Book Reviews/Comptes rendus 9S anything outside the scope of deliberation, including our ends. Richardson's prob- lem, which is our moral problem today, comes from the confrontation ofthose two distinct kinds of practical reason. I began by suggesting that both reason and the good suffer from the segrega- tion imposed on them by instrumental reasoning, which makes reasoning amoral and the good irrational. Richardson exhibits the rationality of goodness as he shows how rational our understanding and even choice of ends can be. Readers of this journal will hope for a sequel which addresses the converse problem and shows how practical reasoning can be ethical. Eugene Garver, Regents Professor of Philosophy, Saint John's University, Collegeville, MN 56321 U.S.A. egarver@csbsju.edu KNOWLEDGE AND BELIEF by Frederick F. Schmitt London and New York: Routledge, 1992. Pp. x + 278. ISBN 0-415-03317-9. Cloth US $ 55. TRUTH: A PRIMER by Frederick F. Schmitt Boulder, San Francisco and Oxford: Westview Press, 1995. Pp. xi + 251. ISBN ISBN 0-8133-2000-3. Cloth US $ 49.95. ISBN 0-8133-2001-1. Paper US $ 16.95. Reviewed by Harvey Siegel Although quite different from each other in subject matter and level of sophistica- tion, these two books by Frederick F. Schmitt will both be of interest to portions of the readership of Informal Logic. Knowledge and Belief offers a complex and elabo- rate defense of externalism in epistemology, and (more specifically) of a reliabilist theory of epistemic justification. It will be ofinterest mainly to theorists, in particu- lar to those informal logicians and argumentation theorists most concerned with the epistemological dimensions and presuppositions ofthese fields. Truth: A Primer is an introductory discussion of the main philosophical theories of truth, which de- fends a version of the correspondence theory. It would be useful in courses in infor- mallogiciargumentation (as well as in philosophy) seeking a richer discussion of truth than those offered in standard texts. The former is really a book for specialists in epistemology; the latter a clear (albeit sophisticated) introduction, suitable for a range of introductory courses. Since the field of informa.llogic includes both those 96 Informal Logic concerned with epistemology and those engaged with the design and teaching of such courses-and of course these two groups overlap to a considerable extent- both of these books will be of interest to (at least some) members of the informal logic community. Knowledge and Belief appears in the excellent book series, "The Problems of Philosophy: Their Past and Present," published by Routledge under the general editorship of Ted Honderich. Like the other books in this series, this one places contemporary philosophical discussions in their historical context, and demonstrates the relevance of earlier discussions to contemporary treatment of basic philosophi- cal issues. Schmitt propounds a radical historical thesis: that, contrary to virtually uniform common opinion, the history of Western epistemology is dominated by externalist rather than internalist theorizing, and that Cartesian skepticism, rather than being representative of mainstream epistemology in the West, is actually an atypical "extreme internalist moment in the substantially externalist history of skepticism" (p. 9). Schmitt's complex defense of this provocative historical thesis, and his readings of Plato, the great majority of Greek epistemologists, and Hume, as (albeit mainly unselfconscious) externalists and reliabilists, are well worth the read- er's attention. But Schmitt is not content to defend only this historical thesis. He also defends externalism and reliabilism as the correct epistemic positions. I cannot here provide a detailed account of Schmitt' s position and arguments, but I will try to give brief characterizations of them; the reader is urged to consult the text for the details. Roughly, then: Schmitt begins by identifying the central problem of epistemology as that of clarifying the nature of justified belief, and the relation- ship between it and true belief. He offers an instrumentalistic conception ofjustifi- cation, according to which justified belief "contributes to the [epistemically good] end of true belief' (2). If justified belief is seen as a means to true belief, then it is natural to think of it as "reliably formed belief, or belief that results from the exer- cise of a reliable cognitive belief-forming process, a process that tends to yield true beliefs" (3). Schmitt argues that reliabilism is the historically dominant position in the Western tradition, the frequent characterization of that tradition as Cartesian, a prioristic, internalist and foundationalist to the contrary notwithstanding. He de- fends his own version of reliabilism, which is developed in the context of detailed analyses of contemporary defenders and critics of it. Reliabilism is externalist in the sense that the justificatory status of a belief is a function (according to the reIiabiIist) solely of the tendency of the process which produces the belief to yield a sufficiently high proportion of true beliefs; the (non)existence of this tendency is a fact about the world which is external to the believer and independent of her appraisal of the justificatory status of the belief. Accordingly, my belief that p is justified if and only if it is produced by such a process; whether I believe that it is so produced, or believe that the relevant process is a reliable one, or have any (meta)belief at all concerning the justificatory status of my belief that p, is irrelevant to that status: my belief s justificatory status is com- . pletely independent of the presence, absence, or correctness of any such (meta)beUef Book ReviewslComptes rendus 97 concerning it. Externalism thus takes issue with intemalism, according to which my beliefs justificatory status depends in some way on my having access to that sta- tus-for example, by being aware of its relation to a body of evidence which sup- ports it. Different internalists fill out this notion of access in different ways, of course. But the general idea that to be justified in believing something, the believer must have access to the conditions of justified belief, and must believe that her belief meets those conditions, is a powerful one which many contemporary epistemolo- gists find at least plausible. Consequently, one of Schmitt's main aims is to offer a detailed criticism of internalism, in particular the version Schmitt labels "accessi- bility internalism," according to which a subject's being justified in believingp "is accessible to her in the sense that she can tell by reflection alone ... that she is. .. justified in believing p" (84). His discussion is wide-ranging, distinguishing be- tween Descartes' antecedent skepticism (which is based not on any assumptions about the world, but rather on what might conceivably be the case), and consequent skepticism (the sort addressed by the Greeks, Hume and Quine), which is based on claims about the world, in particular those concerning the reliability of our actual belief forming processes. His main thesis here is that, Descartes to the contrary notwithstanding, justified belief does not require an answer to antecedent skepticism, because this sort of skepticism requires, but there is no reason to accept, accessibil- ity internalism. Whatever one thinks of his fmal conclusions, Schmitt's linking of the internalisml externalism controversy to these distinct, historically important forms of skepticism is illuminating. En passant, he also addresses related important contemporary posi- tions; of p~cular interest is his defense of (Humean) "naturalized epistemology" (67-83). His discussion of Alston's treatment of justification, and of the latter's explicit linking of justification with argumentation practices (110-15), will be of interest to many readers of this journal. And the final chapter, on the instrumental value of the evaluation of beliefs in terms of their justificatory status, is original and provocative. As with any such book, a reviewer is bound to have some reservations; while I won't try to defend my doubts here, I should at least mention them. Besides the controversial nature of Schmitt's historical thesis and analyses, I find his criticism of internalism problematic, for two reasons. First, he argues that accessibility internalism is deficient because it "bars any account of justified belief capable of explaining the value we ascribe to justified belief' (89), but I find his arguments for this conclusion uncompelling, because his discussion of this point fails to distin- guish the value of justified belief from its character: while the internalist doesn't analyze justification in terms of truth, she can surely value justified belief because of its tie to truth; insofar, she can explain the value of justified belief in exactly the same terms as the externalist. Seconci, Schmitt's criticism of internalism, and his positive defense of reliabilism, suffer because of the conflicting argumentative strategies he relies upon in his de- velopment of them. In many places he relies, in standard "analytic epistemology" 98 Informal Logic fashion, on intuitions about cases (e.g., 111, 167, 176); but in other places he rejects such intuitions in favor of theoretical concerns (e.g., 197). This ambiva- lence (discussed more fully in Moser (1994ยป between relying on intuitions in some cases and rejecting them in others strikes me as arbitrary and ad hoc. The ad hoc nature of Schmitt's argumentative strategy is also evident in his solution to one of the key problems facing reliabilism, that of individuating the cognitive proc- esses relevant to justification. He offers a series of five constraints to be utilized in determining the relevance, to the evaluation of the justificatory status of beliefs, of the various processes that were at work in the forming of the belief; these con- straints are to guide our decision to individuate the relevant processes broadly (e.g., the process of perception) or narrowly (e.g., a certain sort of perceptual process, operative in such-and-so circumstances). The constraints are motivated by the consideration that evaluating beliefs in terms of their justificatory status helps to achieve the epistemic end of attaining true beliefs-a plausible enough motivation for the reliabilist. But while Schmitt's proposed resolution of the individuation problem in terms of his constraints is novel and in some ways prom- ising, the apparatus appears to me to be ad hoc in that the constraints point in opposite directions-some recommend a broad individuation of processes, others a narrow individuation-and the theorist can always handle troublesome counter- examples by strategically allowing the right constraint(s) to govern the example at hand. By this expedient, it appears that reliabilism cannot lose: there will always be an individuation of processes such that reliabilism yields the right (intuitive) an- swer to the question of a beliefs justificatory status. But Schmitt's attempt to provide a principled way of selecting constraints in particular cases, and thereby individuating processes, does not I think succeed. As should be clear, Knowledge and Belief offers a very rich, complex discussion of fundamental epistemological issues. The preceding description of its contents is selective and superficial; there is much more to the book than I've been able to indicate here. It is not for the faint of heart, or for those unfamiliar with contempo- rary scholarship in the theory of knowledge. But for those who (like myself) regard epistemology as fundamental to argumentation and its theory, or who are interested in a radical interpretation of some ofthe key historical figures in the Western episte- mological tradition, this is a book bound to reward and enlighten the careful reader. Truth: A Primer is, in the author's words, "a brisk introduction to philosophical thinking about truth" (ix). It contains introductory discussions of the main extant theories of truth, and of the related issues of realism vs. idealism, and absolutism vs. relativism. The discussion is indeed brisk; as Schmitt says, "[t]or each theory I have gone immediately to the heart of the matter and treated only the most central points" (ix). His discussions begin with a characterization of each main theory's "intuitive idea about the nature of truth" (for the correspondence theory the "intui- tive idea is that truth is a relation of correspondence between a proposition and the way the world is"; for the pragmatic theory "the idea is that truth is essentially Book Reviews/Comptes rendus 99 related to useful belief," etc.) (3), after which he presents and evaluates the most central arguments offered for and against each theory. Schmitt sees the various theories of truth as agreeing "that truth involves a relation to thinkers" (3), but as disagreeing on just what that relation is; this disagreement is, in his view, "the central point of contention between classical theories of truth" (4), and attention to it serves to organize and unify the discussions of each theory. Each chapter is highly focussed, dealing with only a small number of basic arguments for and against each theory; this focus makes the book especially useful as a supplemen- tary text in courses in which the main concern lies elsewhere. Given the briskness and focus of the chapters, one might easily feel that Schmitt's discussions are a bit too quick: for example, can relativism really be decisively rejected (and absolutism established) in twenty-two pages, on the basis of consideration of three objections to it, one of which ("The Platonic Objection") is ineffectual but the other two of which ("The Meaning Objection" and "The Regress Objection") are not? The same worry might be raised about Schmitt's discussion of the coherence theory, which finds that theory wanting in nineteen pages on the basis of two arguments for and six against, and for the other theories discussed. Many readers will think that Schmitt has dismissed their favored positions too quickly; even where they agree with his conclusions, many will find the discussions too fast and superficial. Schmitt may well agree; as he points out, the book is a "primer," designed to introduce and comment upon a handful of positions, issues and arguments, rather than a comprehensive treatment. In any case the cleanness and briskness of the discussion is well-suited to introduce students to this basic but difficult subject. I should note that, despite its briskness, Schmitt's book is not always easy- his defense of(Hartry Field's version of) the correspondence theory, for example, and his discussions more generally, are philosophically sophisticated, and will chal- lenge unprepared students (and instructors). But the difficult aspects of the argu- ments and positions considered benefit from Schmitt's streamlined presentation. Schmitt's four-page conclusion addresses recent criticisms of truth, put for- ward by a variety of voices in the contemporary academy, to the effect that the notion of truth functions as a tool of oppression, domination, and cultural imperi- alism, and, in light of its unsavory political effects, ought to be exorcised from the philosophical lexicon. Schmitt's defense of truth against these charges is a model of clarity, moraVpolitical sensitivity, and good sense, and will by itself make a worthy contribution to undergraduate courses Reference Moser, Paul K. (1994). Review of Knowledge and Belief, by Frederick F. Schmitt. Philosophia, vol. 23, pp. 387-392. Harvey Siegel, Department of Philosophy, University of Miami Coral Gables, FL 33124 U.S.A. hsiegel@umiami.ir.miami.edu