RESPONSIBLE BELIEVING

S. Joel Garver

CHAPTER THREE

Believing as One Ought:
Well-Established Practices

Section IV

We will make a rather in depth consideration of Alston's and others' doxastic practice approaches to epistemology in the following chapter. For now let us be content with a thumbnail sketch.

The thrust of Alston's doxastic approach is to urge the practical rationality of "socially established" practices--those that are engaged in by a significant portion of a population. It is, then, doxastic practices that we can take to be well-established that can also be taken to be justification-conferring. What is it for a practice to be well-established and how are we to determine that?

First, we can eliminate as well-established any practice that produces "sufficiently extensive and persistent internal contradictions in [its] output" (1991:171). Second, in some cases--where internal contradiction may be absent--there will be "a massive and persistent inconsistency between the outputs of two [or more] practices" (1991:171). In this case we can only assume that at least one of the two or more inconsistent practices is unreliable--but how do we decide which? Alston suggests that we "should give preference to the more firmly established practice", which includes being more widely accepted, more structured, more important to us, more innate, more difficult to give up, and more obviously true (1991:171). Opportunity to examine and critique these points will be forthcoming.

Well-established practices (and, no doubt, even quite a few not-so-well-established practices) can provide significant self-support. Since self-support "can be enjoyed only by modes of belief-formation that are actually utilized" (because they must do something to support themselves), only practices can have significant self-support and not abstract grounds or principles which do nothing. Alston concludes, then, that a firmly established practice is rationally engaged in [1] "unless the total output of all our firmly established doxastic practices sufficiently indicates its unreliability" or, in other words, [2] "provided it and its output cohere sufficiently with other firmly established doxastic practices and their output" (1991:175).

What has Alston shown in this discussion? He has not shown that the rationality of a practice entails reliability or that it does, in fact, confer justification. Rather he has shown that the rationality of a practice entails the (practical) rationality of supposing the practice to be reliable. This, however, is sufficient for showing that the outputs of such a practice are reasonably taken as prima facie justified. It provides an outline of how the defense of one's beliefs against criticism functions and so how justifying one's beliefs can be rational and successful. It also provides a basis upon which practical ascriptions of justification can be reasonably made. Those practices that we can take to be well-established can also be rationally taken to be reliable and justification-conferring. We also know how to show that to be the case and are thus able to justify our beliefs.

It turns out, then, that while coherentism may be an incorrect view of what it is to be justified in a particular belief, it may well be that our main guiding paradigm--as to what beliefs are to be reasonably and practically taken as justified--is an indirectly and negatively coherentist one. It is indirect in that the coherentism does not concern the actual possession of justification by particular beliefs, but only the rationality of taking well established doxastic practices to be justification-conferring. It is negative in that coherence between practices is not necessary for their being well-established, but only that their be no incoherence. Positive coherence--and especially mutual support--contributes to being well-established, more so in cases of competing, conflicting practices. Coherence considerations, then, will loom large when we attempt to defend beliefs, especially with a second-line defense. Even in a first-line defense, where a practice is mutually shared and approved, that judgment doubtless arises because the practice in question has shown little evidence of not being well-established.

Furthermore, the doxastic practice approach allows us to think differently about the relations and distinctions between a belief being such that it has justification, its being rational to take as justified, and justifying a belief or making practical determinations and ascriptions of a belief's justification (we might also consider showing whether or not a belief has justification as distinct from justifying; see the previous section). When we think of a subject's belief as being justified, we are thinking primarily about conditions that are rather abstract or internal to the "wiring" and store of beliefs of the subject:

This last condition begins to bleed over into the issue of justifying a belief (and showing justification; it is what underlies what was specified as the "first line of defense," above, i.e., indicating grounds). In fact, it is precisely for this reason that Alston includes the last condition in his analysis of justification. Part of our concept of justification, he argues, draws on "the concept of what would have to be specified to carry out a successful justification" (1989:236).

While recognizing and understanding Alston's point here, my own tendency is in a more reliablist direction regarding justification, at least in a wide range of cases. I would not, therefore, include accessibility to grounds among the conditions needed for justification or as part of the concept of justification. I do think that grounds are in fact the sorts of things accessible to the subject when a belief has justification --but this is, for me, a matter of fact and not a conceptual inclusion. Whether or not I am correct in my reliablist tendency is not of great import, for the reasons that underlie it are independent of the conclusion. The reasons are, however, important for the distinction between having justification, the rationality of taking a belief to have justification, and justifying belief (and thus, I think, between a conceptual analysis of justification and a doxastic practice approach).

As already noted, what conditions are included in a belief's having justification are, by and large, internal to the believing subject. On the other hand, we have invoked the doxastic practice approach to discuss considerations of rationality and, from that, to deal with cases of defending justification, showing when a belief is rationally taken as justified, and making practical determinations and ascriptions of justification. Thus the doxastic practice approach, standards of rationality, and the activities of justifying belief, while not excluding what is internal to the subject, also are fundamentally social in nature and range well beyond what is internal to the subject.

There is significant fallout from this way of distinguishing matters. Most importantly for our purposes is the fact the notions of prima facie justification and overriders shift when one moves from the context of having justification (or the concept of justification) to the context of justifying belief and the doxastic practice approach to rationality. According to Alston's conception of justification a belief is unqualifiedly justified only if it is formed on the basis of grounds that are adequate (i.e., sufficiently indicative of the truth of the belief) and there are no overriders to the contrary. A belief has prima facie justification if it is based on adequate grounds apart from the presence of overriders.

Alston distinguishes between two types of overriders: [1] rebutters and [2] underminers. Rebutters are reasons accessible to the subject for supposing the prima facie justified beliefs to be false. For instance, if I believe there to be something burning in the house on the basis of a strong smell of smoke, that belief can be rebutted by the fact that I can see no signs of fire or smoke and can feel no heat. These serve as reasons for supposing my belief to be false. Underminers, on the other hand, are reasons accessible to the subject for supposing the conditions for prima facie justification are inadequate in this case. For instance, if I believe a friend of mine to be upset, that belief can be undermined if I am aware of the fact that I am presently depressed and have a tendency to project. That provides a reason for thinking the justification possessed by my belief is inadequate.

In the contexts of the rationality of taking a belief to be justified and of showing that a belief is rationally taken as justified, a doxastic practice approach may be taken. Our focus shifts, placing the issue of belief's being based on adequate grounds within the context of the belief being formed by a well-established practice. In any case, we are not so restricted to the subject's own perspective. This necessitates us to reevaluate the function of overriders and, perhaps, of what counts as prima facie justification (since prima facie justification has been defined as justification minus overriders). Consider the following case. I acquire a particular set of perceptual beliefs as an output of engaging in perceptual practice. The beliefs, thou gh not obviously defective, are also not justified nor are they ones we would take to be justified. The reason for this that I am, unbeknownst to me, suffering from a brain lesion.

In terms of the concept of the belief having justification, it is unjustified because the ground is inadequate--due to the brain lesion, sense perception is not sufficiently indicative of the truth of the beliefs produced. It is not a matter of a rebutter since I have no reason to take the beliefs to be false. It is not a matter of an underminer since I do not, as a subject, have access to any reasons that taken with grounds accessible to me would indicate the inadequacy of those grounds. It is simply that the grounds are, in fact, inadequate.

In terms of doxastic practices and practical ascriptions of justification, the belief would be reasonably taken to be unjustified--but not simply because the ground is inadequate (though that is true). Since the practice of sense perception is quite well-established it is reasonable for us to take it to be justification-conferring. Thus my perceptual beliefs, despite the brain lesion, are to be taken to be justified when we are considering them only in light of the practice involved (that is, in a prima facie manner). As we pursue the doxastic practice further, however, the belief is to be reasonably taken as lacking justification because the justification conferred on it by the practice is overridden.

It might be overridden in one of two ways: [1] it might be rebutted by outputs of perceptual doxastic practice as engaged by others--their perceptual beliefs contradict mine and are a clear indication of the falsehood of my belief; [2] it might be undermined by the determination that my brain lesion is affecting my perception--this discovery would indicate that sense perceptual practice is not to be taken as justification conferring in this case. In terms of doxastic practices, overriding mechanisms must be seen--at least in some cases--as a social phenomenon. (Alston's discussion of overriders in connection with doxastic practices [in 1991:158-57] moves in this more social direction as contrasted with his discussions in connection with the concept of justification [in 1989:104-6, 177-79, 191-92]).

Thus it seems that overriders function in different ways depending upon whether we are considering justification conditions or whether we are practically engaged in making rational ascriptions of justification. These points will also, therefore, affect how we go about showing the reasonableness of taking a belief to be justified--at least when overriders are at issue. It also follows that concept of prima facie justification will have to be reconsidered. In the case of a belief having prima facie justification, it is a matter of whether the minimal justification conditions for the belief are met, disregarding overriders. In the case of a belief's being reasonably taken to be justified (the doxastic practice approach) it is not so clear what counts as prima facie justification. For one thing, in a practical process such as justifying a belief, the idea of "prima facie" would commonsensically be thought of in terms of "at first glance" rather than in terms of meeting minimal conditions. But just consider how fluid the notion of "at first glance" might be in terms of practical determinations of justification.

It might well be that in one case a belief's status is questioned and overriders enter in: reasons for taking the belief to be false (rebutters) or reasons for thinking its justification conditions to be inadequate (underminers). The status of the doxastic practice by which it was formed is a further matter, since underminers are within a practice while the reliability of a doxastic practice is in part an interpractice issue. It might, in some case, be "prima facie" clear that there are no overriders. In another case it might well be that "prima facie" the belief is rebutted since it is obviously false. In another case, overriders and rebutters might be bypassed in favor of an attack upon the status of the doxastic practice by which it was formed. In such a case the belief in question is obviously--so a critic might rationally think--formed by some deviant practice that is "prima facie" unreliable. The conditions, then, under which a belief is "at first glance" to be taken as possessing or lacking justification (in terms of practical determinations of such) are not so clear.

The notion of practical ascriptions of "prima facie" justification cannot, then, be usefully conceived in terms of actual situations of "at first glance" assessment. What we need, then, is some principled way to make a practically relevant distinction between under what conditions it is rational to take a belief to be prima facie justified and under what conditions it is to be taken as unqualifiedly justified. If it is to be a principled distinction, it will probably parallel the distinction drawn in the case of conditions on being justified. If so, then that would seem to dictate that overriders, as they function within the doxastic practice approach, would play some important role in the distinction. Thus, a belief is to be rationally taken as prima facie justified if it meets the minimal conditions for taking it to be justified, disregarding overriders. Alston, as already noted, has suggested that a belief's being the output of a well-established practice is minimally sufficient for taking the belief to be prima facie justified. I concur with this, save one caveat. I propose the following: a belief that p is to be taken to be prima facie justified if and only if p is the output of a well-established practice and that practice has been engaged in properly, disregarding further overriders.

Allow me to provide some support for this caveat. It would seem that engaging in a belief-forming practice improperly would serve as an underminer--a reason for taking an otherwise reliable, justification conferring practice to fail to confer justification. This is correct. Nevertheless, I still wish to include it among the conditions for taking a belief to be prima facie justified. There are several reasons.

First, since there is no one correct way to individuate practices, and since they can, for certain purposes, be rather discriminatingly distinguished, improper engagement of a practice might make for a distinction of practices. After all, if one forms beliefs by engaging in activities that are not sanctioned by the standard form of the practice P that is in question, then perhaps the practice that was engaged in is not really to be taken as practice P. For instance, if the use of hallucinogenic drugs (e.g., peyote) is used in the practice of sense perception (e.g., to be able to see one's animal-spirit guide), then perhaps that practice of sense perception is to be distinguished from the standard practice and submitted to independent criteria to determine how well established it is. There is, however, no real fact of the matter regarding how such distinctions are to be made. Thus, the caveat concerning proper engagement will catch those cases that are grouped together when less finely-grained distinctions are made between practices and that are distinguished under more fine-grained distinctions. Yet it will do this up front, without relegating the issue to the secondary matter of underminers.

There is a second argument for this caveat as well. In terms of practical determinations of justification we are concerned with the dialectics of challenge and defense of beliefs. This matter is very much one of what people do with practices and what the practices do. It is, first of all, a matter of which practices one engages in--well-established ones as opposed to not-so-well-established ones. We have a great deal of--albeit indirect--responsibility for this. Secondly, even with those practices in which we have no choice but to engage, we do have control over HOW we engage in them (depending how one individuates the practices, the matter of "which" practice and the matter of "how" it is engaged may be interchanged). I see these two matters as very closely connected.

They lead to a further step in this line of argument: we can distinguish what we do in respect to practices from what happens. Using hallucinogens is something we do. Developing a brain lesion from an accident is something that happens. Both might be taken as underminers, but their character is quite different. If there is any connection between responsibility for belief and whether a belief is reasonably taken to be justified, then, in terms of practical rationality, what we do in respect to belief-formation is more fundamental (and so, more "prima facie") than what happens to us. Thus, not only which practices we engage, but also how we engage them is more fundamental to the doxastic practice approach than what merely happens to us. They are more fundamental in how we individuate practices. They are more fundamental to the notion of something's being a practice (rather than, e.g., abstract grounds). They are more fundamental in terms of accessibility to the subject--since they involve what the subject does--and accessibility is important to showing that a belief is to be taken as justified. Finally, they are more fundamental in terms of praise and blame. Thus I include proper engagement of a practice among the conditions for taking it to confer prima facie justification.