From Laurence BonJour, The Structure of Empirical Knowledge, 26-33

2.2 The varieties of foundationalism

The common thesis of all versions of empirical foundationalism is that some empirical beliefs have a degree of noninferential epistemic justification, justification that does not derive from other empirical beliefs in a way which would require those beliefs to be antecedently justified. One way of distinguishing specific versions of foundationalism, though not in the end the most revealing, is in terms of the precise degree of non‑inferential epistemic justification which these "basic beliefs" are held to possess. In this regard there are three main views. The most obvious interpretation of the foundationalist response to the regress problem yields a view which I will call moderate foundationalism. According to moderate foundationalism, the noninferential warrant possessed by basic beliefs is sufficient by itself to satisfy the adequate‑justification condition for knowledge. Thus on this view a basic belief, if true, is automatically an instance of knowledge (assuming that Gettier problems do not arise) and hence fully acceptable as a premise for the justification of further empirical beliefs. By virtue of their complete justificatory independence from other empirical beliefs, such basic beliefs are eminently suitable for a foundational role.

Moderate foundationalism, as the label suggests, represents a relatively mild version of foundationalism. Historical foundationalist positions typically make stronger and more ambitious claims on behalf of their chosen class of basic beliefs. Thus such beliefs have been claimed to be not just adequately justified, but also infallible, certain, indubitable, or incorrigible. Unfortunately, however, the meanings of these four terms have very rarely been made clear. It is infallibility which is most obviously relevant to epistemological concerns. To say that a specified sort of basic belief is infallible is to say that it is impossible for a person to hold such a belief and for it nonetheless to be mistaken, where the impossibility might be either logical or nomological. Historical versions of foundationalism have virtually always been interested in logical infallibility, in part at least because a claim of nomological infallibility would presumably depend for its justification on empirical evidence for the law of nature in question, so that a belief whose justification depended on such a claim could not be basic. If a basic belief which is actually held is logically infallible, then it is of course necessary that it be true. Thus it is clear that the logical infallibility of such a belief, if known by the believer, provides the best possible epistemic justification for accepting it.

In contrast, the relevance of certainty, indubitability, and incorrigibility to issues of epistemic justification is much less clear insofar as these concepts are understood in a way which makes them distinct from infallibility. Certainty is most naturally interpreted as pertaining to one's psychological state of conviction, or perhaps to the status of a proposition as logically or metaphysically necessary, with neither of these interpretations having any immediate epistemic import. Indubitability should have to do with whether a proposition can be doubted, incorrigibility with whether a belief in it can be corrected, and in both cases the epistemic significance is again not clear, assuming that the reason that a belief possesses such a status is not that it is infallible or perhaps nearly so).

Thus the interesting claim for my purposes is the claim that basic beliefs are logically infallible. And in fact such a claim seems to be what was intended by most of the historical proponents of foundationalism in employing these terms, even though, for largely accidental reasons, they often couched their claims in these other ways. Since the justification resulting from known logical infallibility surely is adequate for knowledge, the view which advances this thesis is a subspecies of moderate foundationalism, what I will here call strong foundationalism. Most historical discussions of foundationalism and even many quite recent ones, both pro and con, have focused almost exclusively on strong foundationalism. This is, however, very unfortunate, for two correlative reasons.

First, there are a number of persuasive arguments which seem to show that, whether or not foundationalism in general is acceptable, strong foundationalism is untenable. Here I will mention only one such argument, due in its essentials to Armstrong, which has the virtue of general applicability. Consider the state of affairs of a person A having a certain allegedly infallible basic empirical belief B; call this state of affairs S1. B will have as its content the proposition that some empirical state of affairs S exists. Now it seems to follow from the logic of the concept of belief that S1 and S2 must be distinct states of affairs. Beliefs may of course be about other beliefs, but beliefs cannot somehow be directly about themselves. My belief that I believe that P is distinct from my belief that P; the content of the latter is simply the proposition that P, while the content of the former is the different and more complicated proposition that I believe that P. And thus it would seem to be logically quite possible for S1 to occur in the absence of S2, in which case, of course belief B would be false. A proponent of logical infallibility must claim that this is, in the cases he is interested in, not logically possible, but it is hard to see what the basis for such a claim might be, so long as S1 andS2 are conceded to be separate states of affairs.

Second, and more important for our present purposes, strong foundationalism, even if it were otherwise acceptable, seems to constitute philosophical overkill relative to the dialectical requirements of the foundationalist position. Nothing about the foundationalist response to the regress requires that basic beliefs be more than adequately justified. (Indeed, as will be explained shortly, many recent foundationalists believe that an even weaker claim is sufficient.) There might of course be other reasons for requiring that basic beliefs have some more exalted epistemic status or for thinking that in fact they do. But until such reasons are provided (and I doubt very much that any can be), the question of whether basic beliefs are infallible will remain a relatively unimportant issue. And hence discussions of foundationalism, both pro and con, which concentrate on this stronger but inessential claim are in serious danger of bypassing the main issue: whether moderate foundationalism is acceptable.

Thus an adequate consideration of foundationalism need concern itself with nothing stronger than moderate foundationalism. Indeed, many recent proponents of foundationalism have felt that even moderate foundationalism goes further than is necessary with regard to the degree of intrinsic or noninferential justification ascribed to basic beliefs. Their alternative is a view which may be called weak foundationalism, according to which basic beliefs possess only a very low degree of epistemic justification on their own, a degree of justification insufficient by itself either to satisfy the adequate‑justification condition for knowledge or to qualify them as acceptable justifying premises for further beliefs. Such beliefs are only "initially credible," rather than fully justified. Such a view was first advocated explicitly by Bertrand Russell and, somewhat later, by Nelson Goodman; recent advocates have included Roderick Firth, Israel Scheffler, and perhaps Nicholas Rescher.

Weak foundationalism is a version of foundationalism because it holds that there are basic beliefs having some degree, though a relatively low one, of noninferential epistemic justification. But weak foundationalism differs substantially from historically more orthodox versions of foundationalism. In particular, the weak foundationalist response to the regress problem (which is usually not made particularly clear) must differ significantly from that of the moderate foundationalist. The weak foundationalist cannot say, as does the moderate foundationalist, that the regress of justifying arguments simply comes to an end when basic beliefs are reached. For the weak foundationalist's basic beliefs are not adequately justified on their own to serve as justifying premises for everything else. The weak foundationalist solution to this problem is to attempt to augment the justification of both basic and nonbasic beliefs by appealing to the concept of coherence. Very roughly, if a suitably large, suitably coherent system can be built, containing a reasonably high proportion of one's initially credible basic beliefs together with nonbasic beliefs, then it is claimed, the justification of all the beliefs in the system, basic and nonbasic, may be increased to the point of being adequate for knowledge, where achieving high enough degree of coherence may necessitate the rejection of some of one's basic beliefs.

Thus understood, weak foundationalism represents a kind of hybrid between moderate foundationalism and the coherence theories mentioned earlier and is often thought to embody the virtues of both and the vices of neither. The weak foundationalist does appear to have an adequate answer to the first of the standard objections to the coherence theory set forth earlier: the choice between equally coherent systems is made by determining which system contains a greater quantity (however exactly this is to be measured) of these basic beliefs. And while it is much less clear how the other two objections to coherence theories are to be answered, especially the second, the weak foundationalist seems at least to have a good deal more room for maneuver. Finally, it is usually thought that weak foundationalism, by virtue of making a weaker claim on behalf of the foundational beliefs, is more defensible than moderate foundationalism.

This last suggestion is, however, very dubious. Although there may well be certain objections to moderate foundationalism which weak foundationalism can avoid, the most fundamental and far‑reaching objection to foundationalism‑‑namely that there is no way for an empirical belief to have any degree of warrant which does not depend on the justification of other empirical beliefs (to be considered at the end of this chapter)‑‑applies just as much to weak foundationalism as to moderate foundationalism. Moreover, weak foundationalism faces at least one serious objection which does not apply to moderate foundationalism, namely that the underlying logic of the weak foundationalist's account has never been made adequately clear. The basic idea is that an initially low degree of justification can somehow be magnified or amplified by coherence, to a degree adequate for knowledge. But how is this magnification or amplification supposed to work? How can coherence, not itself an independent source of justification on a foundationalist view, justify the rejection of some initially credible beliefs and enhance the justification of others? Weak foundationalism seems to presuppose some kind of tradeoff between retention of basic beliefs and increased coherence, but neither the precise exchange rate for this tradeoff nor its underlying rationale are at all obvious. Without further amplification, on this last issue especially, it is very hard to assess the view seriously.

For these reasons, weak foundationalism will be set aside, and my critical discussion of foundationalism will be formulated in application to moderate foundationalism. As will become clear, the key test for any version of foundationalism is whether it can solve the regress problem which motivates its very existence without resorting to essentially ad hoc stipulation. The distinction between the various ways of meeting this challenge both cuts across and is more basic than that between moderate and weak foundationalism. This being so, it will suffice to concentrate here on moderate foundationalism, leaving the application of the discussion to weak foundationalism largely implicit.

2.3 A basic problem for foundationalism

The fundamental concept of moderate foundationalism, as of empirical foundationalism generally, is the concept of a basic empirical belief. It is by appeal to basic beliefs that the threat of an infinite regress is to be avoided and empirical knowledge given a secure foundation. But a new problem now arises: how can there be any empirical beliefs which are thus basic? For although this has often been overlooked, the very idea of an epistemically basic empirical belief is more than a little paradoxical. On what basis is such a belief supposed to be justified, once any appeal to further empirical premises is ruled out? Chisholm's theological analogy, cited earlier, is most appropriate: a basic empirical belief is in effect an epistemological unmoved (or self‑moved) mover. It is able to confer justification on other beliefs, but, in spite of being empirical and thus contingent, apparently has no need to have justification conferred on it. But is such a status any easier to understand in epistemology than it is in theology? How can a contingent, empirical belief impart epistemic "motion" to other empirical beliefs unless it is itself in "motion"? (Or, even more paradoxically, how can such a belief epistemically "move" itself?) Where does the noninferential justification for basic empirical beliefs come from?

This difficulty may be developed a bit by appealing to the account of the general concept of epistemic justification which was presented in Chapter I. I argued there that the fundamental role which the requirement of epistemic justification serves in the overall rationale of the concept of knowledge is that of a means to truth; and accordingly that a basic constraint on any account of the standards of justification for empirical knowledge is that there be good reasons for thinking that following those standards is at least likely to lead to truth. Thus if basic beliefs are to provide a secure foundation for empirical knowledge, if inference from them is to be the sole basis upon which other empirical beliefs are justified, then that feature, whatever it may be, by virtue of which a particular belief qualifies as basic must also constitute a good reason for thinking that the belief is true. If this were not so, moderate foundationalism would be unacceptable as an account of epistemic justification.

This crucial point may be formulated a bit more precisely, as follows. If we let F represent the feature or characteristic, whatever it may be, which distinguishes basic empirical beliefs from other empirical beliefs, then in an acceptable foundationalist account a particular empirical belief B could qualify as basic only if the premises of the following justificatory argument were adequately justified:

(1) B has feature F.

(2) Beliefs having feature F are highly likely to be true.

Therefore, B is highly likely to be true.

If B is to actually be basic, then presumably premise (1) would have to be true as well, but I am concerned here only with what would have to be so for it to be reasonable to accept B as basic and use it to justify other beliefs.

Clearly it is possible that at least one of the two premises of the argument might be justifiable on a purely a priori basis, depending on the particular choice of F. It does not seem possible, however, that both premises might be thus justifiable. B is after all, ex hypothesi, an empirical belief, and it is hard to see how a particular empirical belief could be justified on a purely a priori basis. Thus we may conclude, at least provisionally, that for any acceptable moderate foundationalist account, at least one of the two premises of the appropriate justifying argument will itself be empirical.

The other issue to be considered is whether, in order for B to be justified for a particular person A (at a particular time), it is necessary, not merely that a justification along the above lines exist in the abstract, but also that A himself be in cognitive possession of that justification, that is, that he believe the appropriate premises of forms (1) and (2) and that these beliefs be justified for him. In Chapter I and the previous section, I argued tentatively that such cognitive possession by the person in question is indeed necessary, on the grounds that he cannot be epistemically responsible in accepting the belief unless he himself has access to the justification; for otherwise, he has no reason for thinking that the belief is at all likely to be true. No reason for questioning this claim has so far emerged.

But if all this is correct, we get the disturbing result that B is not basic after all, since its justification depends on that of at least one other empirical belief. It would follow that moderate foundationalism is untenable as a solution to the regress problem‑and an analogous argument would show weak foundationalism to be similarly untenable.

It will be helpful in the subsequent discussion to have available a slightly more explicit statement of this basic antifoundationalist argument:

(1) Suppose that there are basic empirical beliefs, that is, empirical beliefs (a) which are epistemically justified, and (b) whose justification does not depend on that of any further empirical beliefs.

(2) For a belief to be epistemically justified requires that there be a reason why it is likely to be true.

(3) For a belief to be epistemically justified for a particular person requires that this person be himself in cognitive possession of such a reason.

(4) The only way to be in cognitive possession of such a reason is to believe with justification the premises from which it follows that the belief is likely to be true.

(5) The premises of such a justifying argument for an empirical belief cannot be entirely a priori; at least one such premise must be empirical.

Therefore, the justification of a supposed basic empirical belief must depend on the justification of at least one other empirical belief, contradicting (I); it follows that there can be no basic empirical beliefs.

In order to reject the conclusion of this argument, as he obviously must, the foundationalist must reject one or more of the premises. But premise (1) is merely a statement of the basic foundationalist thesis, and premise (2) has, I will assume, been adequately justified in Chapter 1. Moreover, as has also been already discussed, there seems to be no way to plausibly defend the rejection of premise (5). Thus a tenable version of foundationalism must apparently reject either premise (3) or premise (4). Both of these approaches have in fact been attempted. I will conclude the present chapter with an initial, brief sketch of these two alternatives.

First. The basic gambit of many recent foundationalist positions is to reject premise (3) of the foregoing argument by claiming in effect that although it is indeed necessary in order for a belief to be justified, and a fortiori for it to be basic, that a justifying argument be in a certain sense available in the situation, it is not necessary that the person for whom the belief is basic know, or justifiably believe, or even believe at all, the premises of such an argument. Indeed, it is not necessary that anyone know or justifiably believe those premises. Instead, for basic beliefs at least, it is sufficient that the premises for some favored variety of such argument merely be true, whether or not anyone realizes in any way that this is so. D. M. Armstrong, one of the leading proponents of this view, calls it externalism, because what ultimately justifies such a belief is some appropriate set of facts which are (in the most typical case) external to the believer's conception of the situation, and I will adopt his label here. Other proponents of externalism include Alvin Goldman, William Alston, and Fred Dretske. I will consider externalist views in Chapter 3, focusing initially on Armstrong's version.

Second. The older and more traditional foundationalist view concedes, implicitly at least, that in order for a belief to be basic it is necessary both that a justification of the sort sketched above exist and that the person holding the belief be in some sense in cognitive possession of that justification. But this view rejects premise (4) of the antifoundationalist argument by arguing that the believer's cognitive grasp of the premises required for that justification does not involve further empirical beliefs, which would then themselves require justification. What is involved is rather cognitive states of a more rudimentary type which do not themselves require justification, despite having the capacity to confer justification on beliefs. It is these more rudimentary states which are thus, according to this position, the ultimate source of epistemic justification; although basic beliefs are indeed the most basic beliefs, they are thus not the most basic cognitive states. The basic states in question are described as intuitions, immediate apprehensions, or direct awarenesses; the objects of such states are usually said to be given or presented. I will discuss this view, the doctrine of the empirically given, in Chapter 4. (As a result of this discussion, a third possible position will emerge, one which is at least debatably a version of foundationalism; that position, which might be said, though with important qualifications, to attempt a purely a priori justification of basic beliefs, will be considered in the final section of Chapter 4.)