Monday, March 27, 2017

Evening Note for Monday, March 27

Thought for the Evening: David Braine on the Nature of Knowledge

The Catholic Herald recently had an article on the philosopher David Braine, who died last month, so it has started me thinking about some of his work. One of his more important articles is "The Nature of Knowledge" [Proceedings of the Aristotelian Society, New Series, Vol. 72 (1971 - 1972), pp. 41-63]. In it he argues against Cartesian and empiricist accounts of knowledge, and justification-based accounts in general regardless of the precise mode of justification that is allowed as legitimate, and in favor of a broadly Aristotelian account.

The key idea of the paper, at least on its critical side, is that skepticism about our ability to know is closely connected to this notion of justification; if you look at how accounts of knowledge interact with various skepticisms, you find that justification is regularly the weak point. In particular, Braine's diagnosis is that skepticisms very often take advantage of the assumption that if a justification is conditional, it is inadequate for knowledge. In his example, if Smith knows (as we would usually say) that his house, which he last saw a fortnight ago, has five rooms, obviously this depends on things like nobody secretly expanding his house while he was gone. This is very weakly conditional -- Smith's knowledge depends on something not occurring, and, indeed, something not occurring that usually doesn't occur. But this suffices for the skeptical strategy. Because the knowledge of the five rooms depends on there having been no such occurrences, the skeptic will say, it is no better than knowledge of those nonoccurrences; so to have an adequate justification for the five rooms, one must also have justification for the nonoccurrences, and so on and so forth, so that by a generalization we conclude that anything with conditional justification is inadequately justified and therefore not a case of knowledge. And, Braine notes, philosophers typically go along with this -- essentially concede the whole point and look for unconditional justifications. This has the problem of (1) putting the theory of knowledge farther and farther away from any day-to-day account of knowledge and (2) more and more implausible accounts of specific kinds of knowledge. Braine takes these as signs of misstep; a concession is being made, right at the beginning, that should not be made.

Suppose that you are trying to determine whether you are justified in a particular judgment that X is true. What do you look at? Exactly the same evidence that you look at if you are trying to determine whether X is true, and exactly the same evidence that you look at if you are trying to determine whether you know that X is true. There is no fundamental difference of evidence here. Talk of justification, where it does not simply become a synonym of knowledge, is actually just parasitic on our first-order determination of whether things are true. You don't have to look first to see whether you are justified in judging that X is true; you already are looking at everything you need to look at in trying to figure out whether X is true. Likewise, you do not need to start with epistemology in order to figure out what you know, or rely on a special kind of knowledge-inference; you start with the things themselves. By the time you start doing epistemology, you are already knowing particular things in a crude, rough, case-by-case, unsystematic way, and epistemology is the study of those things, not the precondition for them.

It is precisely this that looks broadly Aristotelian -- Aristotle held that serious inquiry generally starts with endoxa, which are, loosely speaking, the common, stable, reasoned judgments arising out of experiencing many times the way things appear to be. Braine follows this line of thought to propose an alternative account of knowledge (I am pulling together Braine's explanations into one definition):

(A) For X to know P is for X to be in a relational state that is an acquired disposition, involving rational thought or inference, that is satisfactory (for a being of X's kind) with regard to P for intellectual reasons.

This definition gives a genus for knowledge -- intellectually satisfactory relational states -- and distinguishes it at least from obvious cases of non-knowledge in the genus. It avoids circularity. It avoids the problems noted above with regard to justification. And, more importantly, it gives a more fundamental way of evaluating whether and what we know than accounts of justification do. As Braine says (p. 57), "In any exercise of human faculties or capacities, it is always some goal or satisfactory end-position which provides the standard by reference to which we (a) assess defects in the exercise of the faculties concerned, and (b) define what it is for this exercise to be free of defect in this or that respect." To determine whether you are justified, you have to know what errors you could make; to determine what errors you could make, you have to know what you are doing; to know what you are doing, you need to know the goal. To know whether the arrow was well-shot, and how well-shot it was, you need to know what the target was. What's more, this goal-focused account allows something that gets lost in discussions of justification. Because we can compare actual seeing to the satisfactory end-position of seeing, we can recognize that sight is not a matter of an on/off switch: someone with poor eyes can see, just not well; thus we can determine that for some things their sight might count as not seriously different from blindness, and yet also recognize that it may be good enough for some kinds of seeing. You don't have to look for a minimum threshold of seeing in order to determine what successful seeing is; and you don't have to look for a threshold of justification in order to determine what knowledge is.

This brings us back full circle, in a sense: there is no need for unconditional justifications; all the talk of justification was in fact covering over the sheer richness of how knowledge is grounded. In order to look at knowledge in a field, we need to recognize this richness, by looking at the kinds of ways people are relating to the world and the kinds of ways that can provide an appropriate completion to the intellectual impulse. We do not have to make any assumptions as to what these things will look like beforehand; indeed, we should not.

Various Links of Note

* Babbage's Difference Engine in LEGO.

* A number of psychologists note that the evidence that there are distinct 'learning styles' is very poor.

* Sean Davis, Fourteen Things Everyone Should Understand about Guns

* Terry Eagleton, Revolutionizing Ourselves, on Wittgenstein and politics

* Caitlin Green, A very long way from home: early Byzantine finds at the far ends of the world

Currently Reading

Dante, Purgatorio
Christ Our Pascha -- Catechism of the Ukrainian Catholic Church
Theodore the Studite, Writings on Iconoclasm
Donald Ainslie, Hume's True Scepticism
C. S. Lewis and E. M. W. Tillyard, The Personal Heresy: A Controversy
Michael Flynn, Eifelheim