justified, entitled, reliably formed, certain, and the like; here I'm using a sprinkling of
merit-denoting words from the epistemologist's lexicon. The epistemologist wants to
illuminate the conditions for the presence and absence of such merits.
Let me bring to the surface some assumptions in what I have just said. I assume, in the
first place, that propositional attitudes do have merits and defects; that seems just
obvious. I assume, in the second place, that whereas some of these have something to do
with truth, others do not. One good thing about beliefs is that they are components in
desirable emotions; some beliefs, for example, make one happy. But making one happy
has nothing to do with truth; false beliefs are just as good at making one happy as true
ones. By contrast, the merit in a belief of being reliably formed, to take just one example,
obviously does have something to do with truth. Third, I assume that beliefs are not
unique in possessing truth-relevant merits but that other propositional attitudes possess
such merits as well; just as a belief may be rational or not, so also a hope may be rational
or not. And last, I assume that there's not just one truth-relevant merit that we find in
beliefs and other propositional attitudes but a plurality of such. The point of this last
remark is that, until rather recently, epistemology in the twentieth-century analytic
tradition was almost always conducted under the assumption that there's just one truth-
relevant merit, sometimes called “justification,” sometimes called
end p.246
“rationality.” The literature was then filled with competing theories of justification or
rationality. The assumption seems to me decisively false. Here is perhaps the clearest
example. It's one thing for a person to hold a belief that he's not entitled to hold, one that
he ought not to hold; it's quite another thing for a person to hold a belief that's not been
reliably formed. Yet both entitlement and reliable formation are truth-relevant merits in
beliefs.
I said that it belongs to the work of the epistemologist to illuminate the conditions under
which truth-relevant merits are present in propositional attitudes generally. As a matter of
fact, however, epistemologists have concentrated almost entirely on beliefs; discussions
of the epistemology of hope, for example, are rare indeed. So also when it comes to the
epistemology of religion; prominent though hope, trust, regret, and so forth are within
religions, the epistemology of religion has focused almost entirely on religious beliefs. I
speculate that the reason for this is that epistemology has been in the clutches of the
preoccupation of philosophers with knowledge—and because knowledge, in the
twentieth-century analytic tradition, has been understood as a species of belief. In
contrast to the expansive account that I have just given of the subject matter of
epistemology, many writers would have led off by saying that epistemology is theory of
knowledge. They would have had etymology on their side; episteme in Greek means
knowledge, and logos means theory, hence, theory of knowledge. My response is that if
one actually looks at how epistemology has developed, one sees that it long ago
outstripped the etymology of its name. Though John Locke, for example, was concerned
to articulate an account of knowledge, he was at least as concerned, if not more, to
articulate an account of what I am calling “entitlement.”
I regard it as regrettable that epistemology has concentrated almost entirely on belief and
knowledge, to the neglect of other propositional attitudes—particularly regrettable in the