Aporetics: Rational Deliberation in the Face of Inconsistency

2. Coherentism: An Aporetic Approach to Empirical Inquiry

9   2 Coherentism An Aporetic ApproAch to empiricAL inqUiry  The Ways of Inconsistency An acknowledgment of contradictions in nature goes back to the pre-Socratics.1 And if not Hegel himself, then at any rate many of his followers maintained the realization of contradictions in the world.2 Marxists of various sorts have more recently been strident supporters of such a view. It is a major historic position that merits careful evaluation. Consistency is unquestionably a prime desideratum in inquiry, but there is nothing guaranteed about it. Individually plausible contentions often disagree with one another. As the ancient sceptics stressed, experience confronts us with an inconsistent world: sight tells us the stick held at an angle underwater is bent; touch tells us it is straight. Each eye presents a somewhat different picture of the world: the brain alone enables us to “see” it consistently. To assert the consistency of nature as we grasp it is to express one’s faith that 10 Coherentism the mind will be able ultimately to impress theoretical consistency upon experienced complexity. But this confidence may in the final analysis prove to be misplaced. It is at best a hope that all such apparent discords are merely that and admit of ultimate reconciliation, a hope for whose realization a good deal of theorizing is required. And there is no reason of a priori general principle to think that this will always prevail at any and every stage of inquiry—that we might in fact not always be able to propel the state of theorizing to a point where all awkward conflicts admit of smooth theoretical reconciliation . After all, theorizing itself involves speculation and is thereby also open to inconsistency. Overall, the risk of inconsistency is an ineliminable fact of epistemic life. Its shadow dogs every step of the quest for “a true picture of reality.” Every theoretical extrapolation from the data runs the risk of clashing head-on with some other. The data themselves may conflict and cry out for theoretical reconciliation . After all, error avoidance is not the be-all and end-all of the cognitive enterprise: “Seek truth!” is no less important an injunction than “Avoid error!” and these two desiderata stand in inextricable interrelation: the prospect of truth itself unavoidably carries with it the risk of error—and even inconsistency. There is nothing regrettable, and nothing irrational, about adopting epistemic policies that allow occasional errors—and even inconsistencies—to slip through the net, provided that the general quality of the catch is high enough. The overall synthesis of our knowledge (i.e., what we think we know) with our metaknowledge (our knowledge about this knowledge ) affords an interesting illustration of the impetus toward inconsistency. The so-called Preface Paradox formulated by D. C. Makinson affords a vivid view of this phenomenon: Consider “the writer who, in the Preface to his book, concedes the occurrence of errors among his statements. Suppose that in the course of his book a writer makes a great many assertions, which we shall call S1, . . . , Sn. Given each Coherentism 11 one of these, he believes that it is true. . . . However, to say that not everything I assert in this book is true, is to say that at least one statement in this book is false. That is to say that at least one of S1, . . . , Sn is false, where S1, . . . , Sn are the statements in the book; that (S1 & . . . & Sn) is false; that ~( S1 & . . . & Sn) is true. The author who writes and believes each of S1 . . . , Sn and yet in a preface asserts and believes ~ (S1 & . . . & Sn) is, it appears, behaving very rationally. Yet clearly he is holding logically incompatible beliefs: he believes each of S1, . . . , Sn, ~ (S1 & . . . & Sn), which form an inconsistent set. The author is being rational though inconsistent.3 We begin with the series of statements in the text or main body of the book: S1, S2, . . . , Sn. (For simplicity and convenience we shall suppose that there are just two of these, i.e., n = 2.) Now the preface maintains that not all of these are true: ~ (S1 & S2). The resulting overall assertion-set {S1, S2, ~(S1 & S2)} is clearly inconsistent. Nevertheless , there is a strong impetus to accepting the whole of this set, and the tendency of this impetus is by no means irrational. In this connection, Keith Lehrer has written: The addition of such a [preface paradoxical] belief is not worth the loss measured in terms of the...