t do that! It’s mean. I argue that certain kinds of relationships—those involving interpersonal reciprocal love—invite us to have certain doxastic as much as practical commitments we bear to each other. It’s hurtful not to recognize our friends, lovers, and companions as having that sort of status for us. This requires, as a kind of precondition, recognition of their existence; of believing that they exist as much as believing in them. I should believe, of a good friend, that she’s my friend, and vice-versa, and thus that she exists (and for her that I exist). Given our experiences at least, it’s disrespectful to act as we do but withhold judgment. We owe each other the recognition that belief makes good on. Since compliance with the sort of doubting norm that comes out of radical skepticism would have us do otherwise, we’d be tasked with doing something wrong.

Thus, the main conclusion is that radical skepticism has significant ethical costs. If we complied with the relevant norms, it would lead us to either an inauthentic, extremely deprived, or wrongful life. Radical skepticism is thus intolerable and indecent. What I mean is that, if we complied with radical skeptical norms of belief, we would do what’s bad for us or else wrongful to others.

But what does this ethical conclusion mean for the correctness or incorrectness of radical skepticism? Where does the skeptical argument go wrong exactly?

I also take up these questions in the paper. I distinguished between two versions of the view I defended. According to Abrogation moralism, skepticism is ethically bad, even if it is epistemically correct. On balance, we should learn to be content with the epistemic irrationality of our beliefs if there are significant ethical costs that comes with avoiding it. It’s not that the epistemic unjustifiedness of retaining our beliefs doesn’t matter, but it doesn’t matter as much as the ethical drawbacks of such extreme doxastic revision. The other is Encroachment moralism, on which the ethical costs of belief resistance and suspension of judgment affects the epistemic standards necessary for justified belief. The encroachment mechanism here is that, as the ethical costs of belief revision increase, the threshold for epistemic justification decreases, within some minimum limit.

From a Moorean perspective, this could mean that what it takes for me to be justified in believing “my son loves me” or “my son exists” is less than what is necessary by the skeptic’s lights (failing underdetermination principles or being introspectively distinguishable from the bad case). From a Relevant Alternatives perspective, the fact that radical skeptical hypotheses are incompatible with our important, interpersonal beliefs doesn’t make them relevant. Now, treating a live possibility as remote can be morally risky, and thus moral factors can contribute to whether a hypothesis is relevant. But the opposite is true as well: treating a remote possibility as live can be morally risky, making it seem like, to be epistemically justified in one’s belief, one must eliminate that possibility. Even raising radical skeptical hypotheses is morally hazardous. Maybe it doesn’t bring you to doubt what you should not, but it might bring others to do this.

I also address how my views relate to neighboring views, like those of Susanna Rinard and Eli Hirsch, the closest I’ve discovered in recent literature. I also explain the relationship between my ethical response and more traditional epistemic responses. One is that I think the two can harmonize. If, for example, our evidence supports believing what we do, that’s great, but if it doesn’t, that’s not so bad because skepticism would remain broadly ethically problematic.

I conjecture at the end of my paper that radical skepticism is politically harmful as well. This brings us full circle to comparing radical doubt with the doubt promotion of certain conspiracy theorists, trolls, and gas-lighters. If my core argument is sound, wouldn’t it also follow that, were skepticism correct, we’d need to give up believing significant political claims, or mixed historical/justice claims, such as that the Holocaust occurred, or that the U.S. instituted Chattel slavery? If skepticism were right, how could we live with each other? I see this as social-political apraxia.

My main message is that radical skepticism shouldn’t be thought of as an idle epistemological fantasy, but an extremely radical normative position, one with significant ethical ramifications. It is also increasingly approximated in the real life. We live among imperfect radical skeptics. Research on radical skepticism stands to profit from further inquiry in moral philosophy, social-political philosophy, social epistemology and the ethics of belief as much as traditional epistemology.