School Work

POST, John. Infinite Regresses of Justification and of Explanation

John Post
of 23
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Infinite Regresses of Justification and of ExplanationAuthor(s): John F. PostReviewed work(s):Source: Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition, Vol. 38, No. 1 (Jul., 1980), pp. 31-52Published by: Springer Stable URL: . Accessed: 12/09/2012 18:50 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at  .  . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact  . Springer   is collaborating with JSTOR to digitize, preserve and extend access to Philosophical Studies: An International Journal for Philosophy in the Analytic Tradition.  JOHN F. POST INFINITE REGRESSES OF JUSTIFICATION AND OF EXPLANATION (Received 23 January, 1979) According to William Alston, the weakest link in the regress argument or foundationalism s the rejection of infinite regresses of justification.' I am not so sure;some other links look as weak. But Alston is right that the reasons typically given for rejecting regress eave much to be desired. What follows is (1) an argument against infinite justificational regresses that is free of problems in the arguments to date; and (2) an application of this result to show that for a wide variety of concepts of explanation, including some according to which an explanation is not a justification, an infinite regress of explanations s also impossible, for reasons that have the additional effect of undermining eading versions of the Principle of Sufficient Reason. In exploring (1) we shall see that there are logical or conceptual grounds, contained in any plausible concept of rational justification, for rejecting infinite justificational regresses. This contrasts with arguments hat make the pathology of regress either a practical matter, such as the finiteness of our faculties, or pragmatic, such as the circumstances n which it would be appropriate o request or give a justification.2 If, instead or in addition, t is conceptually impossible for there to be such a regress, then it makes no difference whether we consider an infinite intellect, who could actually justify every statement in the regress, or a finite one, who would only be able to justify any particular statement on request. In either case there cannot be a regress n which every statement is justified by prior statements. Thus my argument takes into account the objection that only an unreasonable hesis about justification would require us to reject the regress, namely that a person must actually justify every statement in the regress, as opposed to being able to justify any particular one on request.3 Rejoicing by foundationalists would be premature. No foundationalist moral follows from the rejection of justificational regresses unless the remaining inks in the regress argument are sound. In Section II I enumerate these links and sketch objections to the most important one. Whether he ob- Philosophical Studies 38 (1980) 31-52. 0031-8116/80/0381-0031$02.20 Copyright ? 1980 by D. Reidel Publishing Co., Dordrecht, Holland, and Boston, U.S.A.  32 JOHN F. POST jections can be met is an issue too complex to settle here (indeed epistemolo- gists seem generally to have underestimated its complexity). Instead I conclude the paper with a couple of morals that can be drawn far more readily. Both extend beyond epistemology, into metaphysics. One moral (Section III) is that there cannot be an infinite regress of explanations, or a very diverse family of concepts of explanation, including some according o which explanations are not justifications or even arguments. Again the reasons are not practical, such as the finiteness of our faculties, but logical or conceptual, entailed by the very notions of explanation nvolved. Even for an infinite intellect, regresses f such explanations must end. The other moral (Section IV) is that leading versions of the Principle of Sufficient Reason are either demonstrably false or question-begging n their intended applications in Cosmological Arguments or God. What conditions should inferential ustification satisfy? In particular, what is it for person P to be inferentially ustified at time t in believing statement Y on the basis of statement X, or for X to justify Y for P at t? No answer appears as yet to have achieved consensus. Fortunately none is required or our purpose. Most of the lists of conditions proposed n the literature nclude something like the following: at time t, (a) P believes Y (dispositionally or occurrently); b) P is justified in believing X; (c) P believes that X adequately supports Y; (d) P is justified in believing X adequately supports Y; (e) P believes Y because he believes both X and that X adequately supports Y; and (f) there is no defeater; that is, no statement Z such that P is justified in believing both Z and that (X & Z) does not adequately support Y.4 For our purpose, conditions (a)-(f) may be refined or augmented n many ways, according to one's views about inferential ustification. For example, let X, Y, Z be sets of statements; or replace (d) by 'X adequately supports Y', or (f) by 'P is justified in believing here are no defeaters', and so on. Adopt any plausible revision you like. Then construe 'X justifies Y' wherever t occurs below in terms of your revision. My argument against nfinite justificational regresses would still work, with only minor modifications. Suppose, contrary to what is to be shown, that for some person P at a time t, and some statement XO,  INFINITE REGRESSES OF JUSTIFICATION 33 (1) ',Xn justifies Xn -1, ..., X1 justifies XO, where no Xi in the regress s justified by any set of Xi<i (to prevent circularity). (1) is a non-circular, ustification-saturated egress (for P at t), meaning that every statement in the regress s justified by an earlier statement, and none is justified by any set of later statements (for P at t). Thus the question of whether there can be an infinite justificational regress s to be construed here as the question of whether there can be a non-circular, ustification-saturated regress. If anything counts as an inferential ustification relation, logical implica- tion does, in a sense to be specified, provided it satisfies appropriate relevance and non-circularity requirements. Let us say a statement X properly entails a statement Y iff X semantically entails Y, where the entail- ment is relevant and non-circular on any appropriate account. Thus if anything counts as an inferential ustification relation, proper entailment does, in the sense that where X and Y are statements rather han sets of statements, (2) If X properly entails Y, then Y is justified for P if X is - provided P knows that the proper entailment holds and would believe Y in light of it if he believed X Next we shall see that if there could be even one justification-saturated regress like (1) - then we could justify any logically contingent statement whatsoever. The point is not new, nor is my argument or it entirely new. But the argument will plug some old holes and help us to see what is new, the implications or regresses f explanation. Let XO be a logically contingent statement, and adopt some (alphabetical) ordering of the infinitely many statements of P's language. Then construct the entailment-saturated egress (3) ., Xn , ... I X, XO, where Xi (i >0) is the (alphabetically) first statement such that (i) Xi properly entails Xi-,; (ii) Xi is not entailed by any Xj<1; and (iii) Xi is not justified for P on the basis of any set of XA < i. Also, assume that for each Xi, P knows (or could come to know) that Xi properly entails Xi- 1 And assume hat P would believe Xi-, in light of this entailment f P believed Xi. The construction of (3) presupposes hat at every step of the regress here is some statement XA atisfying conditions (i)-(iii). This is one of those old


Jul 23, 2017
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks