Truth as the Fundamental Epistemic Good

1 For The Ethics of Belief: Individual and Social, (eds.) J. Matheson & R. Vitz, (Oxford: Oxford University Press, forthcoming).

TRUTH AS THE FUNDAMENTAL EPISTEMIC GOOD

DUNCAN PRITCHARD University of Edinburgh

ABSTRACT. We begin with a puzzle. It used to be commonplace in epistemology to regard truth as the fundamental epistemic good, such that it is the goal of inquiry; and yet now this conception of the fundamental epistemic good is widely rejected. But what prompted this change of view is far from clear. Three considerations are considered which might be thought to count against the idea that truth is the fundamental good. First, that if truth is the fundamental epistemic good, then it is impossible to account for the greater value of knowledge over mere true belief. Second, that since we don’t value trivial truths, so the goal of inquiry cannot be truth simpliciter. Third, that because what legitimately closes inquiry is not truth, so truth cannot be the goal of inquiry. It is argued that all three considerations are unpersuasive, and thus that we should reconsider the possibility that it is truth which is the fundamental epistemic good.

0. INTRODUCTION That truth is the fundamental epistemic good used to be orthodoxy within epistemology. Indeed, isn’t it simply characteristic of the epistemic that it is directed at truth? Relatedly, isn’t epistemology concerned with the good to which a properly conducted intellectual inquiry is directed, and isn’t the goal of such an inquiry truth?1 In the recent literature in epistemology, however, the idea that truth is the fundamental epistemic good, along with the related thesis that truth is the goal of inquiry, has increasingly been treated as dubious. Indeed, I think it is now fair to say that the new orthodoxy in epistemology is that truth is not the fundamental epistemic good.2 What is less clear, however, is just why there has been this radical shift in our epistemological thinking. In what follows I will be examining three considerations which one might offer as grounds

2

for rejecting the truth thesis, and I will argue that they are all unpersuasive. The dialectical strategy of this paper is thus explicitly negative and piecemeal. The former, because I will not be offering a positive defence of the claim that truth is the fundamental epistemic good, but rather arguing that the reasons proffered for rejecting this claim are unconvincing. The latter, because I will only be examining a selection of considerations which are held to count against this claim. Nonetheless, this should be sufficient to demonstrate that we should look again at this ‘old fashioned’ conception of the epistemic good. In §1, I spell out what I have in mind when I talk of truth as the fundamental epistemic good. In §§2-4, I then proceed to consider the three considerations which I take to have led commentators to abandon this thesis and I argue that they are all unconvincing. In §5, I offer some concluding remarks.

1. TRUTH AS THE FUNDAMENTAL EPISTEMIC GOOD When I say that truth is the fundamental epistemic good, I mean that from a purely epistemic point of view it is ultimately only truth that we should care about. Call this the truth thesis. The easiest way to further unpack what I take to be involved in the truth thesis is to see what consequences I draw from it. One type of consequence that I draw concerns epistemic value. To say that truth is the fundamental epistemic good is to be committed to the idea that there is only one finally epistemically valuable epistemic good and that this is truth, with all other epistemic goods being merely instrumentally epistemically valuable relative to truth. So, for example, on this view insofar as epistemic justification is an epistemic good, then it’s epistemic goodness is derivative on the epistemic goodness of truth, in the sense that we instrumentally epistemically value justification because we finally epistemically value truth. Elsewhere, I have characterised this view as epistemic value T-monism, in that: (i) it is a view about epistemic value specifically (that’s the ‘epistemic value’ part); (ii) it says that there is just one finally epistemically valuable epistemic good (that’s the ‘monism’ part); and (iii) it says that this finally epistemically valuable epistemic good is truth (that’s the ‘T’ part).3 A key point to note about treating truth as a finally epistemically valuable good is that it does not follow that truth is finally valuable simpliciter. For example, it does not follow that the truth always generates some pro tanto goodness. The point is just that from a purely epistemic point of view truth is rightly to be valued non-instrumentally. But this is entirely consistent with arguing that the general value of truth is merely instrumental (or indeed, that it has no value at all).

3

Perhaps, for example, the only value of truth from a non-epistemic point of view is practical value, which is presumably a form of instrumental value. I don’t think we should be surprised by this, nor is this point specific to the epistemic domain, since one can draw analogous conclusions about other domains. Suppose one argued that the fundamental aesthetic good is beauty, for example, such that the beautiful is finally aesthetically valuable. One is not thereby committed to claiming that the beautiful is finally valuable simpliciter. That is, one might draw this conclusion, but it would be a further step. In particular, it is open to one to argue that the beautiful, when assessed outwith a specifically aesthetic axis of evaluation, is only instrumentally valuable, or perhaps not valuable at all. Indeed, the point being made here is just an instance of a more general point that one cannot infer, without further argument anyway, that because something is good when assessed along one axis of evaluation then it is also good when assessed along another axis of evaluation, much less that it is good simpliciter.4 A second consequence that I draw from the truth thesis is that truth is the telos of a properly conducted intellectual inquiry. By this I mean that seeking truth is the constitutive goal of a properly conducted intellectual inquiry, in that if one is undertaking an activity which is not directed at seeking the truth, then one is not undertaking a properly conducted intellectual inquiry at all. Of course, it is consistent with this claim that properly conducted intellectual inquiries can have all sorts of other goals too, some of them non-epistemic. It is an interesting question how the claim that truth is the fundamental epistemic good and the claim that truth is the constitutive goal of a properly conducted intellectual inquiry are related. Clearly they are deeply entwined proposals, to the extent that it is hard to imagine someone defending the one without the other. I just characterised the latter thesis as being a consequence of the former thesis, but one could just as well argue that the former is a consequence of the latter (or indeed that they both spring from a common source). For our purposes, we can set this issue to one side, since all that matters is that these theses go hand-inhand and that in defending the truth thesis I will be defending both of them. One final point about the truth thesis is in order. I have characterised this thesis in terms of the claim that truth is the fundamental epistemic good, but proponents of the truth thesis often characterise the view not in terms of truth simpliciter but rather specifically in terms of true belief.5 Read one way, this substitution of truth with true belief is harmless. In order to see this, consider again the idea that beauty is the fundamental aesthetic good, such that from a purely aesthetic point of view it is ultimately just the beautiful that one cares about. Now, the thought might be that just as in valuing beauty we thereby value, and thus seek out, those things (beautiful artworks, the aesthetic experience of beauty, etc.,) which instantiate that property, so in valuing truth we thereby value, and thus seek out, items which have that property, such as true beliefs (which, from

4

a first-person point of view at any rate, are the most obvious bearers of this property). On this reading, it is still truth which is the fundamental epistemic good. That is, on this reading saying that true belief is the fundamental epistemic good is basically a useful way of roughly paraphrasing, rather than redefining, the truth thesis. There could be something more radical going on here, however, in which case it’s worthy of note. Consider again the aesthetic analogy. One might argue that what is of fundamental aesthetic value is not beauty itself but rather that which we seek out as instantiating beauty. So the fundamental aesthetic good is not beauty but rather some common item which has that property, such as the aesthetic experience of beauty. Perhaps the thought is that we somehow unduly ‘reify’ aesthetic value to think otherwise. Similarly, one might be tempted to advance an analogous claim in epistemology!viz., that, at risk of an undue reification, it is not truth that is of fundamental epistemic value, but rather some common item which has that property, such as true belief. If that is what is going on when truth is replaced by true belief as the fundamental epistemic good, then this is not a mere paraphrasing of the truth thesis, but a redefinition. My view is that the reasoning behind this replacement is mistaken. The aesthetic case is illustrative in this regard. From a purely aesthetic point of view, we surely value the aesthetic experience of beauty because we value beauty, which indicates that beauty plays a more fundamental axiological role here. Similarly, I would maintain that from a purely epistemic point of view we value true belief because we value truth.6 This is not the place to get into the merits or demerits of the truth versus true belief conceptions of the fundamental epistemic good. What is salient for present purposes is just that I am construing the truth thesis specifically in terms of truth and not true belief. (That said, I will comment further below on the issue of whether we should substitute truth with true belief on this score).7

2. EPISTEMIC VALUE AND THE VALUE OF THE EPISTEMIC One reason why contemporary epistemologists are inclined to reject the truth thesis concerns the apparent inability of proponents of this thesis to accommodate the widespread intuition that knowledge is more valuable than mere true belief. In particular, the contention is that epistemic value T-monism is inconsistent with the claim that knowledge is more valuable than mere true belief, and thus that, since the latter is undeniable, the former, and with it the truth thesis, must go. This problem is known as the swamping problem. There are two premises to the swamping problem. The first is epistemic value T-monism,

5

which we have just discussed in the last section. The second is a general claim about value. This holds that instrumental value gets ‘swamped’ when the relevant non-instrumental good in question obtains. To take an oft-cited example, let us grant that coffee is ultimately valued for its sensory qualities!its taste, smell, appearance, quantity, and so on. Someone who values coffee in this way will also value those implements which are good means to successful coffee production, such as reliable coffee-making machines (that is, coffee-making machines which regularly produce good coffee). Here is the problem. For while we might well instrumentally value reliable coffee-making machines, on account of their propensity to deliver us great coffee, we don’t care about how a cup of coffee was produced once we have a great cup of coffee before us. To illustrate this point, consider two identical cups of coffee, identical in every sensory respect!same taste, same smell, same quantity, same appearance, and so on!but where only one of the cups of coffee was produced by a reliable coffee-making machine. Would it matter to us which of these two cups of coffee we were given? Put another way, would we be willing to pay more to have one of these cups of coffee over the other? I think it is pretty clear that we would not. The upshot of this is that we only value reliable coffee-making machines as a means to good coffee, but that once we’ve got the great coffee this ‘swamps’ the value of having coffee produced by a reliable coffee-making machine.8 The thought is that what goes for coffee applies to true belief, at least if the conception of epistemic value offered by epistemic value T-monism is right. For take an epistemic standing like reliability. According to epistemic value T-monism, the epistemic value of a belief being formed via a reliable process is instrumental epistemic value relative to the good of truth, but if the truth is already gained!i.e., if our agent already has a true belief in the target proposition!then that it was formed via a reliable process cannot confer any additional value. In particular, as with the coffee cup case, a true belief formed via an unreliable process (or in an unjustified way, etc.,) will be no less epistemically valuable than a true belief formed via a reliable process (or in a justified way, etc.,). But of course, if that’s right, then whatever epistemic standing one adds to true belief in order to get knowledge can confer no additional epistemic value either, and hence it follows on this view that knowledge cannot be epistemically more valuable than mere true belief. One way of thinking about the swamping problem is that it is a reductio of epistemic value T-monism, at least if one grants that it is undeniable that knowledge is more valuable than mere true belief. It’s certainly clear that lots of epistemologists have conceived of the swamping problem in this way.9 But a closer inspection of what is going on here reveals that this problem is in fact entirely illusory.

6

In order to see this, take another look at the swamping argument, which we can formulate as follows: The Swamping Problem (SP1) If the value of a property possessed by an item is only instrumental value relative to a further good and that good is already present in that item, then this property can confer no additional value on that item. (SP2) Epistemic properties are only instrumentally valuable relative to the good of truth. (SC) Knowledge that p can be no more valuable than mere true belief that p.

(SP1) is the general claim about value that we discussed earlier, and (SP2) is essentially a statement of epistemic value T-monism. And yet (SP1) and (SP2) seem to clearly entail the conclusion, (SC), which is in direct conflict with the widely held claim that knowledge is more valuable than mere true belief. So construed, one can see why this argument looks like a reductio of epistemic value Tmonism. But there is a subtle sleight of hand in play here. In particular, we should recall the point made in the previous section that we cannot (without further argument anyway) infer from the fact that something is valuable (or disvaluable) when assessed along a certain axis of evaluation that it is valuable (or disvaluable) along another axis of evaluation, much less that it is valuable (or disvaluable) simpliciter. In particular, for our purposes, it does not follow from the fact that something is lacking in epistemic value that it is lacking in value simpliciter. In short, we should recognise the distinction between the value (or disvalue) of the epistemic and epistemic value (or disvalue). This point is important because the premises of the swamping argument do not in fact entail the conclusion as formulated as above. Instead, what they entail is this weaker conclusion: (SC*)

Knowledge that p can be no more epistemically valuable than mere true belief that p.

This conclusion is entirely compatible with the claim that knowledge is more valuable than mere true belief. Moreover, notice that this is not simply a matter of splitting hairs. There are, after all, many different ways in which knowledge could be more valuable than mere true belief over and above the specifically epistemic axis of evaluation. Indeed, many of the leading proposals for why knowledge is more valuable than mere true belief do not appeal to anything specifically epistemic at all, but rather employ such considerations as the greater practical utility of knowledge (i.e., its greater practical value), the special role that knowledge plays in a life of flourishing (i.e., its greater ethical value), and so on.10 The defender of epistemic value T-monism can appeal to any one of these accounts of the special value of knowledge in order to explain why knowledge is more valuable than mere true belief without in any way contravening her commitment to epistemic value T-monism.

7

Given that the truth of (SC*) is entirely compatible with the falsity of (SC), and thus compatible with the claim that knowledge is in general of greater value than mere true belief, is there anything stopping us from assenting to (SC*)? In particular, is the idea that knowledge is no better than mere true belief from a purely epistemic point of view worrisome, once we realise that it is entirely compatible with the idea that knowledge is more valuable simpliciter than mere true belief? I don’t see why it should be. Or, at least, I don’t see why it would be worrisome unless one had already rejected epistemic value T-monism, in which case the putative counterintuitiveness of (C*) could hardly constitute an argument against epistemic value T-monism. There is a related issue in play when it comes to epistemic value T-monism, which is that opponents of this proposal often characterise the view as being opposed to there being a range of ways in which epistemic standings can be good. The problem is the ‘monism’ part of the thesis, since it has connotations of offering an unduly narrow and restrictive conception of the value of a domain.11 But this is yet another sleight of hand. In one sense, of course, epistemic value T-monism is opposed to a kind of value pluralism about the epistemic domain, since it is opposed to pluralism about the fundamental epistemic goods of that domain. But as we just saw in our discussion of the swamping problem, there is nothing at all to prevent the proponent of epistemic value T-monism from arguing that epistemic standings are valuable in lots of different and interesting ways, such as by having a practical or ethical value. This would explain why knowledge has a special importance to epistemologists, but it wouldn’t be at all in conflict with epistemic value T-monism. Indeed, as I have argued elsewhere, we severely impoverish the debate about the value of epistemic standings like knowledge if we confine our attentions to the swamping problem, for the important issue is not whether knowledge has greater epistemic value than mere true belief, but rather the more general question of whether knowledge (along with other factive epistemic standings, such as understanding) is more valuable than mere true belief.12 In fact, rather than offering a narrow and restrictive conception of the goodness of the epistemic, epistemic value Tmonism is in fact a liberating thesis. If the goodness of the epistemic is to be entirely understood in terms of epistemic goodness, then that blurs out the many and varied ways in which the epistemic can be good. For example, as just noted, epistemic value T-monism is consistent with an account of the ethical value of knowledge. On the epistemic pluralist picture, in contrast, there is the danger that the special value of knowledge in play here would just be treated as another kind of fundamental epistemic value, and the specifically ethical dimension to this value would be lost. Rather than trying to characterise all the ways that the epistemic can be valuable as types of epistemic value, we should instead recognise the distinction between the value of the epistemic and epistemic value. Only in doing so!and in not being so quick to assume that when something

8

epistemic is good, then it is good qua epistemic!can we properly account for the myriad ways in which the epistemic can be good.

3. TRIVIAL TRUTHS A second way in which the truth thesis has been attacked involves the problem that is putatively posed for this thesis by trivial truths. The reasoning goes something like this. If truth is the fundamental epistemic good, then it follows that we should value all truths, even the trivial ones, such as how many grains of sand there are on a certain section of a beach. But clearly we don’t value trivial truths, and hence it follows that truth cannot be the fundamental epistemic good.13 We can express this argument as follows: The Trivial Truths Problem (TP1) If truth is the fundamental epistemic good, then we should value all truths, even the trivial ones. (TP2) We rightly do not value trivial truths. (TC) Truth is not the fundamental epistemic good.

Given the discussion of the previous section we should be immediately suspicious of this train of reasoning. After all, even if it is right that the truth thesis entails that we should value all truths, even the trivial ones, the value in play will be specifically epistemic value. As such, it is entirely compatible with the truth thesis that trivial truths are disvaluable in lots of ways (e.g., are of no practical utility, play no role in a life of flourishing, and so on), and hence are of no value at all (or of vanishingly small value, or of disvalue), all things considered. That is, one could argue that we should modify our formulation of the trivial truths problem as follows: The Trivial Truths Problem* (TP1*) If truth is the fundamental epistemic good, then we should epistemically value all truths, even the trivial ones. (TP2*) We rightly do not epistemically value trivial truths. (TC*) Truth is not the fundamental epistemic good.

Furthermore, one could contend that so construed (TP2*) is false, in that from a purely epistemic point of view we should value all truths, even trivial ones. More precisely, one could argue that all truths, even the trivial ones, have epistemic value (perhaps even pro tanto value) but claim that this is entirely compatible with the obvious truth of (TP2). For when it comes to trivial truths the epistemic value in play is outweighed by other considerations, such that the all things considered value of trivial truths is at most negligible, and typically either non-existent or negative.14 That said, one might be puzzled even by the more modest claim that the most trivial of

9

truths is of epistemic value. I think the source of this puzzlement relates to the relationship between our conception of the fundamental epistemic good and the goal of inquiry that I discussed earlier. Recall that I claimed that the fundamental epistemic good is that to which properly conducted intellectual inquiry is directed, to the extent that an inquiry which was not directed at this end simply wouldn’t count as a properly conducted intellectual inquiry. Here is the worry. If truth is the goal of properly conducted intellectual inquiry, then doesn’t it follow that such an inquiry should be directed at all truths, even the trivial ones? But properly conducted intellectual inquiry is not aimed at all truths but only the significant (i.e., nontrivial) ones. Hence it follows that truth cannot be the goal of properly conducted intellectual inquiry. We can formulate this concern as follows: The Trivial Truths Problem** (TP1**) If truth is the constitutive goal of properly conducted intellectual inquiry, then properly conducted intellectual inquiry aims at all truths equally, even the trivial ones. (TP2**) Properly conducted intellectual inquiry does not aim at all truths equally, since it does not aim at trivial truths. (TC**) Truth is not the constitutive goal of properly conducted intellectual inquiry.

The idea is that the truth thesis, construed as entailing the claim that truth is the constitutive goal of inquiry, entails a commitment to (TP1**), and that it hence flounders on (TP2**). The upshot is that insofar as the truth thesis entails that truth is the constitutive goal of inquiry, then it must be rejected. Whereas arguing that even trivial truths have epistemic value is at least a viable way of responding to the previous formulation of the trivial truths problem, it is hard to see how an analogous claim would help here. For note that the analogous claim would presumably be that a properly conducted intellectual inquiry should be indifferent as to whether the truths it uncovers are trivial or profound, and that hardly seems plausible. Indeed, the exact opposite seems to be the case, in that a properly conducted intellectual inquiry would surely focus on the weighty truths at the expense of the trivial ones. The problem ramifies. For example, insofar as the trivial truths are easier to uncover, as presumably they are, then shouldn’t the efficient intellectual inquirer focus her attentions on them to the exclusion of the profound truths (e.g., shouldn’t the efficient inquiry seek out the information contained in obscure phone books)? Furthermore, presumably some of the trivial truths are the kinds of truths that could be uncovered by pretty much anyone at any time. For instance, consider those trivial truths which are easily gleaned by anyone with an internet connection. Does it follow that insofar as one has an internet connection, then one is under some sort of prima facie cognitive obligation to uncover these truths (i.e., if one has nothing better to do), and thereby promote the

10

epistemic good? Fortunately for the proponent of the truth thesis, however, there is a subtle confusion in play here, one that has been skilfully exposed in recent work by Nick Treanor (forthcominga; forthcomingb). For as he points out, it doesn’t follow from the fact that all truths are epistemically good that all truths are equally epistemically good, such that even trivial and significant truths need to be of equal weight from the perspective of epistemic value. The claim that inquiry should aim at ‘weighty’ truths rather than trivial truths is thus entirely compatible with the idea that truth is the constitutive goal of inquiry. As Treanor points out, the style of argument in play in the trivial truths problem as just formulated is essentially as follows: If inquiry aims at truth, then it aims equally at every truth (even the trivial ones); but inquiry does not aim equally at every truth, and so inquiry does not aim at truth. In order to demonstrate that this reasoning is faulty, Treanor offers us the following example of reasoning which is analogous, and yet which is clearly defective: “If gold mining aimed for gold, it would aim at every piece of gold equally!every piece of gold is equally gold, after all. But gold mining aims for flakes more than for dust, for nuggets more than for flakes, and for great veins more than anything else. So it does not aim for gold.” (Treanor, forthcominga)

In the same way, that inquiry aims at truth does not entail that it aims at all truths equally; instead, it is entirely in keeping with inquiry aiming at truth that it focuses on the important truths at the expense of the trivial ones. The moral is that (TP1**) is false. Now one might initially baulk at this line of reasoning, on the grounds that what differentiates the profound and the trivial truths is not their truth, in that they are both equally true. Accordingly, so the response goes, the analogy with gold breaks down, since we are not being offered a ‘flake’ of truth and a ‘rich vein’ of truth and being asked to choose between them, but simply two individual propositions both of which are true. But as Treanor shows, this way of thinking about the choice that inquiry must make between profound and trivial truths, while superficially compelling, is in fact deeply flawed. Suppose for the sake of argument that one is faced with a straight choice between two propositions, one of which is trivial!regarding the number of grains of sand on a section of a beach, say!and one of which is profound!a fundamental scientific truth, say. While it is undoubtedly correct that there are just two competing true propositions on offer, it does not follow that both propositions offer us an equal amount of truth. A single proposition, after all, can contain compressed within it a whole body of information, and in this sense present us with more truth than a competing proposition which is not so informative, even if both propositions are equally true. This is just what we would expect to be the case when it comes to choosing between

11

a profound truth and a trivial one. A true statement of fundamental science offers us more truth than a true statement regarding grains of sand on the beach. We know more by knowing the true statement of fundamental science than we do by knowing the true statement regarding grains of sand on the beach. Accordingly, if one’s goal is truth, then one will opt for the profound truth every time. Truth is thus more like flakes and nuggets of gold than we realise. To further illustrate this point, consider the following sentences: (1) (2) (3) (4) (5) (6) (7)

The cat sat on the mat. The black cat sat on the mat. The black cat sat on the red mat. The fat black cat sat on the red mat. The fat black cat sat on the square red mat. The fat black cat sat on the heavily-worn square red mat. The fat black cat sat on the large heavily-worn square red mat.

It should be clear that one could continue this series pretty much indefinitely (although the sentences would become rather cumbersome very quickly). Let us stipulate that all these sentences are true. If we ‘weigh’ truth exclusively in terms of whether the proposition expressed by a sentence is true, then we would be committed to saying that all of these sentences contain an equal amount of truth. And yet it is clearly the case that (7) offers us more truth than (1), and that a responsible truth seeker would prefer (7) to (1) for just that reason. As before, we can put this point in terms of knowledge: we know more by knowing (7) than we do by knowing (1). Indeed, a responsible truth seeker would surely prefer (7) even to these two true sentences: (*)

The cat sat on the mat. The cat was black.

But if we are to weigh truth in terms of counting true propositions, then the reverse should be the case, in that (*) will give us two ‘units’ of truth versus (7)’s one. And yet it should be manifest that in knowing (*) we actually know less than by knowing (7). So much the worse for the idea that we should weigh truth by counting true propositions.15 As a final point on this topic, it is worth noting that this claim about weighing truths has a bearing on the issue of whether we should construe the truth thesis in terms of truth or in terms of true beliefs. Opting for the latter has a drawback, in that it counsels one to maximise truth in one’s beliefs, which is naturally rendered as advising one to get as many true beliefs as possible. But this is a mistake. We want to maximise truth, but the upshot of the foregoing is that we cannot measure truth in terms of true propositions, still less by counting how many true beliefs we have in those propositions. Note that the point here is not that proponents of the ‘true belief’ rendering of the truth thesis cannot accommodate any alternative conception of how to weigh truth. The point is rather that the focus on true belief as opposed to truth serves to obscure the

12

fact that it is a mistake to weigh truth by counting true propositions. This is yet another reason to conceive of the fundamental epistemic good as truth rather than true belief.

4. INQUIRY-STOPPERS This brings us to the third and final consideration that we will be looking at which is often offered against the truth thesis. Here is the general line of thinking. From a purely epistemic point of view, it does not suffice to legitimately close inquiry that one has the truth; rather one needs to in addition have properly established that one has the truth.16 Hence, it follows that it is not truth that is the goal of inquiry, but rather whatever epistemic standing that is required in order for one to have properly established that one has the truth. We can express this argument as follows: The Inquiry-Stopper Problem (IP1) If truth is the constitutive goal of properly conducted intellectual inquiry, then truth is what legitimately closes such an inquiry. (IP2) Truth is not what legitimately closes properly conducted intellectual inquiry. (IC) Truth is not the constitutive goal of properly conducted intellectual inquiry.

Let’s start by examining the case for (IP2). Typically!these days anyway!the epistemic standing that is held to be required to legitimately close inquiry is knowledge. Mere true belief is not enough to legitimately close inquiry, so the argument goes, since one can gain a true belief in all manner of inappropriate ways, and one should continue to inquire until the true belief in question is properly epistemically grounded.17 Only then will one have established that one has the truth which one seeks. And an epistemic standing which falls short of knowledge won’t suffice either, and for the same reason, since it wasn’t yet established that one has the truth that one seeks. The obvious case in point here is a Gettierized justified true belief. Of course, when one’s belief is Getterized one takes oneself to have knowledge (and with good reason). That one would think it appropriate to close one’s inquiry at this point is thus evidence in favour of the idea that it is knowledge which legitimately closes inquiry. Furthermore, that one would reopen the inquiry upon discovering the Gettierization, and so discovering that one did not know after all, is also evidence in favour of the claim that knowledge is the ‘inquiry-stopper’. It is only with knowledge, then, that we have an epistemic purchase on the truth that suffices to legitimately close inquiry. Accordingly, it is concluded, knowledge is the goal of properly conducted intellectual inquiry.18 I do not dispute that what legitimately closes inquiry is an epistemic standing, though for what it’s worth I don’t think that knowledge is a very plausible candidate. More precisely, while I

13

grant that inquiry can sometimes be legitimately closed by knowledge, I would maintain that often what is required to legitimately close inquiry is a more elevated epistemic standing!understanding!where this too is conceived of as a factive notion, but one that epistemically demands more of the inquirer than mere knowledge.19 The reason for this is that knowledge can be gained in a fashion such that one lacks understanding of what one knows, and in such cases I think a properly conducted intellectual inquiry will not close but rather continue until understanding is gained. It is, I claim, typically only understanding which sates the curiosity which stimulates inquiry. Consider, for example, someone who wonders what causes the movements of the tides, and who is told, by someone authoritative in this regard, that the movements of the tides are caused by the gravitational pull of the moon on the earth. Our agent thus comes to know that the movements of the tides are caused by the gravitational pull of the moon on the earth. Suppose further, however, that our hero has no real conception of how it could be that the moon’s gravitational pull could have this effect on the movements of the tides. While knowing the answer to the question she has posed, she doesn’t understand it. Should mere knowledge of the answer to her question suffice to close her inquiry in this regard? It seems not. Rather, she should continue to inquire until she understands why cause and effect are related in this way, and so understands the answer she has been given. In this way, I think that it is often the case that it is understanding, rather than knowledge, which legitimately closes inquiry. In any case, let us return to the objection that since what legitimately closes properly conducted intellectual inquiry is not truth, but rather something epistemic like knowledge, it follows that truth is not the goal of inquiry. Given that I grant that what legitimately closes inquiry is not truth but something epistemic (typically understanding), it should be clear that my objection to this line of argument concerns not the antecedent claim in this inference, but rather the inference itself. That is, it is not (IP2) which I reject in the inquiry-stopper problem, but rather (IP1). In particular, I do not think that it follows from the fact that what legitimately closes inquiry is something epistemic, like knowledge or understanding, that it is this epistemic standing, and not truth, which is the goal of inquiry. An analogy will be helpful here. Suppose our goal is to produce a great cup of coffee, and that we engage in certain practices to achieve that end, such as selecting beans, choosing a suitable coffee-making machine, reading up on the art of making a great cup of coffee, and so on. At the point at which our activities produce a great cup of coffee, then those activities will have been successful, in that they will have achieved their goal. But it would not be appropriate to conclude those activities at this point, since in order to do that one must first establish that this goal has been achieved, and that requires tasting (and thereby coming to know) that the coffee is great.20

14

Crucially, however, that knowing that great coffee has been produced is required to legitimately close a great-coffee-directed activity does not entail that this activity was really aimed at the production of knowledge of great coffee after all (or, for that matter, the tasting of great coffee etc.,), rather than just great coffee. The tasting, after all, is just a way of establishing that this activity has successfully reached its conclusion. In the same way, that what legitimately closes inquiry might be knowledge, or some other (factive) epistemic standing like understanding, does not entail that what inquiry is really aimed at is this epistemic standing rather than truth. If one’s inquiry leads one to truth, then at that point one’s inquiry has been successful. But it is only in grasping that the truth has been attained!where this might involve knowledge, understanding, or some other factive epistemic standing, depending on one’s point of view!that one’s inquiry is legitimately closed. There is thus no tension between the idea that truth is the goal of inquiry and that what legitimately closes inquiry is something epistemic like knowledge or understanding. A final note is in order here. I mentioned in §1 that the truth thesis is often formulated in such a way that it is not truth which is the fundamental epistemic good but rather true belief, and at various junctures I have offered considerations against this alternative construal of the truth thesis. Now one reason one might have for opting for a formulation of the truth thesis in terms of true belief is that properly conducted intellectual inquiry is directed at true belief rather than truth. I noted in §1 that I think this is a mistake. For sure, we want inquiry to lead to true belief, but we only care about true belief because we care about truth (just as the aesthete considered above only cares about the aesthetic experience because she cares about beauty). The discussion of this section should have reinforced this idea. First, because even if there is a direct link between what is the goal of inquiry and what legitimately closes inquiry, this won’t lend support to the idea that true belief (as opposed to knowledge, or understanding) is the goal of inquiry. As we have seen, what legitimately closes inquiry, while not truth, is not true belief either. Second, and more importantly, because the discussion of this section reminds us that what is the goal of inquiry and what legitimately closes inquiry is not the same thing anyway. So even if it were right that it is true belief which legitimately closes inquiry, it wouldn’t follow that it is true belief and not truth which is the goal of inquiry.

15

5. CONCLUDING REMARKS I have considered three prominent reasons for rejecting the truth thesis and found them all wanting. Of course, this in itself doesn’t show that the truth thesis is correct, not least because there may be further arguments out there against the truth thesis which are more compelling than the ones I have considered. Moreover, I haven’t offered any positive defence of truth thesis. But my goal here has not been to demonstrate that the truth thesis is correct. Instead, I have been concerned merely to show that this is a thesis that is far more plausible than it is given credit in contemporary epistemology. In particular: (i) it can accommodate the widespread contention that knowledge is more valuable than mere true belief; (ii) it can accommodate, perhaps even better than alternative views, a pluralism about the good of the epistemic (even though it is incompatible with pluralism about the epistemic good); (iii) it is not affected by the problem putatively posed by trivial truths, since that problem trades on a dubious conception of how one should ‘weigh’ truth; and (iv) it is entirely compatible with the idea that what legitimately closes inquiry is not truth but rather a particular kind of factive epistemic relationship to the truth, such as knowledge and understanding.21

16

REFERENCES Brady, M. S., & Pritchard, D. H. (eds.) (2003). Moral and Epistemic Virtues, Oxford: Blackwell. David, M. (2001). ‘Truth as the Epistemic Goal’, Knowledge, Truth, and Duty: Essays on Epistemic Justification, Virtue, and Responsibility, (ed.) M. Steup, 151-69, Oxford: Oxford University Press. !! (2005). ‘Truth as the Primary Epistemic Goal: A Working Hypothesis’, Contemporary Debates in Epistemology, (eds.) E. Sosa & M. Steup, 296-312, Oxford: Blackwell. DePaul, M. (2001). ‘Value Monism in Epistemology’, Knowledge, Truth, and Duty: Essays on Epistemic Justification, Virtue, and Responsibility, (ed.) M. Steup, 170-86, Oxford: Oxford University Press. Elgin, C. (1996). Considered Judgement, Princeton, NJ: Princeton University Press. !! (2004). ‘True Enough’, Philosophical Issues 14, 113-31. !! (2009). ‘Is Understanding Factive?’, Epistemic Value, (eds.) A. Haddock, A. Millar & D. H. Pritchard, 322-30, Oxford: Oxford University Press. Geach, P. T. (1956). ‘Good and Evil’, Analysis 17, 32-42. Gelfert, A. (2011). ‘Expertise, Argumentation, and the End of Inquiry’, Argumentation 25, 297-312. Goldman, A. (1999). Knowledge in a Social World, Oxford: Oxford University Press. !! (2002). ‘The Unity of the Epistemic Virtues’, in his Pathways to Knowledge: Private and Public, 51-72, Oxford: Oxford University Press. Goldman, A., & Olsson, E. J. (2009). ‘Reliabilism and the Value of Knowledge’, Epistemic Value, (eds.) A. Haddock, A. Millar & D. H. Pritchard, 19-41, Oxford: Oxford University Press. Greco, J. (2009). ‘The Value Problem’, Epistemic Value, (eds.) A. Haddock, A. Millar & D. H. Pritchard, 313-21, Oxford: Oxford University Press. Grimm, S. (2006). ‘Is Understanding a Species of Knowledge?’, British Journal for the Philosophy of Science 57, 515-35. —— (Forthcoming). ‘Understanding as Knowledge of Causes’, Virtue Scientia: Virtue Epistemology and Philosophy of Science, (ed.) A. Fairweather, Dordrecht, Holland: Springer. Jones, W. (1997). ‘Why Do We Value Knowledge?’, American Philosophical Quarterly 34, 423-40. Kappel, K. (2010). ‘On Saying that Someone Knows: Themes from Craig’, Social Epistemology, (eds.) A. Haddock, A. Millar & D. Pritchard, 69-88, Oxford: Oxford University Press. Kelp, C. (2011). ‘What’s the Point of ‘Knowledge’ Anyway?’, Episteme 8, 53-66. Kvanvig, J. (2003). The Value of Knowledge and the Pursuit of Understanding, Cambridge: Cambridge University Press. !! (2005). ‘Truth is not the Primary Epistemic Goal’, Contemporary Debates in Epistemology, (eds.) E. Sosa & M. Steup, 285-96, Oxford: Blackwell. !! (2008). ‘Pointless Truth’, Midwest Studies in Philosophy 32, 199-212 —— (2009). ‘The Value of Understanding’, Epistemic Value, (eds.) A. Haddock, A. Millar & D. H. Pritchard, 95-112, Oxford: Oxford University Press. !! (2010). ‘The Swamping Problem Redux: Pith and Gist’, Social Epistemology, (eds.) A. Haddock, A. Millar & D. H. Pritchard, 89-111, Oxford: Oxford University Press. !! (2011). ‘Millar on the Value of Knowledge’, Proceedings of the Aristotelian Society (suppl. vol.) 85, 83-99. !! (2012). ‘Curiosity and a Response-Dependent Account of the Value of Understanding’, Knowledge, Virtue, and Action, (eds.) T. Henning & D. Schweikard, London: Routledge. Millar, A. (2011). ‘Why Knowledge Matters’, Proceedings of the Aristotelian Society (suppl. vol.) 85, 63-81. Pritchard, D. H. (2007). ‘Recent Work on Epistemic Value’, American Philosophical Quarterly, 44, 85110. !! (2009). ‘Knowledge, Understanding and Epistemic Value’, Epistemology (Royal Institute of Philosophy Lectures), (ed.) A. O’Hear, 19-43, Cambridge: Cambridge University Press.

17

!! (2011). ‘What is the Swamping Problem?’, Reasons for Belief, (eds.) A. Reisner & A. SteglichPetersen, 244-59, Cambridge: Cambridge University Press. !! (2012). ‘Anti-Luck Virtue Epistemology’, Journal of Philosophy 109, 247-79. !! (Forthcominga). ‘Knowledge and Understanding’, Virtue Scientia: Virtue Epistemology and Philosophy of Science, (ed.) A. Fairweather, Dordrecht, Holland: Springer. !! (Forthcomingb). ‘Veritism and Epistemic Value’, Alvin Goldman and His Critics, (eds.) H. Kornblith & B. McLaughlin, Oxford: Blackwell. Pritchard, D. H., Millar, A., & Haddock, A. (2010). The Nature and Value of Knowledge: Three Investigations, Oxford: Oxford University Press. Pritchard, D. H., & Turri, J. (2011). ‘Knowledge, the Value of’, Stanford Encyclopaedia of Philosophy, (ed.) E. Zalta, http://plato.stanford.edu/entries/knowledge-value/. Riggs, W. (2002a). ‘Beyond Truth and Falsehood: The Real Value of Knowing that P’, Philosophical Studies 107, 87-108. !! (2002b). ‘Reliability and the Value of Knowledge’, Philosophy and Phenomenological Research 64, 79-96. Shah, N. (2003). ‘How Truth Governs Belief’, Philosophical Review 112, 447-83. Shah, N., & Velleman, D. (2005). ‘Doxastic Deliberation’, Philosophical Review 114, 497-534. Sosa, E. (2001). ‘For the Love of Truth?’, Virtue Epistemology: Essays on Epistemic Virtue and Responsibility, (eds.) A. Fairweather & L. Zagzebski, 49-62, Oxford: Oxford University Press. Treanor, N. (Forthcominga). ‘The Measure of Knowledge’, Noûs. !! (Forthcomingb). ‘Trivial Truths and the Aim of Inquiry’, Philosophy and Phenomenological Research. Wedgwood, R. (2002). ‘The Aim of Belief’, Philosophical Perspectives 16, 268-97. Williamson, T. (2000). Knowledge and Its Limits, Oxford: Oxford University Press. Zagzebski, L. (2001). ‘Recovering Understanding’, Knowledge, Truth, and Obligation: Essays on Epistemic Justification, Virtue, and Responsibility, (ed.) M. Steup, 235-52, Oxford: Oxford University Press. —— (2003). ‘The Search for the Source of the Epistemic Good’, Metaphilosophy 34, 12-28; and reprinted in Brady & Pritchard (2003), 13-28.

18

NOTES 1 See David (2001, 151-2) for a long list of quotations illustrating how widely held this view once was in epistemology. The list of contemporary epistemologists that he quotes includes William Alston, Laurence Bonjour, Roderick Chisholm, Richard Foley, Alvin Goldman, Keith Lehrer, Paul Moser, Alvin Plantinga, and Ernest Sosa. 2 For two recent high-profile attacks on the idea of truth being the fundamental epistemic good, see DePaul (2001; cf. David 2001) and Kvanvig (2005; cf. David 2005). 3 See, for example, Pritchard, Millar & Haddock (2010, ch. 1) and Pritchard (2011). Following Goldman (e.g., 1999; 2002; Goldman & Olsson 2009), this proposal is also known in the literature as veritism. 4 It is worth bearing in mind here Geach’s (1956) distinction regarding predicative and attributive adjectives. In particular, from ‘x is a big flea’ it does not follow that ‘x is a flea’ and ‘x is big’, since the claim being made is only the attributive claim that x is big for a flea. (Compare: ‘x is a red flea’). Similarly, in claiming that something is, say, epistemically valuable, one is on the face of it just making the attributive claim that the item in question is valuable from a specifically epistemic point of view, and not the predicative claim that the item is both epistemic and valuable. I discuss the topic of epistemic value in a number of places, but see especially Pritchard (2007; 2011), Pritchard, Millar & Haddock (2010, ch. 1), and Pritchard & Turri (2011). 5 See, for example, David (2001; 2005). 6 Here’s another reason to prefer true belief as the fundamental epistemic good to mere truth, pressed to me by John Turri. If one goes for the latter thesis, then what is to stop someone merely desiring the truth or hoping that certain propositions are true, while never actually seeking out true beliefs? But that would be akin to someone regarding beauty as the fundamental aesthetic good and yet eschewing the opportunity of actually experiencing beauty for no other reason than that she prefers to merely desire such beauty from afar, or hope that there is beauty in the world. That doesn’t look like a coherent state of affairs. Moreover, I take it as an advantage of the view that it is truth that is the fundamental epistemic good that it can account for the epistemic goodness of propositional attitudes other than belief. That is, the epistemically good agent will surely be someone who not only seeks out true beliefs and avoids false ones, but who also has various other propositional attitudes too, such as desiring that others should form their beliefs with similar epistemic care. 7 One might also try to motivate the truth thesis by appealing to the relationship between belief and truth. In particular, many commentators have suggested that truth is the telos of belief. For reasons of space, I will not be considering this issue here. I briefly discuss its relevance to questions of epistemic value in Pritchard (2011, §1). For more on the idea that belief is in some sense ‘aimed’ at truth, see Wedgwood (2002), Shah (2003), and Shah & Velleman (2005). 8 The coffee example is due to Zagzebski (2003). 9 For some of the key statements of the swamping problem, see Jones (1997), Swinburne (1999), Riggs (2002a; 2002b), Kvanvig (2003; 2010), and Zagzebski (2003). Note, however, that, at least in early work on this issue, the swamping problem was usually thought of as impacting not on a particular conception of epistemic value but rather on specific theories of knowledge (usually reliabilism). I think it is clear in the contemporary literature that the problem is now recognized as being properly understood as being directed against a certain theory of epistemic value. I should note that in earlier work I understood the swamping problem as posing a reductio for epistemic value T-monism too. See Pritchard, Millar & Haddock (2010, ch. 1) and Pritchard (2011). As the present article suggests, my view has shifted on this point. See Pritchard (forthcomingb) for an overview of how I respond to the swamping problem now. 10 For a recent example of a defence of the special value of knowledge that is of the first sort, see Millar (2011); for a recent example of a defence of the special value of knowledge that is of the second sort, see Greco (e.g., 2009). 11 This is a guiding theme of Kvanvig’s work on this topic, for example, which has been extremely influential in contemporary epistemology. See, for example, Kvanvig (2003; 2005). See also DePaul (2001). 12 See Pritchard, Millar & Haddock (2010, ch. 1) for a detailed defence of why we should disentangle the general question of the value of the epistemic from the specific problem posed by the swamping problem. 13 Versions of this general line of argument abound in the contemporary epistemological literature. For a sample of high-profile endorsements of this reasoning, see DePaul (2001), Sosa (2001), and Goldman (2002). (The ‘sand’ example is due to Sosa). 14 For a defence of this general line of argument, see Kvanvig (2008). 15 For a subtle discussion of how we should weigh truth once we move away from the (clearly false) model on which we simply count true propositions, see Treanor (forthcominga). 16 The caveat that we are talking about what closes inquiry from a purely epistemic point of view is of course vital, since there might be all kinds of non-epistemic reasons to legitimately close an inquiry (lack of time, opportunity cost, and so on). In what follows, I will take this caveat as implicit. 17 Of course, since we’re here talking specifically about a ‘properly conducted’ intellectual inquiry, it’s actually not obvious that such an inquiry could lead to mere true belief, but let’s set that issue to one side. 18 For a particularly clear and recent statement of this line of objection to the truth thesis, one that explicitly substitutes knowledge as the goal of inquiry for truth, see Millar (2011), to which Kvanvig (2011) is a response. Here is Millar:

19

“Suppose that you are inquiring into whether something is so. Your aim is to find out whether it is so. Since finding out is nothing less than coming to know, what you aim for is knowledge.” (Millar 2011, 63) Millar goes on to say that this conception of the goal of inquiry is compatible with inquiry aiming at truth, but it is clear from how he clarifies what he means by this remark that he views the goal of inquiry as being knowledge of the truth rather than truth itself. For example, he remarks that “the aim [of inquiry] is to grasp the truth, and knowledge is the form that such grasping takes if the aim is achieved.” (Millar 2011, 63) Note that Millar draws heavily on the socalled ‘knowledge-first’ programme in epistemology as defended by Williamson (2000), though it’s unclear whether Williamson himself would endorse the idea that knowledge rather than true belief is the goal of inquiry. For some recent defences of the claim that it is knowledge which legitimately closes inquiry, see Kappel (2010), Gelfert (2011), and Kelp (2011). 19 Actually, this isn’t quite right, since I also hold that understanding can sometimes be epistemically less demanding than knowledge, in that it is compatible with a form of epistemic luck that knowledge isn’t compatible with (though even then understanding is still a factive notion). For more on my conception of understanding, and how it relates to knowledge, see Pritchard (2009; forthcominga) and Pritchard, Millar & Haddock (2010, ch. 4). For some key alternative views on the epistemology of understanding, see Elgin (1996; 2004; 2009), Zagzebski (2001), Kvanvig (2003; 2009; 2012), and Grimm (2006; forthcoming), 20 It needn’t be you who tastes it of course; indeed, it might be better for a coffee aficionado to taste it and confirm that it’s great coffee. 21 Thanks to Allan Hazlett and Nick Treanor for helpful discussions of topics related to this paper. Special thanks to Jon Matheson and John Turri who each read and commented on an earlier incarnation. An earlier version of this paper was presented in June 2012 at the ‘Dimensions of Normativity’ conference at Goethe University, Frankfurt. I am grateful to the audience on this occasion, and especially Marcus Willaschek, who commented on my paper at length.