Epistemic Calibration Bias and Blame-Aversion

People typically treat having an importantly false belief as much more problematic than failing to have an importantly true belief.  They're more concerned about being over-confident than being under-confident in their credences.  But why?  Is such an epistemic asymmetry warranted?

I'm dubious.  The ideal is to be epistemically well-calibrated: to have just the degree of confidence in an important proposition that is warranted by your evidence, such that in the long run exactly X% of your "X% confident" beliefs turn out to be true -- no more and no less.  Moreover, it seems to me that we should be equally concerned about miscalibration in either direction.  If we are underconfident (or withhold judgment entirely) when our evidence strongly supports some important truth, that's just as bad, epistemically speaking, as being correspondingly overconfident.

Three Options in the Epistemology of Philosophy

You pursue your arguments as far as they go, and eventually reach your bedrock assumptions: foundational premises that you accept (and might describe as seeming 'intuitive' to you), but that you can give no further argument for.  Further, you realize that coherent philosophical diversity is possible: others could, coherently, accept (and find 'intuitive') different starting points from yours.  What should you do about it?

One might deny the stipulated set-up, and insist that there is ultimately only one internally coherent philosophical world view.  But that seems unlikely, so I will put aside that ambitious view for now.  One might also reject the "foundationalist" structure I've assumed above, defending instead the idea that our beliefs might form an interlocking (ultimately circular, but "mutually supporting") web, with no privileged starting points.  But I think what I say below will be easily translatable into coherentist idiom: the question just becomes what to think of our web of beliefs as a whole, given that coherent alternatives are possible.

Self-Undermining Skepticisms

Radical skepticism has a curious tendency to undermine itself:  If you can know (or justifiably believe) nothing at all, then you cannot know (or justifiably believe) even that.  So it seems that one cannot coherently take oneself to be incapable of forming justified beliefs.

More limited forms of skepticism might hope to avoid this fate. But it can prove difficult to halt the slide once you've started down that route.  Consider Sharon Street's epistemic argument against moral realism, which we might reconstruct as follows:

Judgmentalism vs Non-commitalism

Call Non-commitalism the view that we sometimes ought to suspend belief, assign imprecise credences spanning the entire interval [0,1], or otherwise refrain from doxastic commitment.

Opposing this, we have Judgmentalism, the view that we're never required to suspend judgment: there's always some doxastic commitment or other that we could at least as reasonably hold.

We might go further and consider Strong Judgmentalism, the view that there is always some doxastic commitment (e.g. some level of credence) that's rationally superior to suspending judgment entirely.

Which of these views is most plausible?  And (for any epistemologists in the audience) is there any existing literature on the topic?  (I just made up these names, so they might go by different labels if so...)

Non-Normative Epistemology?

I've previously cited epistemic nihilism as an example of an intellectual black hole -- a view that rational agents must reject in order to preserve their capacity for rationality.  But the normative nihilist might seek to avoid this implication by offering up a non-normative account of rationality (and epistemology more broadly). One could just stipulate a certain extensional account of "rationality" -- perhaps consisting in conformity to norms of parsimony, logical validity, certain inductive/abductive norms, etc. -- without requiring these norms to be backed by irreducibly normative properties, or to give rise to categorical "oughts".  This modest nihilist says, in effect, "Here are my recommendations if you wish to join me in the game of truth-seeking.  But if truth is not your goal, I have no grounds on which to criticize you."

My main concern about this move is that it isn't clear how the nihilist can consistently regard her own preferred epistemic norms as more truth-conducive than any others (at least so far as non-deductive norms are concerned). I've previously noted that anti-skepticism requires us to regard some possible worlds as (a priori) objectively more likely than others, and that to explain rational induction (e.g. the projectability of "green" but not "grue") requires positing objective structure (something empiricist-inclined nihilists may also regard as unacceptably mysterious or metaphysically "queer"). If all possible epistemic norms are metaphysically on a par, as the epistemic nihilist seems committed to, then why regard any particular set of norms as more likely to lead to truth than any other?

The Best Case for Uncertaintism

Liz Harman's excellent Madison Metaethics talk on 'The Irrelevance of Moral Uncertainty' [pdf]* got me thinking again about whether false normative beliefs can be justified.  As I previously noted, we might think that the fundamental normative truths, being non-contingent and presumably a priori, are always what it'd be ideally rational to believe.  Then again, insofar as they can also be highly non-obvious, we might think there's an important (less "strict") sense in which we rationally ought to apportion our credences broadly over a range of competing normative views.

Justification and Explanatory Normalcy

In his very interesting 'Justification, Normalcy and Evidential Probability', Martin Smith argues that justification requires one's belief to be normally true, rather than just very likely true, given one's evidence.  The relevant sense of 'normalcy' is explanatory rather than statistical: for the belief to turn out false would call out for explanation.  Beliefs based on perception, for example, are generally "normically supported" in this way -- for the belief to be false, something weird must have happened (perhaps the agent hallucinated, or was tricked somehow -- there must be some further explanation).  By way of contrast, the belief that your lottery ticket is going to lose might be as likely as can be (stretch the odds of the lottery as much as you like), still it would require no special explanation were this ticket to turn out to be the winning one -- some or other ticket must win, after all, it's just a matter of random chance.

Unreliable Philosophy?

Jason Brennan's 'Scepticism about Philosophy' raises a very interesting challenge, arguing that "[w]idespread disagreement shows that pursuing philosophy is not a reliable method of discovering true answers to philosophical questions."

I want to explore three ways of responding to the challenge (I should flag that these extend beyond the scope of Jason's paper, and so should be taken as "semi-related thoughts" rather than "objections").  We may question (1) whether there is any better alternative to doing philosophy as best we can -- however unreliable that may be; (2) the extent to which "discovering true answers" is an important goal of philosophy; and (3) whether "philosophy" in general is the relevant reference class for assessing reliability, or whether we should instead be assessing a range of more fine-grained methods.

The Kripke-Harman Dogmatism Paradox

Some thoughts inspired by yesterday's epistemology seminar...

If I know that h is true, I know that any evidence against h is evidence against something that is true: so I know that such evidence is misleading. But I should disregard evidence that I know is misleading. So, once I know that h is true, I am in a position to disregard any future evidence that seems to tell against h.

-- Gil Harman, Thought, p.148.

Apparently the standard solution is to distinguish what you know at various times: at t0 you know that h is true, and hence also know that any evidence against h is misleading. But once you actually acquire such evidence at t1, your total evidence no longer supports either fact, which is why you're no longer in a position to disregard this new evidence against h.

I think this is right, but it might help to say a little more to further dispel the air of paradox.

Williamson on Intuitions

In general, the objection "That's only an intuition" is ill-posed in the same way as the objection "That's only a judgment". Some judgments are indeed objectionable, but the mere fact that a proposition is judged is not even a prima facie reason for doubting it.

-- Timothy Williamson, The Philosophy of Philosophy, p.220.

Knowing What Matters (draft)

A pdf draft of my paper, 'Knowing What Matters' (forthcoming in Peter Singer (ed.), Does Anything Really Matter? Responses to Parfit) is now online.

Abstract:

Parfit's On What Matters offers a rousing defence of non-naturalist normative realism against pressing metaphysical and epistemological objections. He addresses skeptical arguments based on (i) the causal origins of our normative beliefs, and (ii) the appearance of pervasive moral disagreement. In both cases, he concedes the first step to the skeptic, but draws a subsequent distinction with which he hopes to stem the skeptic's advance. I argue, however, that these distinctions cannot bear the weight that Parfit places on them. A successful moral epistemology must take a harder line with the skeptic, insisting that moral knowledge can be had by those with the right kind of psychology -- no matter the evolutionary origin of the psychology, nor whether we can demonstrate its reliability over the alternatives.

Along the way, I also (1) argue against Street's "normative lottery" analogy; (2) argue that her constructivist view is self-defeating; (3) introduce an "internalist" version of reliabilism; and (4) offer an analysis of when it makes a difference whether some possible disagreement is actual.

I still have a couple of months to put on the finishing touches, so any suggestions would be most welcome!

The Moral Lottery

There are any number of logically consistent moral belief sets one might have, e.g. whereby the putative badness of pain waxes and wanes with the moon, or even where pain is seen as intrinsically good rather than intrinsically bad. Suppose that moral realists are right to think there's a mind-independent fact of the matter as to which of these possible moral views is correct. We don't think that the abstract moral facts played any causal role in explaining why we ended up with the particular moral beliefs that we have (there will instead be some psychological, sociocultural, or evolutionary story of that). So shouldn't we think it extraordinarily unlikely that our moral beliefs are (anywhere near) true, if realism is correct? So argues Sharon Street in 'Objectivity and Truth: You'd Better Rethink It' [pdf].

Huemer's Self-Defeat Argument

Michael Huemer forcefully advocates Phenomenal Conservatism (PC), the view that intuitive 'seemings' are prima facie justified. That is: "If it seems to S that P, then, in the absence of defeaters, S thereby has at least some justification for believing that P." (We should further add that, for Huemer, any possible defeaters must themselves be grounded in appearances.) He takes the following "self defeat argument" [quoted from our seminar handout] to rule out all competing views:

1. (Almost) all beliefs are based on appearances. (Exceptions: faith, self-deception.)
2. A belief is justified only if it is based on something that is a source of justification.
3. Therefore, there are justified beliefs only if appearances are sometimes a source of justification. (From 1, 2.) (Faith and self-deception are obviously not sources of justification.)
4. If PC is false, then appearances are never a source of justification.
5. So if PC is false, then no belief is justified. (From 3, 4.)
6. So no one is justified in believing any alternative theory to PC. (From 5.)

I'm not sure that (6) follows from (5). Some philosophers who either haven't seen or aren't convinced by Huemer's argument may find that some alternative theory X seems right to them, and if PC is true then this undefeated appearance can justify their belief in X. But never mind that: the conclusion in (5) is strong enough to establish PC (just tollens on the further premise that some beliefs are justified).

A more serious flaw, I think, is premise 4.

Reversing Metaphysical and Epistemic Priority

Ross Cameron asks a great question: what distinguishes legitimate from illegitimate appeals to the claim that "metaphysical priority isn’t epistemic priority"? To illustrate, consider Lewisian modal realism: although the Lewisian takes the possibility of talking donkeys to consist in the concrete existence of another spatiotemporally isolated world in which there is a talking donkey, he can claim to know the concrete fact in virtue of his prior knowledge of the modal one. On the other hand, it would be absurd for a Utilitarian to analogously argue that his prior moral knowledge that murder is wrong can ground his non-moral belief that murdering Bob would reduce unhappiness in the long run. Such reasoning would seem backward: in this case metaphysical and epistemic priority must coincide. Why the difference?

Philosophical Data

I've previously noted that some of the deepest philosophical disagreements are traceable to differing assumptions about what needs to be explained.

A reductionist might assume that we simply need to explain the third-personal empirical data: e.g. the facts that we have certain brain states, make certain vocalizations, etc. They then deny the reality of any further philosophical phenomena (e.g. irreducible qualia, normativity, etc.) on the grounds of parsimony: none of that is necessary to explain the scientific data. You don't need to posit moral facts in order to explain the occurrence of moral intuitions.

That's true enough. I don't think scientific inquiry should lead us to believe in moral realism. But I also don't think that science exhausts all there is to know. The philosophical 'data' may be more expansive than the empirical data, since we may know things other than third-personal observational facts. For an obvious example, I know that I'm phenomenally conscious. This datum calls out for explanation. For another example: we can (I assume) also know various moral/normative truths, in which case the search for a systematic moral/normative theory would be well-motivated (despite the fact that it wouldn't help explain the limited subset of data that the reductionist exclusively concerns himself with).

There are also various facts about epistemic normativity which call out for explanation. For example, I take it as given that rational induction is possible. We ought to believe that an emerald first observed after 2020 will be green, not grue. Such differences in 'projectability' may motivate positing objective structure: arguably, the reason why we can project 'green' but not 'grue' is because the former property is metaphysically privileged in an important sense. It is a more "natural" property, in the sense that collections of green things are (in respect of their colour) objectively more similar than collections of grue things.

So, realists (of various stripes) will dispute the reductionist's limited construal of 'the data to be explained'. This raises the difficult question: how can we make dialectical progress in resolving such fundamental disagreements? Are there generally acceptable moves for supporting or undermining the inclusion of a particular claim as "given"?

It seems there must be some constraints here. Something has gone badly wrong with the person who takes the existence of the Flying Spaghetti Monster as a datum. This seems like a belief that should be undermined by skeptical probing, if it does no essential work in explaining anything else. But then why can't the skeptic likewise dispense with any other self-contained cluster of beliefs (say, normative beliefs; or even empirical beliefs!), if they do no essential work in explaining anything else? What is it that makes empirical beliefs (and, arguably, normative beliefs) indispensable in this way?

I'm unsure what to say about this. My best attempt at an answer is to appeal to transcendental / "may as well" arguments. At least in case of epistemic normativity: anti-skepticism is a precondition for successful inquiry, so we may as well take it on faith. If we're wrong, we're screwed anyway; but if we're right -- if we have any chance at all of attaining knowledge -- then this assumption positions us to make the most of it. (Cf. Intellectual Black Holes.)

A similar argument applies to practical normativity: we may as well assume there's something we ought to do (then work out what it would be, and act accordingly). It's only on this assumption that our choices matter -- that we ever act rightly or wrongly -- so this is the only possibility that is worth taking into account.

It's less clear whether any such neutral justification can be given for the assumption of first-personal subjective experience. In this case, it's more of a Moorean fact: something we're more certain of than we are of any argument to the contrary. The inescapability thus looks to be merely psychological: there's little I could say to someone with different beliefs -- e.g. a self-identified "zombie" -- to convince them that they're really conscious. (One of my undergrad professors insisted that he has no idea what all this talk of 'subjective experience' is about.)

This shouldn't necessarily undermine the belief. We have little option but to reason from what personally strikes us as true -- even if others think differently, that shouldn't make it impossible for us to come to justified beliefs. But, on the other hand, nor should we want subjective certainty to suffice for justification: someone might be irrationally certain that the FSM exists, after all. The difficulty is that, at the bedrock level, there are no neutral arguments left to distinguish between the various positions. It looks like we must simply make a stab in the dark and insist that certain views (our own, if we're lucky) are brutely justified and others aren't.

It would be nice if we could say something a little more illuminating.

Personal vs. Objective Justification

I wonder if we can distinguish two important senses in a which a philosophical (a priori) belief may be justified. Most obviously, there is a person-relative sense, whereby some particular person is justified in their holding of a token belief. Such 'personal justification' may be attributed to an agent who has exhibited sufficient epistemic virtue in the forming of their belief, even if the proposition in question generally ought not to be believed. This latter assessment, by contrast, concerns what we might call 'objective justification' or warrant, and attaches to the proposition in general rather than anyone's particular believing of it.

One may object that this latter notion makes no sense: each person should believe whatever is best supported by their evidence, and that's all there is to say about epistemic justification. There is no more sense to be made of 'general justification' than 'general evidence' (unless this simply means 'evidence shared by most people', but there's nothing especially significant about that -- my notion of objective justification is not merely that of widespread justification), or so the objection goes. I think we actually may be able to elucidate an epistemologically significant notion of generally available (in circumstances C) evidence for a proposition, i.e. evidence that an epistemically virtuous inquirer (in C) would quickly acquire, such that any person (in C) in some sense really ought to believe the proposition even though -- due to their negligence in acquiring all relevant evidence -- their personal epistemic situation excuses or even licenses their contrary belief.* But I don't need to make so strong a claim. It suffices for present purposes to restrict our attention to a priori justifiable propositions, and so avoid these complications.

* (Clayton might say that their belief is rational but unjustified.)

The propositions of a priori philosophy and mathematics don't seem to be supported by evidence in the ordinary sense. Compare: our contingent experiences provide evidence for some empirical beliefs over others, and what we ought to believe about the world depends on these contingent observations. But truths of reason seem importantly different. Whatever "evidence" or reasons there are to believe one philosophical claim over its negation don't seem contingent in quite the same way; rather, they are presumably available (broadly speaking) to any rational agent as such -- hence the label 'truths of reason'. This allows us to drop all that complicating talk of "evidence available in circumstances C"; a priori justifications are "generally available" in the strong sense of being universally available (to sufficiently rational agents). We can thus understand talk of a proposition's being 'objectively justified' as simply a matter of its being a priori justifiable.

Importantly, we are not perfectly rational agents, and so our cognitive limitations might prevent us from subjectively grasping the objective justifications for believing some philosophical proposition. This creates a need for "non-ideal" epistemological theory (by analogy to non-ideal moral and political theory). Although the proposition may (ideally) warrant believing in virtue of the existence of an objective justification, we personally - as non-ideal agents - may justifiedly believe its negation, say due to our having a non-culpably incomplete or distorted understanding of the matter. Our non-ideality is revealed by the fact that we don't believe the objectively justified proposition. Nonetheless, our believing of its negation may be epistemically virtuous in its own way -- we may be successfully following epistemic rules (or heuristics of reasoning) that are generally more fit to our limited rational powers, even if they lead us astray in this particular case.

So I think we can make sense of a distinction between 'objective' and 'personal' justification, at least in relation to a priori beliefs. Note that this now opens a new position in logical space -- besides old-fashioned foundationalism and coherentism -- for the hybrid view that objective justification has a foundationalist structure, whereas personal justification has a coherentist structure.

(Aside: I'm inclined to attribute something like this hybrid view to Sidgwick, for reasons detailed here [II].)

P.S. Do epistemologists standardly make this distinction, or something like it? I don't think I've ever seen it explicitly discussed, but if you have, please pass along the reference!

Skepticism, Rationality and Default Trust

Intellectual virtue lies between the two extremes of dogmatism and radical skepticism. The irrationality of dogmatism is generally appreciated; less so the irrationality of radical skepticism (at least among my intro M&E students!). Let's attempt a remedy.

The global skeptic insists that we should question everything, trusting nothing: apparent perception, memory, "rational intuition", even methods of reasoning -- anything that can be doubted must be suspended from our minds and considered 'guilty until proven innocent'. The obvious problem with this stance - if followed assiduously - is that it entails rejecting everything, leaving one with no mind at all. This is the crucial point: just as a pure 'blank slate' cannot learn from experience, so must we rely on various assumptions if we are to learn, reason, and act rationally at all.

That's not to say we should dogmatically refuse to question our beliefs and practices. It's just that we cannot question them all at once. Any individual belief or assumption may be tested in light of the other things we (provisionally) take to be true, and revised if some incoherence is found. But we have to reason from the beliefs we have -- as Neurath famously put it, our mind is like a ship at sea, and even as we replace a faulty plank we must trust our weight to others.

The mere fact that those other beliefs can likewise be questioned does not suffice to show that we can't reasonably accept them (again, provisionally) in the meantime. The skeptic's claim to the contrary is itself a questionable belief, and not one we have any reason to accept. (It may seem intuitive at first, but once we fully understand the implications of this claim, it would be crazy to accept it -- or so I argue.)

To begin with, it's worth emphasizing that global skepticism is self-defeating, insofar as it implies that we should not trust its recommendation to trust nothing. Or, as Railton writes (in 'How to Engage Reason', Reason and Value, p.186):

Hume observed that [the sceptic] displays a touchingly unsceptical attitude towards the power of argumentation and his own powers of thought and memory. We might add: toward his own command of language and the content of this thoughts. Remove this default confidence, and he can no more declare his words to be 'giving an argument for scepticism'--an intentional action--than 'giving a recipe for haggis' or 'scat-singing without a tune'.

Hence my earlier claim that one who truly internalizes global skepticism is no longer capable of intentional thought or action at all. To accept global skepticism is to forsake any hope of rationality. It is the ultimate intellectual black hole. Railton (p.187) draws an important lesson:

Default trust, however 'blind', is not inherently blinding. On the contrary. If the sceptic trusts his ability to speak English and draw the conclusions demanded of his premisses, and I trust my own appreciation of his argument, we will both see (no longer be blind to) a problem that I have in defending my beliefs: Where I previously had hoped to be able always to have a reason for whatever I believe, taking nothing 'blindly' or 'without reason', I now realize that this hope is impossible--some things cannot, without regress or circularity, be argued for.

I trust the reader will by now agree that global skepticism is a non-starter, as any rational agent must take some things 'on trust' if they are to be capable of reasoning at all. Still, one might ask, what of slightly restricted forms of skepticism? Couldn't one consistently hold the more traditional skeptical view that it's just our sensory experiences and/or inductive practices that shouldn't be trusted? The traditional skeptic is willing to trust in reasoning as much as anybody is. They simply have different expectations about the external world (or "affirm a different prior") from the rest of us. In particular, they hold that all possible worlds are (a priori) equally probable, whereas we anti-skeptics consider some (perhaps simpler or more regular-seeming) distributions of properties across space and time to be more probable than others.

Sure, there may not be anything inconsistent about traditional skepticism, so defined. (Though such a skeptic has no reason to expect that they will continue to exist long enough to finish their thought.) But nor is there much reason to accept it, or indeed to find it any more credible than the claim that the world just came into existence 5 minutes ago. Admittedly, traditional skepticism is motivated by a premise that seems plausible at first glance, namely: "the reasonable 'default' view is to start off by assigning each possible world an equal probability of being actual." That sounds fairer and more reasonable than an a priori bias in favour of, say, worlds where memories are typically true--representing times and events that really did happen. But I suspect the intuitive plausibility of this is an instance of us being misled by overly-abstract principles. When we consider every more particular judgment or knowledge-attribution we are inclined to make, is it really plausible to think that the one abstract principle is more credible than all our conflicting particular judgments - and practices - combined? Colour me (cough) skeptical.

Upon reflection - in light of our actual beliefs - it seems we have most reason to reject the skeptic's principle. Though it seemed plausible at first, it is inconsistent with other claims that most of us find much more plausible. Here's another: rational agents should learn from experience. Skeptics -- even of the merely 'traditional' variety -- can't. As Railton writes (in 'Rational Desire and Rationality in Desire'):

It is well known that in order to learn about one's environment (whether one be human, animal, or teachable computer) it is not enough to have ample sensory input and plenty of memory registers to fill. The learner must also bring some expectations--such as expected dimensions of similarity (the "implicit quality space"). Otherwise, experience will simply accumulate in its infinite diversity, and all experiences will be equally relevant or irrelevant to one another. No lessons will be extracted.

Carnap gave an elegant demonstration of this point within the theory of logical probability. He asked us to consider a confirmation function that began (sensibly, it would seem) with no "prior bias", i.e., that assigned the same non-zero probability to every possible state of the world (what he called "state descriptions"). This function would, even given indefinitely large amounts of information about past states of the world, still assign the same probability to every logically possible way of extending this history into the future. In a fundamental sense, it could not learn from experience.

So I think it's pretty safe to conclude that (even the traditional form of) radical skepticism isn't rational. There's no guarantee that the rest of us are any better off, of course, but at least we have a chance. We do the best we can -- and we're yet to see any good reason to think that some other way is better.

One may be left feeling unsatisfied: the best we can do may not seem good enough. But since turning to skepticism is even worse, it seems we will just have to learn to live with tentatively trusting in the reliability of our perceptions, despite the uncertainty of it all. At least we can learn such things -- and that is something to be thankful for.

Rational Akrasia

I must say, I'm a huge fan of Nomy Arpaly's book Unprincipled Virtue. In the second chapter she argues against the common assumption that our deliberative judgments about what's rational (for us to do or believe) have any special normative authority or significance in determining what would in fact be rational for us. That is:

[S]ometimes an agent is more rational for acting against her better judgment than she would be if she acted in accordance with her best judgment... there are cases where following her best judgment would make the agent significantly irrational, while acting akratically would make her only trivially so. (p.36)

This is so in cases where one's "best judgment" is itself completely irrational -- a point I previously made in my post on 'Subjective Oughts'. Though this fact is often neglected, I don't think there's any sane view of rationality on which S believes that φ-ing is rational implies that S rationally ought to φ. Even on the most 'internalist' views, e.g. coherentism, it's possible to have false beliefs about what would be most coherent. That is, I might think that believing P is rational (and would increase the coherence of my belief-set), when in fact this is false and my belief set would be more coherent on the whole were it to contain not-P instead. Coherentism then straightforwardly entails that I ought to believe not-P, despite the fact that I believe that I rationally ought to believe P. The higher-order judgment is just wrong. And the same may be true in case of practical rationality, i.e. I may hold false views about what I rationally ought to do or desire. Following my judgment, in such a case, would not in fact be rational.

Once stated explicitly like this, the point seems very obviously correct. I guess one possible point of resistance could arise in readers who fail to distinguish what Arpaly calls an "account" of rationality from an "instruction manual". (Cf. the traditional distinction between the 'criteria of rightness' and a 'decision procedure'.) Obviously an instruction manual can't contain advice of the form, "Go against your all-things-considered better judgment", or "Do X, for a reason other than that this rule advises it", or "Don't think of an elephant". None of this is advice you can follow. Nonetheless, they may be true statements of what it would in fact be most rational for you to do in the circumstances. As I keep stressing, trying hard is no guarantee of rationality. Sometimes (though hopefully not often) one's efforts may even be counterproductive* -- this is perhaps most familiar in case of neurotic "over-analyzing", but another example would be overriding one's reliable (reasons-responsive) gut instincts with bad reasoning or rationalizations.

One virtue of Arpaly's discussion is that she highlights how (implicitly) familiar this point is in everyday life. We're all familiar with the idea of "a man who has some 'crazy notions' sometimes but whose common sense prevails 'in real life.'" (p.49) Most people aren't philosophers, or even particularly competent reasoners, so their explicit judgments may end up being downright dopey. If they fully internalized and acted on these dopey explicit beliefs, we might consider these folks fanatics. But because Uncle Bob doesn't really 'live out' or act on his dopey "best judgment" -- being instead restrained by his implicit "common sense", practical wisdom, and basic human decency (though he doesn't consciously realize it) -- we may judge that he is rational enough on the whole. His irrationality is restricted to his explicit beliefs, and he's otherwise ("in real life") not so bad. Akrasia -- failing to act on his explicit 'better judgment' -- thus makes him comparatively more rational than the fanatic he otherwise would have been. (Though of course he'd be even more rational if he didn't have such dopey explicit beliefs in the first place.)

* Stronger still: sometimes any attempt at explicit deliberation might prove to be essentially less-than-optimally rational. This point is more familiar in the context of ethics, where we may be required to respond directly to a person in need rather than mediating our response by any kind of deliberate moral theorizing. The person who considers the permissibility of saving his drowning wife before jumping in clearly has "one thought too many", as Bernard Williams puts it. We want people to be sensitive to moral considerations, but that doesn't require -- and sometimes precludes -- consciously deliberating about such things.

Epistemic Supererogation

It's commonly thought that epistemic normativity lacks the structure of moral normativity. We should believe in proportion to the evidence, and that's that. But most moral theorists will carve things up further, e.g. into the impermissible, the (minimally) permissible, and the supererogatory.

I'm not convinced there's any great difference here, however. In both cases, we ought to believe/do whatever we have most reason to believe/do. Nevertheless, we can distinguish between reasons that are more or less stringent, and failures or imperfections that are more or less forgivable. This is so even in the epistemic sphere. Some biases or errors in reasoning are worse - more blameworthy - than others. Young-earth creationists, holocaust deniers, etc., might be considered especially egregious offenders, for example. Moral and epistemic offenders alike may warrant censure -- blame in the one case, and ridicule or scorn in the other, perhaps.

Further, it seems pretty clear, in both ethics and epistemology, that not all imperfections (or failures to believe/do as we ought) call for such censure. There is a significant range of 'permissible', or tolerable, belief and action -- even though within this range, some options may be determinately superior to others.

I think we can even extend the notion of supererogation, or going "above and beyond the call of duty", to the epistemic sphere. Consider such ordinary optimistic biases as that of parents exaggerating the virtues of their children. What should Mother believe about the aptitude of her little Joe-Average? She certainly shouldn't consider him a genius -- that would be a bit too unhinged. But she may be forgiven for considering him a little above average, even if he's not (and the evidence is sufficiently clear about this). Indeed, we probably can't reasonably expect or demand much more from her than that.

But now suppose Mother is unusually dedicated to overcoming her biases, and reasons thusly: "Joe seems above-average to me, but mothers are known to overestimate such things, so I really should compensate for my own bias here. This suggests that Joe is really about average." Here it seems to me that Mother is going above and beyond the call of (epistemic) duty. A normal, adequately reasonable agent would probably have just stuck with their conveniently over-optimistic initial judgment. But she really went the extra mile. Doesn't that sound like supererogation?

Subjective Oughts

People often assume that there's some genuine sense in which what we ought to do (believe) is determined by whatever we believe we have most reason to do (believe). Call this 'subjectivism'. Subjectivist positions seem common in debates over peer disagreement and normative uncertainty (to name just a couple of examples). But I think it is mistaken.

Granted, there may well be wide-scope requirements e.g., to not believe (i) that the evidence conclusively supports P without also believing (ii) that P is true. But it doesn't follow from my believing of (i) that I ought to also believe (ii). Perhaps I should instead give up my belief in (i).

As I pointed out in 'Rational Objectivity', rational status is not perfectly transparent: we can be irrational without realising it. In particular, it's possible to believe that I rationally ought to φ [e.g. believe P] without this truly being so. This possibility of error is essential to any non-trivial rational norm, thus ruling out the possibility of subjectivism. (Bootstrapping cases are helpful to illustrate this objection more vividly. We can describe a scenario in which it an agent is patently unreasonable in believing P. But subjectivism implies, absurdly, that their belief may be justified by the mere fact that they erroneously take their evidence to support the ludicrous proposition.)

If subjectivism is so daft, why are so many people initially tempted to accept it? I think there are three main reasons. The first, noted above, is the confusion of narrow- and wide-scope requirements. The second is that in bootstrapping cases, the agent is at least exhibiting some (perhaps limited) procedural epistemic virtues. A good epistemic agent will, after all, align their beliefs with their judgments about the evidence. The problem is that this is woefully insufficient to qualify as a good epistemic agent, if one's judgments about the evidence are not themselves reasonable. Indeed, taken in isolation, partial "virtue" may simply lead one further astray. (Compare: an instrumentally rational psychopath at least displays certain 'executive virtues', but their competence actually becomes a bad thing given how warped their ends are.)

The third - and I think most important - reason has to do with considerations of 'action guidance'. The theoretical role of rational norms is, after all, to guide us when we can't tell what we (objectively) ought to do. So there has got to be something a bit more subjective about them. The considerations that make one option rationally superior to another must be considerations that are accessible to us. Subjective beliefs are the obvious candidates: they're accessible to us in a way that external facts are not. And, indeed, there are independent theoretical motivations for accepting a kind of 'internalism' about rationality, i.e. the thought that what's rational for me depends entirely upon facts internal to my mind, not the external world.

But it's simply a mistake to think that internalism implies subjectivism. After all, subjectivism restricts itself to a very specific subset of my beliefs, namely my normative beliefs about what I ought to do. What about my ordinary non-normative beliefs? If I know that a generally reliable source just told me "P is false", but I irrationally interpret this as evidence that P is true, subjectivism licenses my irrational belief that P. But we needn't go along with this. There's a perfectly accessible fact which counts against the belief, namely the testimonial evidence I just heard (and perfectly well remember). Again, I know full well what the source said -- this information is as accessible to me as any -- my error is one of normative interpretation. I unreasonably interpreted this basic fact as evidence for P when really it is (as I should have known) evidence against P. My mistake, right?

This is the key issue. The subjectivist claims that what's rational is determined by what the agent treats as evidence. These normative judgments are themselves taken as 'given' and beyond dispute. But I contend that we cannot get any worthwhile action-guiding norms when so much is taken as given. If my normative judgments are sufficiently unreasonable, then their implications are of no rational help. (Garbage in, garbage out.) There is no sense in which I 'ought', automatically, to do whatever it seems to me I ought to do. Even in the most subjective of genuine rational norms, tailored for non-ideal agents, my beliefs about what I ought to do are always open to question, and so might be rationally trumped by certain of my other beliefs -- even if I'm too irrational to realize it.