In his paper “What Does ‘Epistemic’ Mean?” Allan Hazlett offers a series of reflections on the use of that term, ending with a fairly radical conclusion: we should stop using it. Just as Hume's Enquiry Concerning Human Understanding closes on a jarringly intemperate note – where he advocates the burning of books that are neither mathematical nor experimental – so, too, does Hazlett's ending seem abruptly extreme:
Take any contemporary philosophical essay, and consider each occurrence of “epistemic.” I submit that each is either superfluous – in which case to be stricken – or replaceable with alternative jargon – in which case to be so replaced. This will give you a picture of what philosophy without “epistemic” looks like.Footnote 1
Not only does Hazlett want to deprive epistemologists of a treasured word, he also hints that philosophy itself might be better off without epistemology altogether. In its place would be a discipline that attempts to answer, in the broadest way possible, the question “what should I believe?”.
As someone who has comfortably settled into regarding himself as an epistemologist, I cannot permit this shot across the bow to pass unanswered. In what follows, then, I'll first try to reconstruct the thread of Hazlett's argument before defending what I take to be a sturdy, traditional understanding of what epistemology is and why, as the discipline devoted to the epistemic, it is an essential and ineliminable part of philosophy.
1. WHAT DOES “EPISTEMIC” MEAN?
Hazlett's line of argument rests on three major points. First, the term ‘epistemic’ has two uses. (Alternatively, he says that the concept epistemic is disjunctive.) Second, the best explanation for these two uses is that epistemology focuses on beliefs that “fulfill their aim.” And, third, this focus is too narrow for epistemology – or whatever its successor discipline should be. Philosophy should answer the most fundamental questions, and there are considerations relevant to the question what should I believe? that arise independently of belief's “aim.”
I'll take these points up in the order in which they appear.
2. KNOWLEDGE AND BELIEF
Hazlett argues that the term ‘epistemic’ has two uses. The first use has to do with knowledge, while the second has to do with belief. Not much needs to be said about the first use; this is the usual understanding of what the epistemic is. Epistemology is the theory of knowledge, and the epistemic would be what has to do with knowledge and related phenomena, like justification, evidence, certainty, and the like.
The second use, according to Hazlett, can be seen in a couple of examples he gives. In characterizing how Sextus Empiricus talks about the role of appearances in Pyrrhonism, Jonathan Barnes distinguishes an epistemic from a non-epistemic sense of ‘appear’.Footnote 2 If the skeptic means to be using the non-epistemic sense when talking about how things appear to her, then we can see that she is simply reporting the character of her experiences. To use an anachronistic but helpful phrase from Roderick Chisholm, the skeptic might say, for example, that she is being appeared to redly. There is no implication at all that her environment actually contains anything red. By contrast, if we should understand the skeptic as using the epistemic sense, then her report is getting at how things are in her environment, but in a non-dogmatic way. So, she might say, “I think it's red,” where the italics indicate the sort of hesitance we often display when we make tentative and easily defeasible claims.
The other example Hazlett provides comes from Sarah Stroud's argument that friendship can require what she calls “epistemic partiality.”Footnote 3 The idea here is that friendship can place constraints, not only on how we feel or act in relation to our friends, but also on what we believe. On a traditional account, friendship might require you to continue loving a friend, no matter what awful thing you think he's done. On Stroud's more radical account, friendship might require you to believe he didn't do the awful thing the evidence indicates he has done.
Neither example, I think, provides a clear instance of the purported second use of ‘epistemic’. In the case of the Pyrrhonist, an epistemic use of ‘appear’ is one that attempts to indicate how things are – not just what the Pyrrhonist believes. The “appearance” is one that can be assessed for how well it does in fact reflect what is true. Presumably, the Pyrrhonist will think that the appearance does not count as knowledge (depending on what one means by “knowledge”), but the key point is that the claim is in the same game as the straightforward assertions made by the Pyrrhonist's dogmatic rivals. And, in the case of Stroud's account of epistemic partiality, it seems that the clearest interpretation of what she means is that friendship makes demands on the subject as she is going about the business of acquiring knowledge.Footnote 4 That is, friendship constrains one's beliefs because it constrains how one goes about forming beliefs – and the latter is an epistemic enterprise.
So, neither Barnes nor Stroud provide a compelling instance of the second use of ‘epistemic’. But, even if they did, there is a simple answer available to a philosopher who wants to hold on to the mainstream sense of ‘epistemic’ as having to do with knowledge: reject the alternative usage in Barnes, Stroud, or whomever else leans on it. As Hazlett himself notes, this use is less common and already subject to being discarded as loose talk. If it has become prevalent enough to cause confusion, as Hazlett apparently fears it has, then let's simply engage in a bit of boundary-policing: from here on out, we will use ‘epistemic’ only in the sense that has to do with knowledge. In other words, we can reform epistemology without calling for a revolution.
3. THE AIM OF BELIEF
Hazlett proposes that the disjunctive conception of the epistemic stems from our recognition that “belief aims at truth” – a supposed platitude that has been quite influential in the decades after Bernard Williams suggested that this is what is most characteristic of belief.Footnote 5 There are (at least) two different ways in which the platitude has been understood. On the one hand, it might mean that belief by its very nature is directed at the truth, perhaps because humans have evolved in such a way that they have a truth-directed attitude of this kind. On the other hand, it might mean that belief is governed by truth as a norm, in the sense that it is always evaluable with respect to the truth, regardless of whether in any particular case the belief in question actually is true. But, on either understanding, Hazlett thinks that there is an understandable connection between the two uses of ‘epistemic’. One of them has to do with belief, and the other has to do with belief that successfully meets its “aim.”
Much more could be said about the “aim of belief,” but two quick replies will have to suffice for now. First, many epistemologists do not find it helpful to ground theorizing about knowledge in Williams's platitude. Even if we suppose it to be true, it may not provide a fruitful enough basis for epistemology. (Analogously, many moral philosophers may doubt that the nature of action can provide by itself a sufficiently rich basis for morality. Something more – pleasure and pain, autonomy, freedom, etc. – needs to be added before we can get clearer on what is most essential to morality.) Epistemology is a broad field, including everyone from traditional foundationalists to coherentists to reliabilists and modal epistemologists. None of them make much of the “aim of belief.” So, Hazlett's proposed explanation of how philosophers think of the epistemic is unlikely to be correct.
Second, it is far from clear that philosophers should think of the epistemic as grounded in the “aim of belief,” on either understanding of what that might mean. Even if belief by its nature “aims” at truth, that fact by itself need have no significance for us at all. Suppose scientists tell us that humans have violent urges, mental states that evolved specifically to lead to the conquest of surrounding territory and the subjugation of our neighbors. Any attempt to vindicate the more sordid elements of human history by showing how they are rooted in these urges would certainly be guilty of the naturalistic fallacy.
Notice, though, that this objection encompasses not only the natural interpretation of Williams's platitude but also the normative one. Suppose that scientists discover that the long-dead inhabitants of Easter Island had a genetic mutation that prevented them from experiencing the full force of the urges that drive other humans to violence. Where other humans are quick to respond with a vicious bloodlust, the Easter Islanders would instead try to deal their enemies a gentle slap on the wrist. If so, their violent urges do not work properly. Qua violent urges, they are defective or bad. That is to say, they don't meet the norm for violent urges.
These are normative assessments that we would, I hope, be inclined to discard. They provide no guidance for us as we decide how to behave. But that point should apply just as well to the claim that truth is the norm of belief. Beliefs can be assessed with respect to their own norms without that having any consequences for what we should do. Qua beliefs, those that are true are good. But that doesn't yet mean that they are good for us. To put the point differently, the fact that belief “aims” at truth does not mean that we should aim at truth. To be clear, I do think that it is good for us to have true beliefs; the problem is simply that Williams's platitude doesn't give us a good explanation of why this is so.Footnote 6
4. THE IMPORTANCE OF THE EPISTEMIC
Finally, Hazlett thinks that it is wrong for epistemology to focus only on beliefs that successfully fulfill their “aim.” Philosophy should be concerned primarily with the question how should I live? This can be broken into “practical and theoretical sub-questions”: what should I do? and what should I believe?. The practical sub-question is the domain of ethics, but the theoretical sub-question is not the focus of epistemology. In focusing entirely on beliefs that successfully fulfill their “aim,” epistemology ignores beliefs that are false (and unsupported by available evidence) but still good (in some sense) to have. William James gives the classic example of this sort of case: a patient who is told by her doctor that she has a disease that is very likely to kill her might nevertheless be able to raise the odds that she will survive by believing that she will. Her chances won't improve so much that she will be more likely than not to survive, so this won't make her belief epistemically justified. But, arguably, the increased chance of survival gives her a strong practical reason for holding an evidentially unsupported belief.
Epistemology as traditionally conceived tells us only that James's patient would be wrong to believe she will survive – her belief is unjustified and fails to be knowledge. Hazlett finds this objectionable. He proposes that we should either reconceive of epistemology – so that it becomes the discipline that answers the general question, what should I believe? – or create a new “discipline distinct from epistemology, analogous to philosophical ethics,” to answer it.
But, it seems to me, Hazlett has an idiosyncratic understanding of what philosophical ethics amounts to. It is not simply a discipline devoted to answering the very general question, what should I do? There are, in fact, several philosophical disciplines that answer that question in different, often overlapping ways: decision theory, game theory, theories of practical rationality, moral theory, ethics in the broad sense (as in, e.g., Aristotle or Spinoza), and so forth. Some of these attempt to answer the question in what we might call “value-neutral” ways: given that you have a certain goal, and certain expectations about what is likely to be the case, they will help you decide what to do. Others bring substantive values to bear, not only in selecting your goals, but also in constraining which means you can take to reach them. It's clear that the recommendations of decision theory can conflict with the requirements of morality, just as one's personal goals may come into conflict with one's moral obligations. Is there a philosophical discipline that attempts to find an all-things-considered resolution to these conflicts? If so, it is philosophy as a whole.
Should we, then, stop distinguishing between different branches of philosophy – epistemology, ethics, and the rest – and think of ourselves as philosophers simpliciter? This would be a wrong turn, I think, for at least two reasons. Although they hold generally, I'll formulate them as a special plea for epistemology.
First, if we dumped into the same barrel of reasons every possible consideration relevant to whether it would be good (in the broadest possible sense) to have a particular belief, we would be permitting practical considerations to weigh directly against evidence. It is far from clear how this could be done. The case of James's ill patient is compelling because it makes the relatively weak point that at least sometimes you can have practical reasons “count for more” than evidence. But this is still nothing like a systematic view, and it provides no guidance as to how practical reasons are to be weighed against evidence.Footnote 7
Second, and what is worse, accepting Hazlett's view would mean we would lose our ability to say that James's patient is doing something wrong in believing that she will survive her illness. Given our current, traditional understanding of epistemology, it is open to us to say that she is wrong from an epistemic point of view, even though her having the belief in question is best all-things-considered. But, on Hazlett's view, James's patient would not be doing anything wrong from any perspective. In the same way that someone might form a belief where she has evidence on both sides of the question, but a preponderance favoring one side over the other, the patient would be weighing her considerations against one another and forming the best-supported belief. But these cases are nothing like each other.
Imagine filing a criminal complaint against a billionaire who has run over your dog. You present the judge with a video of the traumatic incident, in which the billionaire's license plate is clearly visible on the offending car. The billionaire doesn't bother to mount the usual sort of defense – he just offers the judge a lot of money to rule that he is not guilty. The judge now weighs your evidence against the practical considerations provided by the defendant. As she wavers in making her judgment, the billionaire just offers more money.Footnote 8 Her answer to the question what should I believe? takes all possible considerations into account. Unfortunately, you have no more evidence to offer, and you don't have much money. The judge rules against you. You might complain that her judgment is false, but she replies that this is only one of several considerations – and by no means the weightiest. If Hazlett's view is correct, this is the end of the story. You have no real basis for complaint, as neither the judge nor the billionaire have done anything wrong.
But this can't be the end of the story – there has to be more to living well than leveraging one's money and power effectively. This is why epistemology (and moral theory) is essential. Our concern for the epistemic – for knowledge, justification, epistemic rationality, understanding and even wisdom – is fundamental to the philosophical enterprise.
