Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason [1 ed.] 9780813218298, 9780813215327

Citation preview

Copyright © 2008. Catholic University of America Press. All rights reserved. Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University

Copyright © 2008. Catholic University of America Press. All rights reserved.

Necessity & Possibility

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved. Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kurt Mosser

Necessity & Possibility The Logical Strategy of Kant’s Critique of Pure Reason

Copyright © 2008. Catholic University of America Press. All rights reserved.

e

The Catholic University of America Press Washington, D.C.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Copyright © 2008 The Catholic University of America Press All rights reserved The paper used in this publication meets the minimum requirements of American National Standards for Information Science—Permanence of Paper for Printed Library Materials, ANSI Z39.48-1984. ∞ L ibr a ry of Congr e ss Cata l o ging -in-Publication Data Mosser, Kurt. Necessity and possibility : the logical strategy of Kant’s critique of pure reason / Kurt Mosser. p. cm. Includes bibliographical references and index. ISBN 978-0-8132-1532-7 (cloth : alk. paper)  1. Kant, Immanuel, 1724–1804. Kritik der reinen Vernunft.  2. Logic.  3. Knowledge, Theory of.  4. Causation.  5. Reason.  I. Title. B2779.M83 2008 121—dc22 2008003599

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Den Philosophen wünsche ich, daß es ihnen gelingen möge, ihren größten Gegner, den alles zermalmenden Kant, zu besiegen, damit ihr Katheder für immer und ewig von dem metaphysischen Unsinn erschallen möge. —F. M. Klinger, Fausts Leben, Taten, und Höllenfahrt

e Car, plus nous avançons dans la vie, plus nous préférons le sens commun au génie lui-même, et les

Copyright © 2008. Catholic University of America Press. All rights reserved.

grandes voies où marche l’humanité aux sentiers détournés qui trop souvent aboutissent à des précipices

—V. Cousin, Philosophie de Kant

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved. Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Contents

Preface  ix Introduction: Kant’s Conception of Logic  xiii

1. Kant’s Critical Model of the Subject  1 2. Kant’s Conception of General Logic  34 3. The Historical Background of Kant’s General Logic  56

Copyright © 2008. Catholic University of America Press. All rights reserved.

4. The Metaphysical Deduction  93 5. Kant and Contemporary Philosophy  137 6. The Modesty of the Critical Philosophy  188

Bibliography  211 Index  223

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved. Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Preface

In the essay that follows, I argue that to understand Kant’s Critique of Pure Reason we must read it in terms of its own central conceit: as a book on logic. Although Kant presents a radical conception of space and time in the Critique’s “Transcendental Aesthetic,” the vast majority of the book is entitled “Transcendental Logic.” It is clear enough from even a cursory view of the text that Kant does not employ the term “logic” in the fashion virtually any current philosopher would. Nor, for that matter, does he employ the term in much the same way most of his contemporaries or predecessors did. I argue here that if we take Kant at his word, we will see that just as we can discover rules that must condition thought itself to be possible, we can discover rules that must condition thought about objects—what Kant calls “experience”—to be possible. Such rules constitute a logic, in that they yield a relatively small set of rules that range necessarily over a given domain. Because they are necessary for judgments within that domain to be possible, they can be characterized, in Kant’s term of the art, as “transcendental.” And because they function as a logic, they do not determine the content of a given judgment, but rather determine whether or not it is meaningful and truth-evaluable. The general point to be made, then, is that if the analogy between general logic and transcendental logic (to use Kant’s terminology) is at all plausible, the results of the Critique may be seen as more compelling and more relevant to some contemporary philosophical issues than those results frequently are regarded. ix

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

x   Preface The linchpin of this argument is that section of the Critique known, for better or worse, as the “metaphysical deduction,” and I will spend a good deal of time in what follows in unpacking its argument, as well as providing some historical and philosophical context for it. Yet I want to emphasize that this work does not take up the details in Kant’s text in a way that may be all too familiar to those conversant with the secondary literature on Kant and, specifically, the Critique of Pure Reason. I think the philosophical analysis of such outstanding philosophical discussions as Michael Friedman’s Kant and the Exact Sciences and Beatrice Longuenesse’s Kant and the Capacity to Judge, and the more historically oriented work of the remarkable Giorgio Tonelli and Riccardo Pozzo is extraordinarily useful. I have benefited enormously from the patient and painstaking analysis such Kant scholars have provided. While I certainly believe that this book may well interest—or provoke—those of Kant’s readers who are at home with both his texts and the secondary literature, I also hope that what I have written here will resonate with those having more general philosophical interests. For, as I argue in what follows, I think Kant has much of value to say about current issues in metaphysics, epistemology, the philosophy of mind, and other sub-disciplines of the field. If we choose to regard him as merely a figure in the history of philosophy, however prominent within that history, or if we descend into the kind of analytical detail only the most committed could hope to pursue, we may well risk neglecting the relevance of Kant’s thought for contemporary philosophy. In short, Kant’s philosophical insights are too important to be left solely to Kant scholars; sometimes, accessibility has its merits. Among a number of things that I hope to have shown here is that this emphasis on the logical structure of the Critique requires a reevaluation of the traditional view of Kant’s conception of logic. The reevaluation, I argue, suggests some surprising results, such as a new way of regarding Kant’s own evaluation of Aristotle and the latter’s logic, a more general rethinking of Kant’s relationship to the history of both logic and grammar, and a much more modest conception of what Kant establishes in the First Critique, including his oft-criticized claims for the systematic completeness of his Architectonic. Fundamentally, what I hope to show is that Kant’s strategy is to follow Rousseau’s fa-

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Preface   xi

Copyright © 2008. Catholic University of America Press. All rights reserved.

mous claim, that freedom is obedience to the law that I impose on myself. Kant generalizes this strategy, to show that the fact that we are free to think, judge, and act presupposes a commitment to a set of universal and necessary constraints, which we can discover reflectively, and which themselves make thought, judgment, and action possible. Although I only mention it briefly below, I think that this strategy does not lead to the conservative, rule-obsessed Kant that one frequently encounters (in spite of his controversial enthusiasm for such things as the French Revolution); indeed, I think those who attempt to reject this strategy out of hand risk adopting ideological and political approaches that, even if coherent, hazard extremely risky consequences. It is with a certain degree of humility that I would like to thank a number of people for having helped me with this book, in a variety of ways. I would like to thank the University of Dayton, and specifically the philosophy department there, for financial and psychological support. I would like to thank Una Cadegan, Brad Hume, and Sean Wilkinson for having read and commented on much of what I’ve written here. I would also like to thank Brad Hume and Liesl Allingham for convivial and insightful discussions about postmodernism, contemporary philosophy, history, science, and the other virtually unlimited topics that arose during our frequent conversations. For their willingness and patience in discussing all things philosophical, I would like to thank Robert Batterman, Daniel and Linda Farrell, Penelope Maddy, Lisa Shabel, Lars Fr. H. Svendsen, and William Taschek. I have learned much from the work of Robert Hanna, but unfortunately only after this manuscript had been completed. I also want to note Carol Kennedy’s meticulous handling of this manuscript, which prevented many errors (while those that remain are, of course, my responsibility). On a more personal note, I would like to thank Dr. Robert Mosser and Wanda Mosser for their continuing support, as well as Mark Brill, David Darrow, William Schuerman, Maura Donahue, Eric Suttmann, Marilyn Fischer, C. F. Martin, and William Collings. Most of what I know about Kant I learned from the late Manley Thompson; the lengthy discussions we had continue to mean a great deal to me, both personally and professionally. I would be remiss were

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

xii   Preface

Copyright © 2008. Catholic University of America Press. All rights reserved.

I not also to thank Christine Korsgaard, who has been very encouraging and supportive of my work, in spite of the pressing demands on her time. Emma Naomi Mosser and Henry Jedidiah Mosser have been, and continue to be, an inspiration to me. Finally, Robyn Reed: as they say, without whom, not.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Introduction Kant’s Conception of Logic

Copyright © 2008. Catholic University of America Press. All rights reserved.

e

Agents—specifically, human beings—think, and they do so in accordance with rules. Saul Kripke’s provocative interpretation of Wittgenstein has inspired a rather large literature around the very question of what it would even mean to follow a rule. Yet well before considering that well-known, and vexing, difficulty, it is of some use to determine the precise (if only purported) function of a given rule, as well as its modal “status,” in order to see if any application of that rule can be justified: in Kant’s language, to establish the scope and limits of a rule and, in turn, a set of rules. For Kant, such a set of rules, ranging over a specified domain, can indeed be identified, articulated, and justified— if only through a demanding process of philosophical reflection. That set of rules, again relative to a specific domain, qualifies as a logic. It is along these lines that Kant presents his conception of general, or universal (allgemeine) logic. Agents also judge, and if Kant is right, they do so in accordance with rules. Here, the specified domain—possible experience—introduces an element foreign to general logic, but the fundamental insight remains the same. One can identify, articulate, and justify (again through philosophical reflection) a set of rules relative to the domain. For Kant, the justified application of those rules within the legitimate xiii

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

xiv   Kant’s Conception of Logic Table 1. Structure of the Critique of Pure Reason I. A.

Transcendental Doctrine of Elements Transcendental Aesthetic

(pp. 17–704) (pp. 19–49)



Transcendental Logic Transcendental Analytic Transcendental Dialectic Transcendental Doctrine of Method

(pp. 50–704) (pp. 64–292) (pp. 293–704) (pp. 705–856)

B. B.1. B.2. II.

Copyright © 2008. Catholic University of America Press. All rights reserved.

Note: Page numbers refer to the first (“A”) edition of 1781.

domain of possible experience constitutes Transcendental Analytic; their illegitimate application constitutes Transcendental Dialectic. Taken together, Transcendental Analytic and Transcendental Dialectic constitute Transcendental Logic. By looking at the structure of the Critique of Pure Reason, one immediately sees that the vast majority of the text is devoted to Transcendental Logic. If one excludes the introductory material and the important, but relatively neglected, Transcendental Doctrine of Method, the remaining text consists of the radically disproportionate halves of the Transcendental Doctrine of Elements (see table 1). Clearly enough, Kant seeks to exploit the analogy between such a conception of logic, qua “a logic of possible experience,” and the more traditional conception of formal or general logic, which he calls “universal” or “general” (allgemeine) logic, although, as we will eventually see, identifying general logic with traditional or contemporary conceptions of formal or symbolic logic is problematic. Logic, in its analytic moment, establishes and justifies these rules; the illegitimate application of these rules gives rise to dialectic. The complex modal structure of Kant’s strategy should also be noted here. When general logic is said to provide a set of rules that range necessarily over a specified domain, that domain is to be regarded as possible thought; similarly, the domain for transcendental analytic is possible experience. Consequently,

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Conception of Logic   xv transcendental analytic would constitute a set of universal and necessary rules for the possibility of experience; as the point has been usefully characterized, Kant is interested in showing the “necessity of a possibility.”1 If Kant is correct, then, just as we can determine (reflectively) what rules are necessary for the possibility of thought, we can reflect on experience to determine that a certain set of conditions (rules, presuppositions) must be satisfied for that experience to be possible. The following discussion seeks to explore and, it is hoped, illuminate the analogy Kant draws between general and transcendental logic. It will be argued that this analogy, which both structures the Critique of Pure Reason and drives many of its arguments, yields a position that is immoderate, yet modest. Undoubtedly, the philosophical view in question is immoderate, in that Kant characterizes the concepts and principles it delineates as, among other things, irrevisable, incorrigible, indubitable, necessary, universal, infallible, and certain. At the same time, its results are modest, in that what Kant attempts to articulate in the Critique merely establishes the scope and limits of the application of rules, relative to a given domain, leaving, for example, specific epistemic claims to be examined in light of the “fruitful bathos of experience.”2 In this way, I hope to provide an interpretation of Kant’s Critical philosophy that not only demonstrates his strategy to put legitimate metaphysical inquiry on a firm basis, but also reveals why many of the original hopes of philosophy—the aims of applying pure reason, without critical reflection—“we may have to give up as futile” (Bvii). These results, I believe, will also reinforce the notion that, fundamentally, Kant’s views appeal to a conception of “common-sense” that was popular in his day and that continues to endure, a result that 1. H. J. de Vleeschauwer, La déduction transcendentale dans l’oeuvre de Kant (Paris: Leroux, 1934–37), 3:101; K. Reich, “Die Vollständigkeit der kantischen Urteilstafel” (Dissertation, Rostock, 1932), 27. A key passage for understanding Kant’s strategy is his distinction between “principle” and “theorem” at A737=B765; an especially helpful exegesis of this passage is given in A. Genova, “Kant’s Notion of Transcendental Presupposition in the First Critique,” in Essays on Kant’s Critique of Pure Reason, ed. J. Mohanty and R. Shahan (Norman: University of Oklahoma Press, 1982), 99–126. 2. Prolegomena, Ak. IV, 373 n.; judgments of pure mathematics raise technical issues that I ignore here.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

xvi   Kant’s Conception of Logic may appear at first glance curious for a philosopher with the reputation for forbidding difficulty that Kant has.3 I begin with what I call, for lack of better terms, Kant’s “critical” or “dialectical” conception of the thinking and judging subject. Here we examine what is perhaps the central difficulty of the Critique of Pure Reason, namely, how subjective conditions can claim objective validity, or, to phrase the point differently, why conditions imposed by a subject must at the same time be regarded as universal and necessary. The task here is complicated by the ambiguity with which Kant employs the terms “subjective” and “objective,” as well as their derivatives; further difficulties result from the fact that this issue, as much as any, compelled Kant to rewrite important parts of the Critique, which include some of the most impenetrable pages of that text. We see him, understandably, struggle with his characterization of the subject, and its role, in his published and unpublished writings between the two editions of the Critique; he continues to do so in those works (again, both published and unpublished) that occur after the Critique’s second edition. Here, I will attempt to keep the account to a manageable scope by focusing on the relationship between the thinking subject and logic as given in the Critique itself, drawing on other texts only when necessary to clarify Kant’s central claims relative to this relationship. The central claim to be argued here is that if a coherent model of the thinking and judging subject emerges from Kant’s text (particularly from the Transcendental Deduction and Paralogisms of Pure Reason), then his conception of how this subject can, on reflection, identify a set of rules—a logic—relative to a given domain is largely defensible, as is Kant’s insistence on the universality and necessity of those sets of rules for any relevantly similar agent. Indeed, Kant’s account of logic—both general and transcendental—makes sense only for the kind of thinking and judging subject he considers. If successful, this would deflect the sting of one longstanding complaint against the Critical philosophy, its alleged extreme subjectivism. Kant’s analogy between general and transcendental logic provides 3. For historical details on Kant’s relationship to the tradition of “common sense” philosophy, see Manfred Kuehn’s Scottish Common Sense in Germany, 1768–1800 (Kings-

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Conception of Logic   xvii the central thread tying together the arguments of the First Critique; the premises, arguments, and conclusions Kant offers, I believe, must be seen in light of this analogy. If that is the case, then what Kant himself means by “logic” must be clarified. This is not a particularly easy task, in that Kant himself uses the term in contexts where it is not entirely clear whether in employing the term “logic” he means general logic, transcendental logic, or some still more vague conception that includes both. The project is made still more complicated by the historical consensus that Kant’s conception of general, or formal, logic is coextensive with Aristotelian syllogistic; as C. S. Peirce sums up this consensus, “we are to remember that, according to Kant, nothing worth mention had been contributed to logic since Aristotle,” a view that one can find as easily in Hegel as in contemporary, and competent, histories of logic and philosophy.4 To clarify Kant’s own conception of logic, so fundamental to the analogy we shall examine, I look at the manner in which Kant employs the notion of “logic,” arguing that for Kant, general logic should not be interpreted along the lines of a contemporary formal model. For Kant, “logic” is employed to identify a set of conditions for thought in general—hence the term “allgemeine logic,” used to indicate its universality—in a manner similar to Wittgenstein’s usage in his Tractatus-Logico Philosophicus. I also look at Kant’s notion of a “conceptual scheme”—a notion that I will return to, in discussing some of Donald Davidson’s work— within the context of his view of natural language. Kant has consistently been taken to task, by critics of his day and our own, for having ignored the problem of natural language. The earliest, and still bestknown, objection is that of J. G. Hamann, who argued that in attempting to “purify” reason, Kant neglected the most significant aspect of language, its contingent nature, a point that has been further developed by contemporary scholars. I will argue that, on the contrary, ton [Ont.]: McGill-Queen’s University Press, 1987). Karl Ameriks has emphasized Kant’s “broad and often unappreciated concern with common sense” in his recent Kant and the Fate of Autonomy (Cambridge: Cambridge University Press, 2000). 4. C. S. Peirce, Elements of Logic §39, in Collected Papers of Charles Sanders Peirce, ed. C. Hartshorne and P. Weiss, vol. 2 (Cambridge, Mass.: Harvard University Press, 1931).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

xviii   Kant’s Conception of Logic Kant is concerned with language, albeit at a logical or grammatical level, in his attempt to identify conditions necessary for judgments to take place in any language. Therefore, Hamann and his contemporary followers fail to recognize the relevance of Kant’s approach for providing conditions of rational agency, or his contribution to the history of the search for a universal grammar. (I will sketch as well an abbreviated outline of the tradition of universal grammar that sheds a good bit of light on Kant’s own account.) Hamann’s mistake, I believe, continues to be registered by many contemporary philosophers who marvel at “Kant’s well-known indifference to language.”5 This discussion will then reinforce an earlier result of this study, namely, that on Kant’s conception of logic, we reflectively discover a set of rules that are universally and necessarily binding on thought and cognition, although here in a specific linguistic context.6 I further provide some of the historical background and context of informing Kant’s views, which will include both Stoic and Scholastic contributions to this history, including a brief look at the influential texts of the Port-Royal school, the significance of which must take into consideration not only the Port-Royal “logic” (La Logique, ou l’Art de Penser) but also the Port-Royal “grammar” (Grammaire générale et raisonnée). In the attempt to weave together these various influences on Kant’s thought, we will discover that the almost canonical identification of Kant’s logic with that of Aristotle is at best historically naïve, and has served as a serious obstacle to understanding the First Critique. With a coherent conception of the relationship between the subject and a given set of rules, and a clearer philosophical and historical understanding of what Kant means by “logic,” we take up one of the most controversial and debated aspects of Kant’s project in the Critique, the justification of those rules, specifically the Analytic moment of Transcendental Logic. Here I will begin to explore the exegetical 5. H. Aarsleff, From Locke to Saussure (Minneapolis: University of Minnesota Press, 1982), 11. 6. In this context, it is worth noting Beatrice Longuenesse’s remark that for Kant, “no judgment (as psychological activity) can take place without linguistic expression” Kant and the Capacity to Judge, trans. C. Wolfe (Princeton: Princeton University Press, 1998), 100n47; cf. “Jäsche” Logic Ak. IX, 604–5.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Conception of Logic   xix details of Kant’s arguments by examining the influence of Rousseau’s thought on Kant, specifically in addressing the question of autonomy writ large. While there is now a century-long tradition recognizing the importance of Rousseau for Kant, I will argue, albeit briefly here, that this influence has been misconstrued within that tradition by focusing almost exclusively on Kant’s practical (moral) philosophy. Instead, I will try to demonstrate that Rousseau’s importance for Kant is at least as significant in the strategy Rousseau employs, a strategy grounded in the notion of self-legislation. While there are a number of historical antecedents in the development of this notion, I will argue that Rousseau’s contribution plays the crucial role in how Kant goes about structuring his own arguments for justifying the rules of logic. I then turn to the text that is central to the relationship between General Logic and Transcendental Logic, the “Metaphysical Deduction.” The difficulties here are manifold: Kant scholars have long argued about what, if anything, is established by this “deduction” and about whether it qualifies, even on Kant’s language, as a “deduction”; even were we to grant that it does so qualify, it is not entirely clear how it differs from the much better-known Transcendental Deduction. Making things still worse, Kant uses the term “metaphysical deduction” only once, in the second edition of the Critique (at B159), although the argument to which the term purports to refer is in both editions; not surprisingly, where the argument even is has been the matter of some debate. A now-traditional reading of the metaphysical deduction, as given by Schopenhauer and Jonathan Bennett, among many others, sees Kant as beginning with a set of judgment-forms, from which he then develops (“hacks and wrenches,” in Bennett’s colorful terminology)7 a corresponding set of categories with that earlier table in mind. He can then claim to have derived the table of categories from the table of judgments, both of which he views as complete, universal, necessary, and so on. On this interpretation, the metaphysical deduction becomes an indefensible and arbitrary construct, owing too much at once to both 7. Kant’s “favoured dozen” judgment-forms “serve throughout the Critique only as a Procrustean bed on which he hacks and wrenches his philosophical insights into a grotesque ‘system,’” Kant’s Analytic (Cambridge: Cambridge University Press, 1966), 89. The image is originally Schopenhauer’s.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

xx   Kant’s Conception of Logic the Aristotelian tradition and Kant’s own unwavering loyalty to his architectonic, a view that has been vigorously challenged by Klaus Reich, Reinhardt Brandt, and others. Here I also challenge this reading, from a somewhat different direction, by looking at the table of judgments in light of the history of logic earlier sketched, by calling into question the very notion that Kant “derives” the table of categories from the table of judgments, and by offering an alternate interpretation of Kant’s notorious claim that the two tables are “complete.” I then turn to some issues where Kant’s Critical philosophy engages topics of contemporary interest. I first take up Laurence BonJour’s recent work, where he argues that without some kind of commitment to the a priori—which BonJour develops into a program he calls “moderate rationalism”—we risk “giving up rational thought altogether.” I argue that on the basis of BonJour’s reading, he in fact fails to recognize Kant as providing the strongest arguments available for the a priori, and that Kant’s results, which combine a correspondence theory of truth and a coherentist theory of knowledge and experience, are strikingly similar to BonJour’s own results. Yet BonJour, I think, fails to see that the strategy he adopts, which appeals, at best, to strong inductive arguments, is not sufficient to provide the conclusions he needs. I then return to an earlier point, to consider whether Kant’s conception of logic can be seen as a conceptual framework, imposed by a thinking and judging subject, that yields a set of rules providing minimal, albeit universal and necessary, constraints for the possibility of meaningful thought and for possible experience. Employing the very notion of a “conceptual scheme” is, of course, itself fraught with controversy; here I simply hope to show that a Kantian “conceptual scheme” or “framework,” as construed here, both is unavoidable and does not succumb to the standard kinds of objections in contemporary analytic philosophy, particularly those raised by Donald Davidson. As Davidson remarks about his own work, “Kant’s influence has been the most pervasive, but it runs so deep that I have seldom acknowledged it in print.”8 Davidson has argued, famously, that the no8. Donald Davidson, “Intellectual Autobiography,” in The Philosophy of Donald Davidson, ed. L. E. Hahn (Chicago: Open Court Press, 1999), 64.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Conception of Logic   xxi tion of alternate conceptual schemes is incoherent in its reliance on an ill-posed contrast between organizing scheme and a “given” content to be organized; consequently, the “very idea” of a conceptual scheme is itself untenable.9 Davidson argues: “We have found no intelligible basis on which can be said that schemes are different. It would be equally wrong to announce the glorious news that all mankind—all speakers of language, at least—share a common scheme and ontology. For if we cannot intelligibly say that schemes are different, neither can we intelligibly say that they are one.”10 It is not entirely clear that this is a result Kant would reject; indeed, on one reading, it is one he not only would embrace, but in fact argued for in his own work. At the same time Kant would add that certain notions, or principles, or rules—no doubt including those originally articulated in the table of judgments—would have to be presupposed even to get the kind of background agreement Davidson relies on to get his argument off the ground, although they would be discoverable only reflectively. Such a set of rules, of course, Kant characterizes as a “logic.” The point is that an agent who rejects all principles—that is, who fails to adopt any principle that could be characterized as such a principle of logic—could not qualify, on Kant’s view, as rational. The principles at issue here, then, serve as candidates for being necessary—and not sufficient—conditions of rationality. The question that arises is this: in attributing agency to another, must we attribute some set of minimal logical constraints on meaning and communicability, and, if so, can they be satisfactorily identified? Furthermore, it is worth emphasizing that this claim functions as a normative constraint on reason; frequently agents are, in fact, inconsistent; frequently we commit the simplest mistakes in reasoning and unknowingly embrace contradictory or inconsistent beliefs. But when made cognizant of such mis9. The locus classicus of the view is D. Davidson, “On the Very Idea of a Conceptual Scheme,” reprinted in Inquiries into Truth and Interpretation (Oxford: Clarendon Press, 1984), 183–98. Interestingly enough, Davidson elsewhere notes that, among others, Kant pursued his metaphysical inquiries by studying “the general structure of language”; “The Method of Truth in Metaphysics” in the same collection (199); my emphasis. See also Davidson, “The Second Person,” in Midwest Studies in Philosophy, ed. P. French, T. Uehling, and H. Wettstein, vol. 17 (Notre Dame: University of Notre Dame Press, 1992), 255–67. 10. Davidson, “On the Very Idea of a Conceptual Scheme,” 198.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

xxii   Kant’s Conception of Logic takes, we have an intellectual obligation to eliminate the inconsistency. Whether we choose to call this set of rules a “conceptual scheme” or not makes relatively little difference in this context, in that Kant is not contrasting such a logic or scheme with a given content, as are those Davidson criticizes. For Kant, the status of these rules, as universal and necessary, or a priori, is unambiguous; here I will attempt to explicate how one might view such principles within the context of Davidson’s own work, in short, what a Davidsonian account of a prioricity—if such a thing is not itself contradictory—would amount to. To be sure, Davidson’s unwillingness to countenance the analytic-synthetic distinction, and any of the strong a prioricity required by Kant’s approach, prevents Davidson from presenting a genuinely Kantian picture. At the same time, there seems to be no in-principle conflict between Davidson’s identification of language-users with belief-holding agents and Kant’s insistence that some minimal set of logical—in the sense of both general and transcendental logic—conditions is isolable and justifiable in that identification. In short, for Kant there is some core of beliefs that we must attribute to another if we are to recognize that other as an agent under any sufficiently complex description. I argue here that there may be reasons for seeing that Davidson ultimately cannot avoid introducing some such element into his program, particularly in light of his later work, where he develops a “triangulation” strategy in characterizing the relationship among two agents and a shared stimulus.11 I then look, all too briefly, at those philosophers who have influentially argued that most, if not all, of the traditional commitments of the Enlightenment—necessity, truth, and objectivity among them— need to be eliminated as philosophical goals, and we must rather remain content with the thoroughgoing “postmodern” embrace of ge11. An anonymous reader of an earlier version of this material put it well: what is needed here is a way “to tease apart the Kantian and the Quinean elements in Davidson’s work,” in terms of radical interpretation, agency, and rationality. There seem to be, and certainly are on Kant’s view, in-principle limits in characterizing these notions. From that perspective, it becomes difficult to countenance Quine’s well-known claim that “no statement is immune to revision,” and it is more than simple “logic-chopping” to ask how one might in fact revise this statement itself. See “Two Dogmas of Empiricism,” in From a Logical Point of View (Cambridge, Mass.: Harvard University Press, 1980), 43.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Conception of Logic   xxiii nealogy, convention, and ideology. Looking specifically at a fragment of the work of Michel Foucault, I argue that the reading Foucault and others have proposed fails to conceptualize Kant’s results in the logical way Kant presents them, and thus the postmodern critique of Kant either attributes to Kant positions he in fact rejects or is forced to adopt the logical thrust of his transcendental strategy to avoid a crippling relativism or skepticism. I conclude this study by tying together its various elements—historical, exegetical, and contemporary—in a relatively brief summary, arguing that Kant’s results, while immoderate in the sense that they propose universal and necessary constraints on rationality, must also be seen, in terms of what these arguments establish, as modest. We can regard, in spite of Davidson’s important and suggestive objections, Kant’s logic—the legitimate rules of general logic and transcendental analytic—as a conceptual scheme, imposing a set of unyielding, invariant synthetic concepts and principles employed a priori. But these concepts and principles must be regarded as providing the conditions of possible thought and experience and as fixing the limits within which they occur.12 The structure Kant argues for in the Transcendental Analytic, and the exposure in the Transcendental Dialectic of the errors of attempting to transcend the limits imposed by that structure (along with those of the Aesthetic), is for many difficult enough to accept. Any “defense” of Kant’s project becomes hopeless if his conceptual scheme is taken as establishing anything more than formal conditions for the possibility of thought and experience, or worse, if formal conditions are taken as establishing substantial, material conclusions about the content of that experience. Thus, I think we must regard that conceptual scheme as immodest yet minimal—immodest in establishing absolutely universal and strictly necessary conditions for the possibility of experience, yet as such imposing only a minimal framework within which questions of science, mathematics, and empirical experience are investigated. As Arthur Melnick has succinctly stated, “at least part of Kant’s empirical realism is that everything is ‘left open’ that zen.

12. See Prolegomena, Ak. IV.352, where Kant distinguishes Schranken from Gren-

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

xxiv   Kant’s Conception of Logic could be left open,” where “left open” is construed as “undecideable on a priori grounds or not in any way contributed by the subject.”13 As I hope to have shown, the interpretation of Kant’s project along the lines given in what follows makes that project considerably more attractive, and of considerable more relevance, than its current reception in contemporary philosophy would indicate.

Copyright © 2008. Catholic University of America Press. All rights reserved.

13. Arthur Melnick, Kant’s Analogies of Experience (Chicago: University of Chicago Press, 1973) 156.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Necessity & Possibility

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved. Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Ch a pter One

Kant’s Critical Model of the Subject

Copyright © 2008. Catholic University of America Press. All rights reserved.

e

Kant conceives of general logic as a set of rules—exemplified by the Principle of Non-contradiction—that holds, universally and necessarily, for thought to be possible. For Kant, by reflecting on thought, we are able to identify and articulate those rules that are necessary for thought to be possible, and possible thought itself reveals those rules to be necessary.1 This is the key to Kant’s transcendental method, whereby rules are revealed as universal and necessary relative to a given domain by reflecting on what must be the case for the judgments within that domain to be possible. Within the domain of Kant’s general logic, of course, the very activity of thought provides that content upon which we are to reflect. Kant seeks to extend this idea in developing the analogy between general and transcendental logic, by arguing that a set of rules may be 1. The question of how one might “justify” the rules that the reflection on thought itself requires leads, of course, to one version of the traditional problem of metacritique. Kant clearly recognizes the need for a systematic presentation of the rules of general logic, as one of his complaints about Aristotle is the latter’s lack of systematicity (A81=B107). At the same time, one might well question whether Kant anywhere provides the structural account of general logic that satisfies this requirement for systematicity.

1

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

2   Kant's Critical Model of the Subject identified, articulated, and justified relative to the domain of possible experience. In short, just as a judgment simpliciter must satisfy certain logical requirements to be a judgment, a judgment of possible experience must also satisfy certain logical requirements to be a judgment of possible experience. ( Judgments of experience introduce a whole new cognitive component, of course, sensibility, intuited within the forms of space and time.) As I hope to show here, this analogy makes sense only within the context of a particular conception of the judging subject, a conception of the subject that I will call Kant’s “critical” model of the experiencing subject.2 In an appendix to the Transcendental Analytic of the Critique of Pure Reason, Kant makes clear that his conception of this subject sharply contrasts with other prevailing models, specifically those of Leibniz and Locke: “Leibniz intellectualized appearances, just as Locke ..... sensualized [sensifiziert] all concepts of the understanding, i.e. interpreted them as nothing more than empirical or abstracted concepts of reflection....... [E]ach of these great men holds to one only of the two, viewing it as related immediately to things in themselves, the other doing nothing except to confuse or to order the representations which this faculty yields” (A271=B327). Kant, in rejecting the positions of Leibniz and Locke, offers his own alternative, thus making room for both pure intellectual concepts and an ineliminable sensible component. His insistence on the requirement of both a conceptual and a sensible component as “equal partners” in cognition forms the core of this “critical model” of the experiencing subject. This model must be, at the very least, plausible if Kant’s project is to have any chance of success; indeed, it may be obvious that Kant’s model of the experiencing subject is crucial for his entire philosophical approach. To support his central (transcendentally) idealistic appeal, the mind must make an active contribution in its experience. To 2. Clearly enough Kant regards the human being as the exemplar of this kind of subject, but it should be kept in mind that nowhere does Kant claim that this model reduces to human beings; that is, he leaves open the possibility that other beings may well exemplify the specific characteristics that identify such a subject (see B72). As we will see, the attribution of these characteristics to an agent is itself contingent, not a claim that can be made a priori.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant's Critical Model of the Subject   3 maintain its realistic commitment, an account must be given of that aspect of experience the mind doesn’t create. On Kant’s view, only the model he provides gives both features of experience their due. Yet nowhere does Kant directly argue for this conception of the subject, and he even indicates that a direct argument for his own model is not available. In what follows, I want to piece together, from various things Kant says, such an argument. To be sure, without this model, much of what Kant attempts becomes incoherent. Indeed, his most radical strategic move—what has become known as the “Copernican Revolution”—rests squarely upon it, because at the heart of this revolution is the active subject supplying universal and necessary conceptual order relative to a passively received sensible manifold. Given Kant’s claim that “all necessity, without exception, lies in a transcendental condition” (A106), he will otherwise be unable to carve out the special intellectual space that makes this necessity possible. For the transcendental condition at stake here is manifestly the articulation of that structure this subject imposes a priori on the sensible manifold. Furthermore, that which is so original about Kant’s proposals becomes at best uninteresting and at worst untenable when stripped of this context, for it is precisely the necessity that accompanies the judging subject’s contribution that provides the force to Kant’s claims, among others, in the Transcendental Aesthetic, the Transcendental Deduction, the Analogies of Experience, and the Refutation of Idealism, as well as the critical attacks mounted in the Transcendental Dialectic. Finally, if the connection between Kant’s theoretical and practical philosophy is as intimate as many—including Kant—take it to be, serious problems arise for the latter unless plausible support can be given to what grounds his general strategic approach. The stakes involved here, then, are considerable. Yet since Kant does not—indeed, as he admits, one cannot—give a direct argument for his model of the subject, from where might such support come? The relevant material for this support is found throughout Kant’s writings, in the published works as well as his lecture notes, correspondence, and unpublished writings, and his own notes, the Reflexionen. When these scattered remarks are drawn together, we find

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

4   Kant's Critical Model of the Subject an argument that, while indirect, is structurally a rather straightforward one, and strong enough for his purposes. Kant begins the Transcendental Logic by observing that “our cognition springs from two fundamental sources of the mind, the first of which is to receive representations (the receptivity of impressions), the second the capacity to cognize an object by means of these representations (spontaneity of concepts)” (A50=B74). This division is essential to Kant’s critical model of human cognition. Both sources are necessary for cognition; the two faculties cannot “exchange their functions” and “only through being unified can cognition arise” (A51=B75– 76).3 Human cognition thus has two ineliminable, yet complementary, components, a passively received sensible manifold (conditioned by the contributed—hence a priori—forms of intuition, space, and time) and an active understanding that synthesizes this manifold. Implicit in this characterization is a rejection of both rationalist and empiricist views of the role the subject plays in its experience. On either of these, we neglect a necessary dimension of knowledge and, as we will see Kant argue, succumb to contradictions and skepticism. Furthermore, Kant’s strategy, wherein an active understanding serves in some important sense as “the source of the laws of nature” (A127), has no hope of success save in terms of his own model. In brief, I show here that by taking the three conceptions—the rationalist, the empiricist, and the critical—as exhaustive of the possible alternatives, and showing that the first two are, for various reasons, inadequate, Kant provides an indirect argument in support of his own. At the outset, the limitations of the following discussion should be made clear. Kant’s position, as outlined below, begins with a certain kind of subject capable of experience; the specific example he has in mind, of course, is the human being. Now if the human being is this kind of subject, then it must have the distinct, yet complementary, capacities of receiving sensible impressions and making synthetic judg3. Kemp Smith translates this “Only through their union can knowledge arise,” but Kant uses the reflexive sich vereinigen, emphasizing the fact that this unification is a result of activity on the part of the subject. It is worth noting that Guyer and Wood (“Only from their unification .....”) and Pluhar (“Only from their union . ...”) employ a noun to express this activity, which is at least misleading.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant's Critical Model of the Subject   5 ments universally and necessarily, relative to these impressions. My focus is on the antecedent of this conditional, in the attempt to make explicit an argument that underlies Kant’s discussion.4 Thus here I am not so much concerned with this conditional as a whole, and even less the specifics of what constitutes, and is entailed by, its consequent. In Kant’s Transcendental Psychology, Patricia Kitcher argues in extensive detail for a picture of Kant’s philosophy of mind that reinstates Kant’s psychology as an essential element of Kant’s philosophical approach.5 This account is invaluable for its examination of concept-formation, of the connections between the Transcendental Aesthetic and Transcendental Analytic, and of Kant’s often-underestimated treatment of heuristics. But again, whatever description one provides of the specifics of Kant’s epistemological machinery, it must be embedded in the model of the subject Kant employs. Here I simply want to sketch his implicit, and indirect, argument for employing the model that he does.

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Archetypal Model What I (with Kant) will call the archetypal model of the intellect is described as an “understanding which through its self-consciousness could supply to itself the manifold of intuition.” Such a possible understanding would be one “through whose representation the objects of the representation would at the same time exist” (B138–39). Kant compares this to “a divine understanding which would not rep4. Schematically, the point here is to distinguish a claim Kant does make, at least implicitly, from those he does not. Letting “p” represent the kind of subject capable of experience Kant discusses in the Critique of Pure Reason, and “q” and “r” the sensible and intellectual capacities of that subject, the claim attributed to Kant here is p → (q & r), or equivalently, p → ( q & r). Although they are frequently attributed to him, Kant does not make, cannot consistently make, and admits he cannot make, either of the following: p → (q & r) or p → (q & r). That “p” itself is plausible, given the empiricist and rationalist alternatives, is the result of the indirect argument outlined here. 5. P. Kitcher, Kant’s Transcendental Psychology (Oxford: Oxford University Press, 1990). While Kitcher obviously offers a much fuller account, we share a common opponent, well-represented by Strawson’s dismissive characterization of Kant’s doctrine of synthesis and the “imaginary subject of transcendental psychology.” See P. Strawson, The Bounds of Sense (London: Methuen, 1966), 97.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

6   Kant's Critical Model of the Subject resent to itself given objects, but through whose representations the objects would themselves be given or produced” (B145). This model is distinguished sharply from an understanding through whose representation “I think” or “I am” no material for cognition is given. The archetypal intellect is characterized, then, as having intellectual intuition, whereas the human intellect has only an intellectual consciousness of itself (through the “I think”) and does not possess intellectual intuition (cf. Bxl n.). While we may be able to imagine and describe this archetypal intellect, Kant notes that we “cannot form the least conception” of such an understanding (B139; cf. B72); indeed, in thinking of an understanding that intuits things immediately, or has intellectual intuition, “we should not have the least concept, or of beings of the understanding to which it should be applied” (Ak. IV, 316n; cf. XX, 267). But we can employ this useful fiction not only to see why it fails, but also in order to isolate the “peculiarity of our understanding” that requires a synthesis of a given sensible manifold for cognition.6 The archetypal intellect is creative in that its objects exist by virtue of being thought. It thus creates its own objects and gives them to itself by means of intellectual intuition. For an understanding that possesses intellectual intuition, there is then no distinction between an object as it is thought and an object as it exists. Hence, there is no distinction to be drawn between possibility and actuality, in that what an archetypal intellect thinks thereby actually exists; it is, in effect, perfectly active. For the human understanding, which only thinks and does not intuit (B139), there is a clear difference between the possibility of an object and its actuality. As we have seen, for an object to be possible at all, its concept cannot be self-contradictory; it must satisfy, minimally, the principle of non-contradiction. “I can think whatever I please, provided only that I do not contradict myself, that is, provided my concept is a possible thought” (Bxxvi n.). This condition of general (formal) logic Kant calls logical possibility. This condition is distinct from the con6. Allison claims that the archetypal intellect functions “essentially as a regulative idea in the Kantian sense.” H. Allison, Kant’s Transcendental Idealism (New Haven: Yale University Press, 1983), 19. While this seems to be stretching the notion of a “regulative idea,” the point is simply that Kant is trying to demonstrate by contrast the inadequacy of such a model for describing a human, or more generally, discursive, intellect.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant's Critical Model of the Subject   7 ditions for possible knowledge imposed by transcendental logic (specifically transcendental analytic), the necessary conditions of possible experience, which yield in the theoretical philosophy objective validity, or “real [reale] possibility” (ibid.). Finally, to be able to know an object as actual, we require an immediate perception; not necessarily of the object itself, but “the connection of the object with some [irgendeiner] actual perception, in accordance with the analogies of experience, which present all real connection in an experience in general” (A225=B272). Central to this whole picture is the notion that the subject must make judgments about a passively received sensible manifold (albeit received within the forms of intuition). For a discursive intellect, an object of possible experience is an object of judgment, requiring the a priori application of synthetic concepts—the categories. An archetypal intellect that creates its own objects has no need for making judgments at all. And without the notion of judgment, we lose in a very basic sense the possibility of ascertaining whether a judgment is true or false. The archetypal model of the human intellect fails to subsume adequately the human understanding for the simple and obvious reason that the human being can think an object without that object thereby existing. Kant’s rejection of this archetypal model is more or less stated without argument in the first Critique, but such an argument can be put together from remarks he makes in various places, particularly the second-edition preface, the Postulates of Empirical Thought, and the Amphiboly of Concepts of Reflection. However, Kant conveniently summarizes the point in a well-known passage from the Critique of Judgment: It is unavoidably necessary for the human understanding to distinguish the possibility and actuality of things. The ground for this lies in the subject and the nature of its cognitive faculties. There would be no such distinction (between the possible and the actual) were there not two completely heterogeneous elements [Stücke], understanding for concepts and sensible intuition for objects [Objekte] corresponding to them, for the exercise of these faculties. That is, if our understanding were intuitive, it would have no objects [Gegenstände] than the actual objects. (Ak. V, 401–2 (§76))

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

8   Kant's Critical Model of the Subject

Copyright © 2008. Catholic University of America Press. All rights reserved.

The argument has a form quite common in Kant’s writings—modus tollens with the denial of its consequent seen as so obvious that it requires no further argument.7 If human beings had intuitive intellects, all objects they thought would be actual; since it is clear that not all objects thought of or about by human beings are (necessarily) actual or existent, it is not the case that humans have intuitive intellects. I may be able to think of a forty-six-foot tall green-and-white striped bowling ball that tastes like persimmons and smells like honeysuckle, but that does not mean that there actually is such a thing by virtue of this thought. Kant clearly directs his discussion of the archetypal model against Leibniz, for whom sensible knowledge is simply a confused or inferior version of divine knowledge as obtained through the pure intellect.8 Kant takes the Leibnizian position to treat sensible intuition as “only the confused concept of its object and, consequently, intuition is distinguished from concepts of things only in virtue of the degree of 7. While Kant frequently employs this logical move in the first Critique, perhaps the clearest example comes out at a crucial step in the second-edition Transcendental Deduction: “only in so far as I can grasp the manifold of the representations in one consciousness, do I call them one and all mine. For otherwise I should have as manycolored and diverse a self as I have representations of which I am conscious to myself ” (B134). The assumption here, that it is obvious that the thinking subject represented by the transcendental unity of apperception is not as diverse as its representations, is of course an important part of Kant’s criticism of Hume. Yet it should be noted that this claim is the outcome of a general theme running throughout the Critique—that the unity of the subject is reflected in the unity of objects. This notion has been clearly outlined in Arthur Melnick’s Kant’s Analogies of Experience, 144–49, and more fully in A. C. Ewing, Kant’s Treatment of Causality (London: Kegan Paul, Trench, Trubner, 1924), chapter 5. In any case, the point—which we will see again—is that while the content of Kant’s arguments may be complex and even obscure, the logical structure of those arguments is often surprisingly straightforward. 8. Leibniz is specifically singled out in the Amphiboly. But it is clear that Kant includes under this heading a variety of thinkers, all of whom are in one way or another guilty of “intellectualizing appearances.” Such thinkers would include Plato, Malebranche, and Crusius, as well as Baumgarten and Wolff. Kant summarizes the point in the Transcendental Aesthetic: “The philosophy of Leibniz and Wolff, in thus treating the difference between the sensible and the intelligible as merely logical, has given a completely wrong direction into the nature and origin of our cognition. This difference is quite evidently transcendental” (A44=B61; my emphases). Leibniz himself writes in the preface to the New Essays, directed against Locke, that Locke’s system is “closer to Aristotle and mine to Plato.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant's Critical Model of the Subject   9 consciousness, not specifically” (Ak. XX, 278). While Leibniz doesn’t equate the finite human intellect with the infinite intellect of God, the divine intellect serves as the paradigm toward which the human intellect strives to approach. Kant, on the contrary, claims we can consider such an archetypal intellect only problematically; for the human intellect, intuition and concept must be distinguished “specifically”—that is, shown to be different in kind, rather than degree.9 If the human intellect does not create its objects through intellectual intuition, there must then be for human beings a radically different source, through which an object may be given. If this faculty is not active, it must be passive, merely a faculty of receiving impressions. It is this passive, receptive capacity Kant calls the faculty of sensibility. In stark contrast to the archetypal model, Kant’s denial of intellectual intuition to human beings reveals that we are “thoroughly dependent on an entirely different faculty (of receptivity) for its intuitions, or better, for the material out of which it produces [hervorzubringen] knowledge” (Ak. XI, 54). Admittedly, Kant’s rejection of human intuition as intellectual does not entail that human intuition is sensible. That human cognition has an ineliminably receptive, sensuous element is simply an obvious alternative to this rejected model, and is merely put forth as an assertion by Kant: “Our nature is so constituted [bringt es so mit sich] that our intuition can never be other than sensible; that is, it contains only the mode in which we are affected by objects” (A51=B75). The point is that human beings in some sense rely on sensible intuitions, received under (or, better, within) the forms of space and time, to present a sensible manifold that, through synthesis, can produce objective cognition, or experience. In this way, the human being is in a passive relation 9. For my limited purposes here, I don’t need to go into the details of Kant’s critique of Leibniz. But it should be noted that Kant’s radical division of sensibility from understanding, and the concomitant rejection of the possibility of intellectual intuition for human beings, grounds Kant’s arguments against the identity of indiscernibles, the doctrine of analyticity, the monadology, and the relational theory of space and time (cf. A271=B327ff.). Many of these points are expanded upon in two pieces Kant directed against Leibnizian rationalism, namely Über eine Entdeckung (Ak. VIII) (which has a particularly valuable discussion of Kant’s analytic-synthetic distinction, in response to the version presented in defense of Leibniz by Eberhard), and the Fortschritte.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

10   Kant's Critical Model of the Subject to these intuitions. Hence, we can see that Kant’s characterization of this receptive faculty, as Manley Thompson notes, is “intended to capture something of our sense of being in a world we never made.”10

The Ectypal Model

Copyright © 2008. Catholic University of America Press. All rights reserved.

Whereas Kant’s discussion of the archetypal intellect is directed against Leibnizian rationalism and its variants, the opponent in his critique of the ectypal intellect is Lockean empiricism.11 Kant claims, in the introduction to the Critique, to have “established the fact that our faculty of cognition does have a pure employment,” in judgments as well as concepts (B5). He regards this as clear enough from “any of the propositions of mathematics” as well as in the ordinary employment of the understanding (B4). Even without examples, “it is possible to show that pure a priori principles are indispensable for the possibility of experience” (B5). For without universal and necessary conditions for judgment, all rules for experience would be contingent. Thus universality and necessity, the criteria of a priori knowledge, serve as criteria for the pure employment of the faculty of understanding. But does this argument rest on a petitio, by simply assuming that we do in fact possess a pure employment of the understanding? The 10. Manley Thompson, “Things in Themselves,” Proceedings of the American Philosophical Association 57 (1983): 36. See also Klaus Reich’s discussion of the claim that “the concept of an object that alone has a meaning for me, rests on [the fact that] I am conscious of myself in the ‘I think’ as spontaneous, but also as non-creative [nichtschöpferisch] thus, if one wants, as finite.” “Die Vollständigkeit,” 32–33. At this point, I am undoubtedly oversimplifying an extremely complicated picture Kant presents of the nature of human sensibility. Some of these complications are more fully explored in my dissertation, “Kant and Epistemology” (University of Chicago, 1990). But even at this stage, an interesting question develops. Is Kant, as is often claimed, trying to prove in the Transcendental Aesthetic that intuition is sensible, or is he rather developing his argument in terms of the question that if it is sensible, what must follow from that claim? 11. Kant does refer, in the Critique of Judgment, to “our discursive understanding that has need of images [Bilder] (intellectus ectypus)”; Ak. V, 408 (§77). I employ the term “ectypal intellect” more as convenient (and symmetrical) shorthand for the empiricist’s model than for any significant philological purpose; in any case, we share—insofar as we have a passive component to our cognition—this aspect with the empirical model. As we will see, Kant’s point is that this is not sufficient in giving an account of human cognition, which does not conflict with the claim made in the Third Critique.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant's Critical Model of the Subject   11 objection seems a natural one to make; as Simon Laurie asks, is this “not to beg the question, there may be no ‘certitude,’ whatever certitude may mean?”12 Vaihinger correctly points out that supporting (rather than merely demonstrating) this claim with examples from pure mathematics and pure natural science is itself an a posteriori, inductive strategy; “examples would present a priori procedures only accidentally.”13 Kant admits that an empirical derivation of claims yields only “comparative universality” (B3) and that such a derivation of a priori principles is an “utterly useless enterprise” (A87=B119). Hegel’s criticism gets to the heart of the issue: “That in cognition are found the determinations of universality and necessity is a fact not disputed [even] by Hume’s skepticism. In Kant’s philosophy, it isn’t anything other than a presupposed fact [Faktum]; in the customary language of the sciences one can say that Kant has just offered another explanation of the fact.”14 What is really at stake here is the viability of Kant’s first-person approach to philosophy. It might be said in response to Hegel that Kant does offer an “explanation” of a fact; but it is a unique kind of explanation in being transcendental, and it is a rather peculiar kind of “fact.” Hegel, it is well-known, chides Kant for being a “reflective” philosopher; he is correct in that Kant’s argument is to the effect that a priori principles must be recognized, in reflecting on experience, as necessarily presupposed for that experience to be possible. However, Hegel’s critique raises important questions that are beyond the scope of this discussion, particularly by pointing to the thorny issue of the possibility of a “metacritique” of pure reason.15 The ectypal intellect is described as one that “would derive the data for its logical procedure from the sensuous intuitions of things” (Ak. X, 12. Simon Laurie, “Kant’s Criticism of Pure Reason,” Journal of Speculative Philosophy 6 (1872): 224. 13. Hans Vaihinger, Commentar zu Kants Kritik der reinen Vernunft (Stuttgart: Spemann, 1881), 1:315. This is a problem only for the synthetic method of the Critique, and not for the analytic method of the Prolegomena; cf. Ak. IV, 274–75; 276n. 14. G. W. F. Hegel, Enzyklopädie der philosophischen Wissenschaften, vol. 1: Die Wissenschaft der Logik, §40 (Frankfurt a.M.: Suhrkamp, 1983), 113. 15. See note 39 below, where I return briefly to this topic.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

12   Kant's Critical Model of the Subject

Copyright © 2008. Catholic University of America Press. All rights reserved.

130).16 This ectypal intellect, then, seeks to derive certain general concepts by abstracting them from experience. The difficulty is that such a procedure fails to guarantee the universality and necessity of its general concepts, in that experience “teaches us that a thing is so and so, but not that it cannot be otherwise” (B3). Locke’s error was in attempting to derive or abstract from experience concepts that are necessary and universal for experience. Locke’s “physiological derivation” corresponds only to Kant’s quaestio facti; such a derivation cannot justify the requisite characteristics of universality and necessity (A87=B119). Thus if we do possess a pure employment of the understanding, Locke’s derivation fails to provide the justification this pure employment demands. To take the human intellect as creative, as the model of the archetypal intellect does, is to fail to take into account the receptive faculty through which intuitions are given. But to take the human intellect as solely passive, as the model of the ectypal intellect does, is to fail to take into account the active faculty of combination through which the given sensible manifold is taken up, synthesized, and judged. The full consequences are spelled out in an oft-cited 1789 letter to Herz: [A]ll data of sense for a possible cognition would never, without those conditions [the forms of intuition and the categories], represent objects. They would not even reach that unity of consciousness that is necessary for knowledge of myself (as object of inner sense). I would not even be able to know [wissen] that I have the data of sense; consequently for me, as a knowing being [erkennendes Wesen] they would be absolutely nothing. (Ak. XI, 52)

Here Kant clearly indicates the necessity of the active faculty, reprising a central argument of the Transcendental Deduction. Without the spontaneous “I think” accompanying “all my representations,” such representations “would be impossible, or at least would be nothing to me” (B131–32). Again, the details involved are extremely complex; here I need only note that this awareness of an active capacity (the understanding) distinguishes, for Kant, human beings from other ani16. In the Lectures on Metaphysics L1 (Pölitz), Kant takes this to be the view of Aristotle, and after him the Scholastics, in that all accept the notion that “nihil est in intellectu, quod non antea fuerit in sensu”; Ak. XXVIII.1, 232. Cf. letter to Reinhold, 5.19.91; Ak. XI, 40.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant's Critical Model of the Subject   13 mals. As he continues in the letter to Herz, the sensible data, for an ectypal intellect, could still (I think of myself as an animal) carry on their play in an orderly [regelmäßig] fashion, as representations connected according to empirical laws of association, and thus even have an influence on my faculties of feeling and desire, without my being conscious of my own existence (assuming that I am even conscious of each individual representation, but not of their relation to the unity of representation of their object, by means of their synthetic unity of apperception). This might be so without my knowing thereby the slightest thing, not even my own condition.

Copyright © 2008. Catholic University of America Press. All rights reserved.

As early as 1762, Kant distinguishes between rational animals and animals without reason.17 He continues to employ this distinction, basing it on a capacity human beings possess, and all other animals lack— namely, the active capacity of human beings for discursive thought, attested to by our ability to think “I think.” In one of the sets of Lectures on Metaphysics, Kant argues that human beings are different in kind from animals because human beings have self-consciousness, indicated by our use of “I” (Ak. XXVIII.1, 275ff.).18 In the Lectures on Anthropology, it is the fact that human beings can have the representation “I” that raises them infinitely beyond all other beings on earth (Ak. VIII, 127).19 The point is summarized in the Fortschritte, written sometime after 1790: How it might be possible for the I that thinks to be an object (of intuition) for me, one that enables me to distinguish me from myself, is absolutely impossible to explain, even though it is an indubitable fact; it indicates, however, a capacity so highly elevated above sensuous intuition that, as the basis 17. In the essay on the “subtleties” of the syllogism (Ak. II, 60); the same distinction is given as late as 1796, in “Verkündigung”; (Ak. VIII, 413f.). See also “Deutlichkeit” (1764) (Ak. II, 284f.); Opus Postumum (Ak. XXI, 18). 18. This is the fullest discussion Kant gives of this distinction of which I am aware. See also the Herder lecture set, where Kant distinguishes three dimensions of thinking beings; as he describes this third dimension, “Consciousness of the entire condition of representations and feelings”—“We know [kennen] merely man with these three abilities [representation, representation and feelings, and this consciousness] and more are not thinkable for us” (Ak XXVIII.1, 117). 19. Cf. the preface to the Critique of Practical Reason (Ak. V., 12).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

14   Kant's Critical Model of the Subject

Copyright © 2008. Catholic University of America Press. All rights reserved.

for the possibility of understanding, it has the effect of separating us from the animals, to which we have no reason for ascribing the ability to say “I” to themselves, and results in an infinity of self-constituted representations and concepts. (Ak. XX, 270)20

Without the active capacity for combination, human beings would lack the ability to introduce their representations into the unity of consciousness and unify them under rules. An intellect solely in a passive relationship with its representations would lack this active faculty of combination, and would lack the ability to make judgments about its representations in accordance with universal and necessary rules. It would thus be unable to attain objective knowledge. Furthermore, as we have seen, this ectypal model would fail to respect fully the human capacity to call itself “I,” the consequences of which are developed in Kant’s practical philosophy. While Locke recognizes a role for “pure” concepts, his attempt to support them through an empirical derivation fails to guarantee for them universality and necessity. Hume’s insight, in contrast, is to see that such concepts would require an a priori origin. But because it never occurs to Hume “that the understanding might itself, perhaps, through these concepts, be the author [Urheber] of the experience in which its objects are found,” he as well tried to derive them from experience (B127). Hume thus ascribes the “alleged universality and necessity” (A765=B793) of these concepts to a subjective or psychological compulsion “which comes mistakenly to be regarded as objective” (B127). While he argues consistently from this position, Hume, on Kant’s reading, “gives himself over entirely to skepticism” (B128), and fails to distinguish between those concepts that are objective and necessary for the possibility of experience and those concepts that attempt to go beyond the field of possible experience.21 The former con20. It is interesting to note that while here Kant see this ability as “absolutely impossible to explain,” he sees less of a difficulty in the Critique: “How the ‘I’ that I think can be different from the ‘I’ that intuits itself ..... [raises] no greater nor lesser difficulty than how I can be an object to myself at all, and, more particularly, an object of intuition and of inner perceptions. Indeed, that this is how it must be, is easily shown” (B155–56). He does admit that the problem has the appearance of being a “paradox” (B152). 21. Although it is perhaps worse for Locke, who “opened a wide door to the Schwärmerei” (B128).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant's Critical Model of the Subject   15 cepts, of course, are for Kant pure concepts, which are contributed by the subject and whose universality and necessity for the possibility of experience must be proved through a transcendental deduction. The latter concepts, in their attempt to go beyond experience, lead to dialectical illusion. Hume thus not only fails to give an accurate assessment of the pure concepts of the understanding, but is unable to draw a distinction “between the well-grounded claims of the understanding and the dialectical pretensions of reason” (A768=B796). Certain concepts are “identified [bestimmt] for pure employment” (A85=B117); through an empirical derivation or through induction these concepts achieve “only comparative universality, that is, widespread applicability” (A91–92=B124). But if pure, these concepts must possess strict universality and necessity, and such a warrant cannot come from experience. We can only note their persistence by attributing their use to a habit or custom, as did Hume, or we must show for them a “certificate of birth entirely distinct from that of descent from experience” (A86=B119). To do this, we must show that these concepts are contributed by the subject.22 As mentioned above, Kant sees pure mathematics as a clear case for the employment of pure concepts; indeed, it “gives us a shining example of how far, independently of experience, we can progress a priori in cognition” (A4=B8). Kant also takes pure mathematics as the crucial test for Hume’s approach, and the best indication of the difficulties such an approach encounters. If we accept Hume’s conclusions, Kant argues, then all we call metaphysics is a mere delusion whereby we fancy ourselves to have rational insight into what, in actual fact, is borrowed solely from 22. This, of course, is Kant’s task in the Transcendental Analytic; in particular, this is the demand he tries to satisfy in the Transcendental Deduction. This also leads to a further complication, the distinction between the justification of synthetic judgments made a priori and the origins of the capacity that makes such judgments possible. This is the traditional distinction between the “objective” and “subjective” sides of the Transcendental Deduction (see Axvi–xvii). Kitcher, in Kant’s Transcendental Psychology, takes the latter issue, the “subjective” question about origins, to be crucial; see 224. On the other hand, it can be (and has been) argued that in Kant’s revisions for the second edition of the Critique, it is precisely this feature that he downplays, or even tries to omit. Whether he is successful in this or not, or whether such an attempt—as Kitcher indicates—would make his account seriously inadequate, raises questions beyond the present discussion.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

16   Kant's Critical Model of the Subject experience, and under the influence of custom has taken the illusory semblance of necessity. If he had envisaged our problem in all its universality, he would never have been guilty of this statement, so destructive of all pure philosophy. For he would have recognized that, according to his own argument, pure mathematics, as certainly containing synthetic propositions a priori, would also not be possible; and from such an assertion his sound understanding [Verstand] would have saved him. (B20)

Copyright © 2008. Catholic University of America Press. All rights reserved.

The argument, then, is this: Hume saw no possibility that the principle of causality could be objectively necessary. If he had expanded his investigation, he would have seen that his procedure would have eliminated the universality and necessity that characterizes propositions of pure mathematics, which Hume accepted (at least for arithmetic and algebra). Hence, if he had recognized such an implication, he would have in turn been able to discover a new solution to the original problem of causality through a relation ampliative of experience, but universal and necessary; that is, a synthetic concept employable a priori. Instead, Hume took pure mathematics to be analytic, and was prevented from making this new discovery. Whether in fact Hume took the propositions of pure mathematics to be analytic is an issue I do not need to pursue here; it is clear enough that Kant understood him to do so.23 Kant argued, on the contrary, that “[a]ll mathematical judgements, without exception, are synthetic” (B14). The thrust of the argument is that objects of pure 23. Mark Steiner has argued that Hume held mathematical propositions to be synthetic and a priori; “Kant’s Misrepresentations of Hume’s Philosophy of Mathematics in the Prolegomena,” Hume Studies 12 (1987): 400–410. R. F. Atkinson is more cautious, arguing only that such a possibility is not excluded from Hume’s treatment of mathematics; “Hume on Mathematics,” Philosophical Quarterly 10 (1960): 127–37. The situation becomes more complicated, naturally, when the relationship between the mathematical discussion of the first Enquiry and that of the Treatise, and what Kant knew of the latter, is taken into account. L. W. Beck notes that Kant “continued to ascribe to Hume only the mathematical teachings of the Enquiry,” but points out the striking resemblance of a passage from the Critique to a passage from the Treatise, both denying that “A straight line is the shortest distance between two points” is analytic (Kant) or a definition (Hume). “A Prussian Hume and A Scottish Kant” in Essays on Kant and Hume (New Haven: Yale University Press, 1978), 117–18, esp. n20. See also Vaihinger, Commentar, 1:361–64. As Manley Thompson points out, Kant in fact says “that the straight line between two points is the shortest,” and does not mention distance; see “Unity, Plurality, and Totality as Kantian Categories,” Monist 72 (1989): 187n16, for discussion.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant's Critical Model of the Subject   17

Copyright © 2008. Catholic University of America Press. All rights reserved.

mathematics must be able to be exhibited in pure intuition (A4=B8); we must go “outside” the concepts of pure mathematics and “appeal to intuition” (B16). The argument is completed, however, only in the Methodology. Here Kant elaborates on the requirement that mathematics must be able to construct its objects in pure intuition, whether in a symbolic construction (as in algebra), or in an ostensive construction (as in geometry) (A717=B745). The distinction between the employment of reason in mathematics and in what Kant refers to as “philosophy” is not drawn, then, between the respective analyticity and syntheticity of their judgments. It is drawn rather between the nature of the objects with which they deal, in that “Philosophical cognition is cognition gained by reason from concepts; mathematical cognition is cognition gained by reason from the construction of concepts” (A713=B741). Mathematics constructs its concepts by exhibiting them in pure intuition; philosophical cognition, which yields a priori the synthetic concepts that make experience possible, must deal with “the synthesis of possible intuitions which are not given a priori” (A719=B747; cf. A240=B299). The details of Kant’s philosophy of mathematics are complex;24 but we can at least see Kant’s reason for the point that mathematics contains synthetic judgments. In the Prolegomena, Kant spells out clearly the argument we have seen—had Hume seen that mathematics contains a priori synthetic judgments, 24. Many of the questions involved here have been examined in an ongoing discussion between Jaako Hintikka and Charles Parsons; for some of the debate, and references, see Parsons’s “Kant’s Philosophy of Arithmetic” in Kant on Pure Reason, ed. R. C. S. Walker (Oxford: Oxford University Press, 1982), 13–40. A good discussion of the issues is chapter 2 of Gordon Brittan, Kant’s Theory of Science (Princeton: Princeton University Press, 1978). After discussing the role of free variables in what he calls the “BethHintikka” reconstruction of Kant’s mathematical approach, Brittan emphasizes the existence claims (hence, for Kant, synthetic) of the Zermelo-Fraenkel axiomatization of set theory, among them the null-set axiom and the axiom of infinity, and their relation to the logicist program in the foundations of mathematics. Brittan’s interpretation has in turn been criticized by Hintikka in “Kant’s Theory of Mathematics Revisited,” Essays on Kant’s Critique of Pure Reason, ed. J. Mohanty and R. Shahan (Norman: University of Oklahoma Press, 1982), 201–15. Thompson’s “Singular Terms and Intuition in Kant’s Epistemology” (Review of Metaphysics 26 [1973]: 314–43) extends the debate by discussing the role intuitions play in the context of empirical judgments, and is also criticized by Hintikka. Houston Smit has recently revisited the debate in “Kant on Marks and the Immediacy of Intuition,” Philosophical Review 109.2 (April 2000): 235–66.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

18   Kant's Critical Model of the Subject

Copyright © 2008. Catholic University of America Press. All rights reserved.

he would have extended his question concerning the origin of our synthetic judgements far beyond the metaphysical concept of causality and included in it the possibility of mathematics a priori also; for this latter he must have assumed to be equally synthetic. And then he could not have based his metaphysical judgements on mere experience without subjecting the axioms of mathematics equally to experience, a thing which he was far too acute to do. (Ak. IV, 272–73 (§2c.2))

Kant repeats the same basic argument in the Critique of Practical Reason, concluding “if one assumes a universal empiricism, mathematics will also be involved” (Ak. V, 13; see Ak. V, 50–55). Kant sees himself in agreement with Hume that pure mathematical judgments must be necessary; Hume’s mistake is to take this to mean that these judgments are analytic. But if such judgments are in fact synthetic, yet still strictly necessary (a priori), they could not be derived from experience. Such a derivation would attain only relative, not strict, necessity and universality. Thus the universality and necessity of mathematical judgments, as synthetic and a priori, can be guaranteed only by having a wholly different origin than experience (here, being able to be constructed in pure intuition). This example can then be generalized relative to those marked out for pure employment in experience. If Hume had seen the consequences of his approach for mathematics, he would also have seen the larger problem, what Kant calls the “proper problem of pure reason”—how are synthetic judgments possible a priori (B19)? As L. W. Beck has characterized the significance of this connection between mathematics and the synthetic judgments made a priori for experience to be possible, Kant saw in mathematics a clue to the objectivity of all a priori knowledge, both analytic and what he considered to be synthetic. This is indeed the sense of the Copernican Revolution: even empirical objects are constructions; and their necessary conditions are geometrical. Had Kant radically sundered mathematical knowledge from the intuitive a priori structures of empirical knowledge, as he criticizes Hume for doing, both would have been rendered unintelligible to him.25 25. Beck, “Can Kant’s Synthetic Judgements Be Made Analytic?” 22.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant's Critical Model of the Subject   19

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Archetypal and Ectypal Intellect As we have seen, then, the ectypal model of the human intellect fails to give an adequate account of human knowledge for two fundamental reasons. This model cannot sufficiently distinguish an active, spontaneous intellect from a solely reactive, passive intellect, which Kant sees as the specific difference between human beings and animals. Nor can it guarantee universality and necessity to the pure concepts Kant argues to be the conditions of possible human experience. Pure mathematics gives the clearest example of the pervasive difficulty that follows the adherence to a “universal empiricism.” The archetypal and ectypal model have in common the failure to secure the universality and necessity of pure concepts. We have seen why an empirical derivation of pure concepts fails to provide such a warrant; we can now look briefly at the similar failing of the archetypal model. Again, as opposed to an ectypal intellect that abstracts inductively its general concepts, the archetypal intellect is one that produces its own objects by means of intellectual intuition. A variety of this latter model is discussed toward the end of the second-edition Transcendental Deduction, where Kant considers different views of how human knowledge is possible in terms of its origins. He first rejects the ectypal model as a generatio aequivoca, attempting to derive pure concepts, which must be independent of experience, from experience.26 In a regrettably brief and obscure remark, Kant suggests his own solution on analogy with the biological theory of epigenesis, in which the a priori concepts that make experience possible are self-thought.27 Finally, a “middle course” is proposed of preestablished harmony, wherein pure concepts are “subjective dispositions of thought, implanted in us from the first moment of our existence.” Kant refers to this picture 26. It is an argument Kant scorns elsewhere as trying to get water from a pumice stone (ex pumice aquam)—Critique of Practical Reason, Ak. V, 12. 27. A brief but informative discussion of Kant’s biological analogy and its historical background is J. Wubnig, “The Epigenesis of Pure Reason,” Kant-Studien 60 (1969): 147–52; a richer, more recent, article focusing on epigenesis and its philosophical ramifications is Günter Zöller, “Kant on the Generation of Metaphysical Knowledge,” in Kant: Analyse—Probleme—Kritik, ed. H. Oberer and G. Seel (Würzburg: Könighausen and Neumann, 1988), 71–90.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

20   Kant's Critical Model of the Subject as “a kind of preformation system of pure reason,” ordered by our Creator in such a way that the employment of these “dispositions” would be “in complete harmony with the laws of nature in accordance with which experience proceeds” (B167). Kant points out two insurmountable problems with such a view. First, there could be no limit set to what “predetermined dispositions to future judgements” might be made. That is, we could not distinguish a set of concepts marked out for pure employment from concepts whose employment would be illegitimate, so-called usurpatory concepts such as fortune and fate (cf. A84=B117). Kant much earlier had derided this hypothesis as a view that “encourages all sorts of wild notions and every pious and speculative brainstorm” (Ak. X, 132).28 Second, and the reason Kant takes as decisive, this preformation model could not account for the universality and necessity required by the pure concepts of the understanding. “The necessity of the categories, which belongs to their very conception, would have to be sacrificed” (B168). The categories are synthetic concepts whose only legitimate a priori employment is in application to intuitions. To claim such concepts to be implanted subjective dispositions is to ignore the crucial relation between these concepts and intuitions; it is through the application of the concepts in judgment that they can achieve objective universality and necessity. To subscribe to the preformation model is to claim merely that “I am so constituted that I cannot think this representation otherwise than as connected.” Because the fundamental notion behind the Copernican Revolution is that the thinking subject itself connects its representations and unifies them according to rules through the transcendental unity of apperception, preestablished harmony would eliminate the contribution of the thinking subject whereby the objective application of pure concepts is determined and limited. It would be this, Kant 28. Kant here further refers to the “harmonia praestabilitam intellectualem” as a “deus ex machina ..... the greatest absurdity one could hit upon in the determination of the origin and validity of our knowledge.” While the specific model of implantation is sometimes associated with Crusius (cf. Prolegomena, §36; Ak. IV, 319n; Reflexion 4275 (Ak. XVII, 492)), the more general view under attack is associated with Plato, Malebranche, and particularly, Leibniz. In Fortschritte, Ak. XX, 284, Kant refers to preestablished harmony as “the most whimsical [wunderlichste] figment philosophy has ever contrived.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant's Critical Model of the Subject   21 claims, that “the skeptic most desires” (B168). This failure of the ectypal model of the human intellect to be able to justify universal and necessary concepts leads, on Kant’s view, to skepticism. From a different direction, but for ultimately the same reasons, the archetypal model is put in a position vulnerable to skepticism. It would hardly be prudent to enter here into a full-blown discussion of Kant’s notion of the thing-in-itself, nor is such a discussion necessary for present purposes. It is, of course, his notion that the thingin-itself cannot be known, as well as the other, positive claims Kant seems to make about it, that has given rise to the greatest controversy of any of his doctrines.29 What I would simply like to note at this point is that the archetypal and ectypal model have in common an additional failure—by not distinguishing things as they appear from things as they are in themselves, both approaches lead to antinomies, and ultimately the “death of sound [gesunden] philosophy” (A407=B434). This failure, on Kant’s view, requires their rejection, and lends further support to Kant’s own “critical” model of the human subject. For something to be able to be an object, it must be an object of judgment. In this way, the human intellect plays a constitutive role in determining its own objects. The forms of intuition universally and necessarily condition the manifold to be judged; the understanding consists of universal and necessary concepts and principles by which that manifold is judged. Both the forms of intuition and the understanding, then, are brought to experience, or are contributions made by the subject in determining its experience. It follows that an object of possible experience, to be judged (or to be able to be judged), must appear under or within the forms of time and, for external objects, space; an object that is not so conditioned could not be an appearance, and hence could not be an object of possible experience for a finite, discursive intellect. Such an unconditioned object, stripped of those features that make it an object of possible human experience, would then have to be regarded as a thing-in-itself. Such a thing, of course, can still be thought; but the conditions requisite for knowing (cognizing) it are lacking. The point is put succinctly by Melnick: “the very notion of an ob29. As an introduction to some of the difficulties involved, see my “Stoff and Nonsense in Kant’s First Critique,” History of Philosophy Quarterly ( January 1993).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

22   Kant's Critical Model of the Subject ject is essentially the notion of how what is given independently of his judgment is connected epistemically to the subject’s understanding.”30 Of course, that epistemic connection is made through judgment; and with judgment there is a judging subject involved. Using this sketch of Kant’s argument for the unknowability of the thing-in-itself, we can turn to the difficulties he sees for both the archetypal and ectypal models. These difficulties stem from the common failure to acknowledge two specifically different faculties that give rise to cognition, namely sensibility and understanding. The archetypal model, which Kant associates specifically with Leibniz and generally with rationalism, views human knowledge as a confused, inferior brand of that achieved by a divine, intuitive intellect. The ectypal model, which Kant associates specifically with Locke, and generally with empiricism, views the human intellect as essentially passive, simply taking its general concepts inductively from experience. Hence, the summary with which I began, where Kant describes the failure of both models in the “Note to the Amphiboly of Concepts of Reflection”: “Leibniz intellectualised appearances, just as Locke, according to his system of noogony (if I may be allowed the use of such expressions), sensualised all concepts of the understanding, i.e., interpreted them as nothing more than empirical or abstracted concepts of reflection” (A271=B327). As Kant goes on to indicate, not to acknowledge two radically different origins for human cognition neglects the essential distinction between a thing as it appears and a thing as it is in itself:31 “Instead of seeking in understanding and sensibility two completely different sources of representations, which can yield objectively valid judgements of things only in conjunction, each of these great men holds to only one of the two, viewing it as related immediately to things in 30. Melnick, Kant’s Analogies of Experience, 153. 31. As is so often the case, Hume’s position is less easily painted by such broad strokes as these. But as Kant makes explicit in the Critique of Practical Reason, Hume also failed to make the critical distinction, which led to his specific denial of the objective status of causality—“I granted that, when Hume took the objects of experience as things-in-themselves (as is almost always done), he was entirely correct in declaring the concept of cause to be deceptive and an illusion”; Ak. V, 53—as well as to the more pervasive outlook Kant repeatedly refers to as “skepticism” (cf. B128; A764f.=B792; but also Critique of Practical Reason; Ak. V, 13ff., where Kant denies that Hume’s is an “unlimited” skepticism, because of his views on mathematics).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant's Critical Model of the Subject   23

Copyright © 2008. Catholic University of America Press. All rights reserved.

themselves, the other doing nothing except to confuse or to order the representations which this faculty yields” (A271=B327). Without going into the details of the arguments,32 it should at least be noted that Kant labels any philosophy that does not distinguish things as they appear from things as they are in themselves “transcendental realism.” Moreover, it is the transcendental realist “who afterwards actually plays the role of empirical idealist. After wrongly presupposing that objects of the senses, if they are to be external, must have an existence by themselves independently of the senses, he finds that, judged from this point of view, all our sensuous representations are insufficient to establish their actuality [Wirklichkeit]” (A369).33 In contrast, Kant offers as an alternative “transcendental idealism.” The transcendental idealist is at the same time an empirical realist, who no longer must be “scandalized” by previous doctrines in which “the existence of things outside us (from which we derive the whole material of knowledge, even for our inner sense) must be accepted merely on faith” (Bxxxix n.). Thus, the necessity of distinguishing things as they appear from things as they are in themselves serves to discredit any model (whether ectypal or archetypal) that fails to make this distinction and that, consequently, will succumb to contradiction and transcendental illusion, as shown in the Antinomy of Pure Reason.

32. A much more detailed exposition of Kant’s arguments on this issue, with particular emphasis on the Antinomies, can be found in Allison, Kant’s Transcendental Idealism, 35ff.; see especially 51–61. 33. A position with which I am in considerable agreement is Gerd Buchdahl’s; to oversimplify dramatically, I take Buchdahl to be saying that Kant’s “realism” provides as robust a sense of the real as one could desire. Among other places, he argues for this in Metaphysics and the Philosophy of Science (Oxford: Blackwell, 1969); “Realism and Realization in a Kantian Light,” in Reading Kant, ed. E. Schaper and W. Vossenkuhl (Oxford: Blackwell, 1989); “Science and God: The Topology of the Kantian World,” Southern Journal of Philosophy 30, Supplement (1991). (The passage cited at A369 comes from a section of the first edition of the Critique that Kant rewrites completely for the second edition. These revisions, while crucial to understanding the shift to the position Kant adopts in the second edition, are not germane to the present discussion.)

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

24   Kant's Critical Model of the Subject

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Indirect Argument for the “Critical” Model I want to conclude this chapter by characterizing Kant’s strategy as an indirect argument in support of his own model of human cognition. Kant seems to assume three possible such models, the archetypal, the ectypal, and his own as described in the Critique of Pure Reason, which I have been calling the critical model. For a model to be supportable, it must present an adequate account of human cognition, and not lead to contradictions. Both the archetypal and ectypal models fail on both accounts, for the various reasons outlined above. The Transcendental Aesthetic and Transcendental Analytic seek to show that, given Kant’s characterization of sensibility and understanding, objective claims of cognition can be given an adequate account, or be well-grounded. The Transcendental Dialectic, particularly the Antinomy of Pure Reason, reveals the transcendental illusion into which the competing models necessarily fall, and how the contradictions to which those models give rise can be avoided through the Critical philosophy. Hence, if the three models present an exhaustive list, there are good reasons to reject two of them (namely, the archetypal and ectypal), and there is a third model (the critical) that is successful both in accounting for the cognitive claims of human beings and in avoiding contradiction, it is this third model we are compelled to accept. The essence of this model is that the cognizing subject has two heterogeneous faculties [Vermögen], an active faculty of the understanding and a passive or receptive faculty of sensibility. It is probably obvious that the claim that these three models— archetypal, ectypal, and critical—present an exhaustive list smacks of sheer dogmatism. It is certainly a notion worth considering at greater length than Kant gives it. But it is also clear, when Kant does address the issue, that he thinks any philosophical position worth entertaining can be fit into this framework.34 Thus, in the little-discussed final chapter of the Critique of Pure Reason, “The History of Pure Reason,” Kant divides various historical approaches with respect to the objects 34. As Christine Korsgaard has pointed out to me, it may be only in light of Kant’s solution that we would begin to consider the others as alternative models (as either entirely active or entirely passive).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant's Critical Model of the Subject   25

Copyright © 2008. Catholic University of America Press. All rights reserved.

and origins of cognition through pure reason, as well as with respect to method. Throughout his brief account, each philosopher he mentions—Plato, Aristotle, Epicurus, Locke, Leibniz, Wolff, and Hume— is assimilated to either the archetypal or the ectypal model. Having satisfied himself that he has shown both of these to be inadequate and misleading, Kant concludes “the critical path alone is still open” (A856=B884). Underpinning this account is Kant’s conception of a faculty (Vermögen; Fähigkeit). On Kant’s view, a faculty—and here, it is crucial to point out, we are considering only sensibility and the understanding—is either active or passive, and must be able to be characterized as one or the other. That is, anything that qualifies as a faculty must be seen as either active or passive. Kant’s lengthiest description of the passive, or receptive, faculty of sensibility and the active faculty of the understanding is given in his introduction to the Transcendental Logic (A50ff.=B74ff.); perhaps the clearest use of this distinction is found in the often neglected §15 of the second-edition Transcendental Deduction, for example, “of all representations combination is the only one which cannot be given through objects” (B130; cf. B134). In sum, what cannot be “given,” or passively received (albeit within the forms of intuition), must be actively contributed. Thus, given Kant’s reading of the alternative models, and what I take to be his relatively innocuous version of “faculty psychology,”35 the exhaustiveness claim comes to this. The Leibnizian ar35. A full defense of this specific claim is beyond the scope of this account. But, as Beck remarks, “It would be vain to deny that Kant is a faculty psychologist, but I would remind you that the German word for ‘faculty,’ Vermögen, is the noun form of the infinitive ‘to be able.’ The discovery and assessment of what one is able to do seem to be a much less mysterious process than the discovery and assessment of faculties, and they do not lead so readily to unanswerable questions about ‘where’ the faculties are (in the phenomenal or noumenal world) and the like.” “Toward a Meta-Critique of Pure Reason,” in Essays on Kant and Hume, 32. Walsh makes, less explicitly, the same point in “Philosophy and Psychology in Kant’s Critique,” Kant Studien 57 (1966): 197. Of course much earlier, in the Essay Concerning Human Understanding, II.xxi.20, Locke writes, “Faculty, Ability, and Power, I think, are but different names of the same things.” Hans Amrhein usefully argues, with specific reference to the thinking subject, “This Kantian expression ‘Vermögen’ should not be interpreted in the sense of ‘Vermögenpsychologie.’ ‘Vermögen’ here is taken only in the sense of the ability [Fähigkeit] to produce the thought of the ‘I,’ that is, to display the legitimacy of the transcendental apperception in thought.” “Kants Lehre vom ‘Bewußtsein überhaupt’” (Berlin: Kant Studien Ergänzun-

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

26   Kant's Critical Model of the Subject chetypal model subsumes sensibility to the (active) intellect, thus “intellectualizing appearances”; the Lockean ectypal model abstracts pure concepts from (passive) experience, thus “sensualizing” them. Presumably, no one rejects both that which is provided through sensibility and that which is provided through the intellect. The remaining possibility is that human cognition possesses two ineliminable—and irreducible—components, one passive, and one active. Given the original assumptions, then, there are four possible combinations, and Kant has, on this account, critically examined all four. While what Kant is saying here is unquestionably quite elaborate, the logical structure employed is a straightforward disjunctive syllogism. In the second premise of this syllogism, the denial of two of the three disjuncts, we can also see Kant make a strategic move that has since come to be known as “Ramsey’s maxim.” If two positions fail to yield satisfactory results, and contradict each other, “it is a heuristic maxim that the truth lies not in one of the two disputed views but in some third possibility which has not yet been thought of, which we can only discover by rejecting something assumed as obvious by both the disputants.”36 The third possibility, of course, is that radical proposal nowadays referred to as the “Copernican Revolution”—that objects must conform (sich richten) to our cognition, thus shifting the focus from the “given” to the subject and its a priori contribution. Ultimately, however, the argument must remain indirect—as Kant frequently admits, a direct argument is not to be had (if, indeed, such sheft 10, 1909), 84n1. A detailed discussion of this issue is Heinz Jansohn, Kants Lehre von der Subjektivität (Bonn: Bouvier, 1969), esp. 264–84. Hermann Cohen also has an informative account in his Kants Theorie der Erfahrung (Berlin: Cassirer, 1925), 292ff. Kitcher provides an up-to-date discussion in Kant’s Transcendental Psychology, with particular attention to contemporary issues in cognitive science. She also mentions the tendency to take Kant’s description of capacities as indicative of a problematic “faculty psychology”: “Faculty terminology makes confusion [between these] quite easy” (14). 36. Frank Plumpton Ramsey, “Universals,” in The Foundations of Mathematics (New York: Humanities Press, 1950), 115–16; my emphasis. I owe this reference to L. W. Beck’s “Kant’s Strategy,” in Essays on Kant and Hume, 11. Kant relies on this move in a variety of ways—in his characterization of the thinking subject, at least the first two Antinomies, and in his general argument in support of Transcendental Idealism; see, e.g., A503=B531. All revolve around Kant being willing to entertain this third possibility, and propose his hypothesis in metaphysics analogous to the “first thoughts of Copernicus” (Bxvi).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant's Critical Model of the Subject   27 an argument as such is needed). For a finite intellect, forms of intuition other than space and time, “even if they should be possible, we cannot render in any way conceivable to ourselves [können wir uns doch keinerleiweise erdenken und faßlich machen]” (A230=B283). Furthermore, he observes in the second-edition Transcendental Deduction that for this finite intellect, which thinks and does not intuit, synthesis is necessary. Indeed, that this act of synthesis is necessary serves undeniably as “the first principle of the human understanding, and is so indispensable to it that we cannot form the least conception of any other possible understanding, either of such as would intuit itself, or of any that may possess an underlying mode of sensible intuition which is different in kind from that grounded in space and time” (B139). Kant puts the general point most strongly in the letter to Herz I have already referred to: “[W]e are absolutely unable to explain further how it is that a sensible intuition (such as space and time), the forms of our sensibility, or such functions of the understanding as those out of which logic develops are possible” (Ak. XI, 51)37 Nicholas Rescher has called the kind of claim involved here a “surd fact,” which every philosophical system “is bound to come up against.”38 We don’t, Kant argues, possess an intuitive intellect that produces, even in a confused, inferior fashion, its objects through the thought of its representations. Nor do we possess a wholly passive intellect, one that simply copies impressions from things themselves. Rather, the critical model of human cognition Kant presents has both a passive faculty of receiving intuitions under (within) the forms of space and time and an active faculty that judges these intuitions employing synthetic concepts a priori. We cannot prove demonstratively that this must be the case; but we can show why other proposed views fail to provide adequate, 37. In his polemic Über eine Entdeckung, Kant spells out some of the details of his position as they contrast with the Leibnizian-Wolffian view; cf. Ak. VIII, 249–50. Here again, after denying that the human being has intellectual intuition and must have two “completely heterogeneous sources of knowledge,” Kant admits we cannot provide a reason for why this is the case; “This we could not (nor can anyone else) further explain.” See also Prolegomena (Ak. IV, 318 (§36)). 38. Nicholas Rescher, “Kant and the ‘Special Constitution’ of Man’s Mind,” in Akten des IV Internationalen Kant-Kongresses (1974), Part II.1, 321. Cf. W. H. Walsh, Kant’s Criticism of Metaphysics (Chicago: University of Chicago Press, 1976), 253.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

28   Kant's Critical Model of the Subject non-contradictory models. Thus, as L. W. Beck remarks, “We cannot show, as Kant repeatedly confesses, why they [these human cognitive faculties] must be so and not otherwise; but it is something to show good reasons why they are as they are.”39

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Judging Subject and Objective Validity At the close of the second edition’s preface, Kant issues a preemptive warning to his readers and critics: “If we take single passages, torn from their contexts, and compare them with one another, apparent contradictions are not likely to be lacking, especially in a work that is written with any freedom of expression” (Bxliv). Central to this context is Kant’s characterization of the human being; stripped of this, many of Kant’s claims sound ludicrous. As he admits, in describing this human subject as the “lawgiver of nature,” “However exaggerated and contradictory it may sound, to say that the understanding is itself the source of the laws of nature, and so of its formal unity, such an assertion is none the less correct, and is in keeping with the object to which it refers, namely experience” (A127). It is not difficult to see why early reviewers of the Critique took such statements as committing Kant to the kind of idealism espoused by Descartes and Berkeley. The best-known version of this response is the notorious Garve-Feder review of 1782, which in fact turned out to have very beneficial consequences for the student of the Critique.40 39. Beck, “Toward A Meta-Critique of Pure Reason,” 37. I am willing to agree with Beck and others that Kant did not explicitly develop the question of metacritique, as first broached by J. G. Hamann. As Frederick Beiser points out, Kant takes up some of the issues involved in the Opus Postumum (Ak. XXI, 81–100), and Beiser is certainly correct that these “scattered and inchoate remarks do not amount to an explicit and general meta-critical theory.” The Fate of Reason (Cambridge, Mass.: Harvard University Press, 1978), 330n14. A relatively early Reflexion indicates, however, that Kant was well aware of the problem: “I seek in an understanding, that requires rules, the knowledge [Kenntnis] of these rules themselves. This is paradox;” Refl. 1592, Ak. XVI, 28. (Adickes is uncertain about the date of this remark, simply putting is somewhere between 1764 at the earliest and 1775 at the latest.) It may turn out that the problem was (or would be) seen by Kant (of the First Critique, at least) as superfluous or possibly— as Beck suggests (“Toward a Meta-Critique of Pure Reason,” 33)—that if there is to be a metacritique, it must be “internal.” 40. For some of the details of this incident, see Kant’s letter to Garve, 8.7.83; Ak. X,

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant's Critical Model of the Subject   29 For this review annoyed Kant (to put it mildly) so much that he felt compelled to correct its misunderstandings in the Prolegomena. Furthermore, many of the important revisions for the second-edition Critique are made in reaction to what Kant regards as a palpable misinterpretation of the type of idealism to which he is committed. If we keep in mind what kind of subject is involved here, it begins to seem less preposterous to regard its intellectual faculty as the “lawgiver of nature.” By filling in this context, we come to see that this claim is radically distinct from claiming that the subject is responsible for any of the content (as created or self-posited) of nature, or experience. Such would be the case where the subject possesses intellectual intuition, as an inferior version of a divine, archetypal intellect. Rather, a subject that has only sensible intuition gives order (“formal unity”) to a world it does not create, through judgment.41 In this way, Kant gives due to the tough-minded empirical side of his philosophy that runs throughout the Critique. We are in a passive relation to the sensible manifold we receive within the forms of space and time; but to make sense of that manifold requires an active contribution by a thinking subject, who unites that manifold in making judgments about it. To perform this active function is to be the lawgiver of nature. That the subject makes some kind of contribution is not controversial. The difficulties arise in determining what that contribution is, what its modal status is, and what its limits are. Kant wants to show that the active role of the subject is essential to experience, yielding a systematic and, in some sense, exhaustive set of rules that are universal and necessary for that experience. At the same time, the subject is constrained by limiting the legitimate application of these rules to the domain of possible experience. Kant here navigates between the Scylla of an uncritical empiricism (by establishing secure foundations of transcendental 336–43. The relevant documents are collected in the appendices to the Philosophische Bibliothek edition of the Prolegomena, edited by Karl Vorländer. 41. Or better, does not wholly create. The notion of “world” takes on a meaning that must be made more precise in the wake of Kant’s philosophical revolution, an issue I have taken up in chapter three of my dissertation (see n10) in connection with some of the views Richard Rorty has advanced. It should also be pointed out that these remarks are not taken to characterize Kant’s description of the practical employment of reason.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

30   Kant's Critical Model of the Subject analytic) and the Charybdis of dogmatic metaphysics (by curbing its dogmatic pretensions). Kant makes clear where he thinks he is headed: “It is the land of truth—a charming name!—surrounded by a wide and stormy ocean, the native home of illusion, where many a fog bank and many a swiftly melting iceberg give the deceptive appearance of farther shores, deluding the adventurous seafarer ever anew with empty hopes, and engaging him in enterprises which he can never abandon and yet is unable to carry to completion” (A235–36=B294–95). A subject that is in a wholly passive relationship to its representations does not unify those representations in making judgments, but simply reacts to what is given. To use Kant’s example, a dog distinguishes bread from roast meat solely on the basis of being affected differently by them.42 The dog has only the ability to connect its representations according to empirical (and hence contingent) laws of association. In contrast, the human being unifies all its representations (or at least can unify them) by means of the “I think,” allowing it to judge its representations according to universal and necessary rules. Kant refers to this capacity as yielding the “objective ground of all appearances,” and calls it “affinity,” sharply distinguishing it from association (A122; cf. A113).43 Without affinity, the animal lacks this objective ground, and would fail to unify its representations. As we will see below, the very notion of what it means to have experience is lost for this (ectypal) intellect. And if “truth” is taken as “the agreement [Übereinstimmung] of our knowledge with objects” (A191=B236; A157=B197; A237=B296; A820=B848; cf. A57–58=B82–83), the notion of truth is equally lost. Hence, the crucial role the active judging subject performs, for “truth or illusion is not 42. The example is from Kant’s early essay on the mistaken subtlety of the four figures of the syllogism; “Spitzfindigkeit,” (Ak. II, 60). Kant very probably has in mind the alleged claim of Chrysippus that even dogs demonstrate an ability to reason syllogistically, declaring that “the dog makes use of the fifth complex indemonstrable [viz., disjunctive] syllogism when, on arriving at a spot where three ways meet, after smelling at the two roads by which the quarry did not pass, he rushes off at once by the third without stopping to smell.” Sextus Empiricus, Outlines of Pyrrhonism, trans. R. G. Bury (Cambridge: Harvard University Press, 1933), I.69. 43. See Henry Allison, “Transcendental Affinity—Kant’s Answer to Hume?” in Kant’s Theory of Knowledge, ed. L. W. Beck (Dordrecht: Reidel, 1974), 119–27. Kant replaces the notion of affinity in the second-edition Deduction, but the basic unifying function is preserved.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant's Critical Model of the Subject   31 in the object [Gegenstande], in so far as it is intuited, but in the judgement about it, in so far as it is thought....... Truth and error, therefore, and consequently also illusion insofar as it leads to error, are to be found in the judgement, i.e. only in the relation of the object to our understanding” (A293=B350; my emphases). From a different direction, the archetypal intellect also has no use of this notion of truth. As pointed out above, this intellect creates its objects in thinking them; hence there is no question about the agreement between knowledge and object—only the clarity of the thought is at issue. Again, what is lacking is judgment, a function that essentially characterizes human beings and the (truth-evaluable) process by which they gain knowledge. It is, I think, remarkable in fact just how little Kant says about truth in the Critique. When he does take up the question, he invariably couches his discussion in terms of conditions for assigning truthvalues—what Gerold Prauss refers to as “Wahrheitsdifferenz.”44 A typical example is given in the Second Analogy: “Since truth consists in the agreement of knowledge with its object, it will at once be seen that we can here only enquire regarding the formal conditions of empirical truth” (A191=B236). Such formal conditions are, of course, those provided by transcendental analytic. Kant, then, is not so much interested in the truth of a given proposition “p”; rather, he is concerned with establishing the universal and necessary conditions that allow “p” to be assigned a truth-value. Assuming “p” is a claim of possible experience—that is, conforms to the rules of general logic and transcendental analytic—the content of that proposition is contingent upon what is given in the relevant sensible manifold. And because of that contingency, “[a] sufficient and at the same time general sign [Kennzeichen] of truth cannot possibly be given. Since we have already called the content of a cognition its matter, we must be prepared to say that of the truth of the matter of cognition, no general criterion can be demanded. Such a criterion would by its very nature [in sich selbst] be self-contradictory” (A59=B83; cf. A480=B508). Beyond the formal conditions imposed by a thinking subject, the 44. Gerold Prauss, “Zum Wahrheitsproblem bei Kant,” Kant Studien 60 (1969): 181; Erscheinung bei Kant (Berlin: De Gruyter, 1971), 275ff.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

32   Kant's Critical Model of the Subject truth of a given empirical proposition must be determined by returning to what Kant calls in the Prolegomena “the fruitful bathos of experience” (Ak. IV, 373n). Thus transcendental analytic cannot by itself yield the truth; it serves only as a “logic of truth” (A62=B87). But this entire scenario is relevant only for a subject that thinks and judges by means of concepts (discursively), relative to a passively received sensible manifold (conditioned by the forms of intuition). For such a subject, “the business of the senses is to intuit, that of the understanding is to think” (Ak. IV, 304 (§22)). When trying to come to grips with what Kant means by various notions—objects of possible experience, the thing in itself, truth, just to name three—it is crucial to keep in mind that their meanings are to be determined relative to this kind of subject. In seeking a satisfactory response to that fundamental question that motivates the Transcendental Analytic in general, and the Transcendental Deduction in particular—how can subjective conditions of thought have objective validity? (A89–90=B122)—it is equally important to see that this question is one that arises only for this specific kind of subject, and must be answered in terms of that subject’s capacities and limitations. To see this, we can look briefly at a passage that contains a central step in the second-edition Transcendental Deduction: “Understanding is, to use general terms, the faculty of cognitions. These consist in the determinate relation of given representations to an object [Objekt]; and an object is that in the concept of which the manifold of a given intuition is united. Now all unification of representations demands unity of consciousness in the synthesis of them” (B137). The key notion here is that an object, in its most general sense, is a product of a unifying synthesis and a manifold to be synthesized. (“The understanding can intuit nothing, the senses can think nothing. Only through being unified can knowledge arise” (A51=B75–76).) And, as Kant notes, that synthesis requires a unity of consciousness. As outlined above, these are features characteristic of the human subject that possesses two heterogeneous faculties. The archetypal intellect that intuits intellectually (albeit imperfectly, in the case of human beings) lacks the essentially sensible faculty by which representations are given. The ectypal intellect lacks the unity of consciousness that

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant's Critical Model of the Subject   33 allows it to take up, combine, and judge this manifold by means of synthetic concepts and principles a priori—activities of an essentially spontaneous, discursive intellect. Only on Kant’s critical model, presenting a subject with heterogeneous, yet complementary, faculties, does it make sense to call an object “that in which the manifold of a given intuition is united.” If this line of reasoning is sound, then seeing what Kant means by an “object” involves taking into account the constitutive role played by the judging subject. And this is precisely what was proposed by the Copernican Revolution—instead of seeing how our knowledge conforms to objects, let us begin with the subject, in order to determine how objects conform to its knowledge. Again, this is a revolution that is applicable only to, or that makes sense only for, a discursive intellect that judges about a world it doesn’t create. A result of this shift is Kant’s view that human beings can obtain cognition, or have experience, of objects only as they appear, not as they are in themselves. Many of Kant’s successors made it abundantly clear that this was not a particularly welcome consequence of that revolution. But as I suggested above, it simply follows from what it means for something to be an object of possible experience—it must meet certain (necessary) conditions: no conditions, no object. Ultimately, this yields a fundamental result of Kant’s revolution—an entirely new conception of the world, and the human being’s relation to it. If we are to understand what Kant undertakes to do in the first Critique, we must come to grips with this fundamental question of the Transcendental Analytic, “how can subjective conditions of thought have objective validity?” One approach is to reformulate this as “how can the thinking subject impose universal and necessary conditions for the cognition of objects to be possible?” This version makes the reference to the subject explicit; as I hope I have shown, this is a question that concerns—is meaningful only for—the specific kind of subject Kant considers, in sharp contrast to the problematic conceptions of the archetypal and ectypal intellects. Indeed, it is only within this context that Kant’s project has any hope for success.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Ch a pter T wo

Kant’s Conception of General Logic e

Copyright © 2008. Catholic University of America Press. All rights reserved.

General Logic and Aristotle Given his preoccupation with logic in the Critique of Pure Reason, it is an understandable hope that Kant might use the term “logic” in a clear-cut, univocal fashion throughout the text. Unfortunately, such a hope is mere fantasy; Kant uses the term in a bewildering variety of ways, at times making it close to impossible to determine whether he is referring to (among others) general logic, transcendental logic, transcendental analytic, a “special” logic relative to a specific science, a “natural” logic, a logic intended for the “learned” (Gelehrter), some hybrid of these logics, or even some still more abstract notion that ranges over all of these uses. The difficulties are compounded when one takes into consideration the history of logic and how Kant’s own conception of logic developed and changed throughout his career. Is logic, as the Stoics argued, a part of philosophy, or should it be viewed as it was by the followers of Aristotle, as merely an instrument (organon) of philosophy, or should one follow the Platonists and consider it both?1 Kant, 1. See the brief comment by Jan Lukasiewicz in Aristotle’s Syllogistic from the Standpoint of Modern Formal Logic (Oxford: Clarendon Press, 1951), 13, and particularly his citation of Ammonius’s Commentary on Aristotle’s Prior Analytics (given in note 1).

34

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant’s Conception of General Logic   35 at various times, took logic to be “an art (Kunst), a science (Wissenschaft), prudence (Klugheit), wisdom (Weisheitslehre), or an instrument (Instrument).”2 This list hardly exhausts the terms Kant uses throughout writings on logic, as one can quickly see in his numerous lectures on logic and his own notes (Reflexionen) on logic and philosophy. I have no intention of trying here to disentangle the complicated developmental history of Kant’s view of logic; in any case, outstanding work has been done on this topic by Giorgio Tonelli, Riccardo Pozzo, and others.3 Rather, I have the more modest goal of coming to grips with Kant’s sufficiently complex use of “logic” within the Critique of Pure Reason. To begin to do so, I will turn to what is almost certainly Kant’s best-known—indeed, notorious—remark about logic. In the Preface to the second edition of the Critique, Kant comments:

Copyright © 2008. Catholic University of America Press. All rights reserved.

Daß die Logik diesen sicheren Gang schon von den ältesten Zeiten her gegangen sei, läßt sich daraus ersehen, daß seit dem Aristoteles keinen Schritt rückwärts hat tun dürfen, wenn man ihr nicht etwa die Wegschaffungeiniger entbehrlicher Subtilitäten, oder deutlichere Bestimmung des Vorgetragenen als Verbesserungen anrechnen will, welches aber mehr zur Eleganz, als zur Sicherheit der Wissenschaft gehöhrt. Merkwürdig ist noch an ihr, daß sie auch bis jetzt keinen Schritt vorwärts hat tun können, und also allem Ansehen nach geschlossen und vollendet zu sein scheint (Bviii). This is only one of a large set of questions that can be raised about the relationship among the history of logic, Kant’s logic, and his understanding of that history. 2. See Ricardo Pozzo, Kant und das Problem einer Einleitung in die Logik (Frankfurt a.M.: Lang, 1989), 42. 3. In addition to Pozzo’s work cited in note 2 above, see “Kant within the Tradition of Modern Logic,” Review of Metaphysics 52 (December 1998): 295–310, his George Friedrich Meier’s Vernunftlehre: Eine historisch-systematische Untersuchung (Stuttgart: Frommann-Holzboog, 2000), and Pozzo and M. Oberhausen, eds., Vorlesungsverzeichnisse der Universität Königsberg 1720–1804: Mit einer Einleitung und Registern (Stuttgart: Frommann-Holzboog, 1999). Tonelli’s remarkable historical research is scattered throughout a number of journals, in Italian, German, English, and French; only part of this is collected in his unfinished Kant’s Critique of Pure Reason within the Tradition of Modern Logic, ed. David Chandler (Hildesheim: Olms, 1994); a useful bibliography of Tonelli’s work may be found in C. Cesa and N. Hinske, eds., Kant und sein Jahrhundert (Frankfurt a.M.: Lang, 1993), 187–204. See also Elfriede Conrad, Kants Logikvorlesungen als neuer Schlüssel zur Architekton der Kritik der reinen Vernunft (Stuttgart: Frommann-Holzboog, 1994).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

36   Kant’s Conception of General Logic I have presented the text in the original, for it is worth looking at how Kant’s central claim here has been translated. As Kemp Smith renders the emphasized passage, Kant states that logic “to all appearances [is] a closed and completed body of doctrine,” while Pluhar’s version reads, “Another remarkable fact about logic is that thus far it also has not been able to advance a single step, and hence is to all appearances closed and completed.” Guyer and Wood render the claim in a way similar to Pluhar’s, that logic “until now ..... has also been unable to take a single step forward, and therefore seems to all appearances to be finished and complete.” It is clear that Kemp Smith employs a term—“doctrine”—that not only is not justified by the text but complicates issues by introducing a technical term from Kant’s philosophical vocabulary. Kant characterizes “doctrine” as “an attempt to provide a priori an expansion for the understanding in the field of pure cognitions” (A135=B174), that is, to extend the range of pure cognition in its a priori employment. While Tonelli argues that general logic serves as a doctrine of the general form of thought,4 it is not at all clear that general logic qualifies under this description. However, more important than this terminological dispute is the overall tenor of Kant’s remark here: is he in fact saying that since Aristotle there need be no more concern about logic as a discipline or a field of study, that Aristotle (with some minor embellishments, in terms of presentation) is the last word in logic? Certainly that is how Kant has almost invariably been understood. Thus Hegel relies on such a view when he writes: “Now if logic has not undergone any change since Aristotle—and in fact, judging by modern compendiums of logic the changes frequently consist mainly in omissions—then surely the conclusion which should be drawn is that it is all the more in need of a total reconstruction.” As cited earlier, C. S. Peirce similarly observes that “we are to remember that, according to Kant, nothing worth mention had been contributed to logic since Aristotle”; P. F. Strawson notes that Kant believed “without question” in the “finality” of Aristotelian logic; E. W. Beth clearly has this standard interpretation in mind when he writes that “Aristotle’s logical 4. Tonelli, Kant’s Critique, 104.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant’s Conception of General Logic   37 work, far from being a closed system, already contained the germs of future development.” Such comments are commonplace in the literature, both on Kant and on the history of logic; as Kneale and Kneale sum up what must be called the “received” view, Kant was “apparently unaware of the value of any contributions made to logic after the time of Aristotle, and that the doctrine which he regarded as the complete and perfect discovery of Aristotle was in fact a peculiarly confused version of the traditional mixture of Aristotelian and Stoic elements.”5 Kant himself, of course, is largely responsible for this interpretation. However, I believe that at best it is historically naive and that at worst it gives a very misleading picture, both of Kant’s understanding of the history of logic and of his conception of general logic and its role in philosophical inquiry. The alternative interpretation I outline below will reveal that Kant’s conception of general logic is considerably more sophisticated and historically informed than he has traditionally been given credit for; more important, on this reading Kant’s strategy in the First Critique, grounded as it is in logic, becomes more plausible, more defensible, and, consequently, more attractive.

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Role of General Logic General logic “treats only the logical form in the relation of cognitions to one another”; its concern is “the form of thought in general” (die Form des Denkens überhaupt) (A55=B79). In Kant’s published work on logic, including the First Critique, as well as “The False Subtlety of the Four Syllogistic Figures,” and in his lectures and notes on logic, two particularly important points emerge. First, Kant struggles to articulate the notion of general logic as the most abstract and strictly formal of investigations, revising and reformulating his characterization of logic to exclude any and all extraneous issues that might inter5. G. W. F. Hegel, “Introduction” to Wissenschaft der Logik 1, ed. G. Lasson (Hamburg: Meiner Verlag, 1932), 32–33; C. S. Peirce, Elements of Logic §39, in Collected Papers of Charles Sanders Peirce, ed. C. Hartshorne and P. Weiss, vol. 2 (Cambridge: Harvard University Press, 1931); Strawson, The Bounds of Sense; E. W. Beth, “Hundred Years of Symbolic Logic,” Dialectica 1 (1947): 332; W. Kneale and M. Kneale, The Development of Logic (Oxford: Clarendon Press, 1962), 355.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

38   Kant’s Conception of General Logic fere in presenting logic as such an investigation. Second, throughout his discussion of general logic, the focus is on rules. To see this first point, it is helpful to keep in mind that one can just as easily refer to general (allgemeine) logic as to universal logic: the point is to identify, articulate, and—if possible—justify the principles of such a logic insofar as they range over all possible thought. Given Kant’s fundamental position that all thought must be expressed in the form of judgments, general logic provides a set of universal and necessary conditions for the very possibility of judgment. In the Critique, Kant particularly emphasizes the notion of “form”: general logic “abstracts from all content of the cognition of the understanding and from all differences in its objects, and deals with nothing but the mere form of thought” (A54=B78); it serves as a “canon of understanding and reason, but only in respect to what is formal in their employment” (A53=B77). On Kant’s view, the thinking subject judges, and the faculty of judgment is the same as the faculty of thought (A69=B94; A81=B106). A given judgment must conform to certain necessary syntactic conditions, or rules, to be well-formed. Such rules are provided by general logic. In general logic we abstract from all content of cognition, and “consider only the mere form of understanding” (A70=B95). Only the formal relations between and among representations are involved, regardless of their source or what they represent. (As Paton reminds us, it is not that general logic has no object, but that “it ignores differences in objects.”)6 Kant’s claim, then, is that the act of judging reveals the capacity to unite the manifold of representations under the unity of thinking in general and that the various syntactic moments of this act can be displayed in a systematic table of judgments that provide a set of general or universal rules for the possibility of thought. By abstracting from all content, the analysis of a given judgment provided by general logic is formal and can simply treat the relations between the judgments themselves. Whether I make the judgment 6. H. J. Paton, Kant’s Metaphysic of Experience (New York: Macmillan, 1936), 1:191n1. As Kant himself notes in the Philosophische Enzyklopädie, “Logic thus teaches us nothing about objects”—literally, it teaches us something about no objects (Ak. XXIX.1.1, 13).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Conception of General Logic   39 “Socrates is mortal” or “Socrates is made out of Fontina cheese,” the task of general logic is to provide a formal characterization of the judgment and an analytical classification of the terms in the judgment. Here the subject term is “Socrates,” the predicate terms “mortal” and “made out of Fontina cheese,” respectively; the judgment is singular, affirmative, categorical, and assertoric. The categorical judgment simply asserts the relation between the subject and predicate terms, and general logic is not even in the position to indicate which is which. (Abstracted from all content, any universal affirmative categorical judgment “All S are P” can always be converted per accidens to the particular affirmative “Some P are S.”)7 It is only when we bring the subject term under its category—a function of transcendental logic—that we can determine what must be taken as subject and never as mere predicate (see B128–29). We see the same point made with other examples Kant discusses: “The soul is not mortal” can be obverted to “The soul is non-mortal,” and the distinction for general logic is simply that between a negative particular judgment (which is said to “distribute” the predicate term) and an affirmative particular judgment (which distributes neither subject nor predicate term). And in a hypothetical judgment, which asserts problematically a relation between two judgments, “it is only the logical sequence [Konsequenz] which is thought by this judgement” (A73=B98).8 Furthermore, it makes no difference for general logic whether the judgment whose form it characterizes is analytic or synthetic, for this distinction, as Kant makes clear in the Prolegomena (Ak. IV, 266), is one of content. Thus, in his discussion of general logic, Kant uses a traditional analytic judgment (“All bodies 7. In general logic, certain logical operations—subalternation, opposition, conversion, obversion—are required to reduce syllogisms to the first figure, the only one Kant accepts as legitimate. The details are given in the short piece on the “false subtlety” of the syllogistic figures (Ak. II, 45–62). Kirk Dallas Wilson raises difficulties for Kant’s reductionist attempt in “The Mistaken Simplicity of Kant’s Enthymematic Treatment of the Second and Third Figures,” Kant Studien 66 (1969): 404–12. An extensive commentary on this early work of Kant’s is Francis Courtes’s French translation (Paris: Vrin, 1972). 8. Here (as elsewhere) there is at least a hint of the Stoic contribution to the history of logic; as Brandt has pointed out, Kant follows Leibniz in extending Aristotelian syllogistic by means of the Stoic form of inference. See R. Brandt, Die Urteilstafel: Kritik der reinen Vernunft A67–76; B 92–101 (Hamburg: Meiner Verlag, 1991), 69n44.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

40   Kant’s Conception of General Logic are divisible”) and a hypothetical judgment composed of two presumably synthetic judgments (“If there is a perfect justice, the obstinately wicked are punished”). General logic, as it were, is blind to the distinction between analytic and synthetic judgments: “The explanation of the possibility of synthetic judgements is a problem with which general logic has nothing to do; indeed, it may not even know its name” (A154=B193). A fortiori, general logic has nothing to do with explaining the possibility of synthetic judgments put forth a priori. The other contribution general logic makes is that of making concepts distinct. Kant summarizes the point in introducing the Analytic of Principles: because general logic abstracts from all content, “nothing remains for it but the business [Geschäft] of analytically dividing the mere form of cognition into concepts, judgements, and inferences, and thereby achieving formal rules for all use of the understanding” (A132–33=B171–72). Kant clarifies this somewhat in the Jäsche Logik, by noting that a concept is “a general representation or a representation of what is common to several objects” (Ak. IX, 91); a concept, that is, collects under a general term a variety of representations that share in that concept. Because no concept refers to an object immediately, a concept refers (sich beziehen) to an object “mediately by means of a mark [Merkmal] which several things have in common” (A320=B377). To clarify a concept, or to make a concept distinct by analysis of those representations that constitute it, is to produce, in Kemp Smith’s terminology, an “analytic exposition,” and this is the task of general logic.9 Thus, given a judgment, it is then general logic’s role to analyze, or make clear, the concepts in that judgment. But these concepts must first be given, through synthesis, for what has not already been combined synthetically cannot be analyzed; an “analytic exposition” presupposes a synthesis to yield that which is to be clarified. “Before we can analyze our representations, the representations must themselves be given, and therefore as regards content no concepts can first arise by 9. Kant says that because general logic abstracts from all content, “bleibt ihr nichts übrig, als das Geschäft, die bloße Form der Erkenntnis in Begriffen, Urteilen, und Schlüssen analytisch auseinander zu setzen.” Kant usually reserves Erörterung for exposition in the technical sense, as in the Transcendental Aesthetic (see B38), where he gives expositio as its Latin equivalent.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant’s Conception of General Logic   41 way of analysis” (A77=B103). As we will see, this provides Kant with a crucial contrast between general and transcendental logic: “By means of analysis different representations are brought under one concept (a procedure [Geschäft] treated by general logic). Transcendental logic, in contrast, teaches how to bring under concepts not the representations but the pure synthesis of representations” (A78=B104; Kant’s emphasis). Here again we see that Kant uses the term Geschäft to characterize the task of general logic—to provide an “analytic exposition” of a concept, or to make a concept distinct. In sum, general logic is a logic of all thinking in general, abstracted from all cognitive content, and from all reference to objects. It provides a set of syntactic conditions ranging over the formal elements that make possible the expression of a judgment, and makes clear (or clearer) the conceptual elements that make up a judgment. But it yields no content through its function, operating only relative to that given to it, and thus presupposes the synthetic activity of the understanding. In essence, then, general logic is a “logic limited to the use of judgements in reference to each other” (A71=B96–97), and serves as a set of formal rules that identify the universal and necessary conditions for the possibility of judgment simpliciter.

Copyright © 2008. Catholic University of America Press. All rights reserved.

General Logic as A Priori Below, particularly in discussing the “Metaphysical Deduction,” I will turn to some of the specific details of Kant’s account of general logic. But here it may be better to postpone looking at those details in order to gain a clearer understanding of what, from a very broad “meta-logical” perspective, general logic is thought to accomplish. Viewed from this perspective, we will see why Kant says what he does about Aristotle’s logic, and will further see how those remarks can be interpreted in a way that makes them considerably more plausible than they have traditionally been regarded. One might describe Kant’s approach to general logic as a reflective attempt to characterize the fundamental conditions that serve as constraints on rationality, that is, on the capacity to reason. In Kant’s own language, such conditions are absolutely necessary and strictly

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

42   Kant’s Conception of General Logic universal, each of which Kant regards as a sufficient and independent criterion for the a priori (B4). Thus general logic provides the a priori conditions for the rational expression of thought, or for the possibility of judgment; indeed, for rationality itself. In his relatively brief discussion of the principles of general logic in the First Critique, Kant focuses—as does Aristotle—on the principle of non-contradiction.10 Aristotle describes the principle of non-contradiction, in Book Four of the Metaphysics, as “the most certain of all principles,” stating that it is “what one necessarily understands who understands anything.”11 Aristotle glosses the principle in this fashion: “For the same thing to hold good [υπαρχειν] and not to hold good [µηυπαρχειν] simultaneously of the same thing and in the same respect is impossible" (1005b19–22). This principle, along with the principle of identity and the principle of excluded middle, have entered the history of logic as the "laws of thought"; they can be summarized briefly as follows:12

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Law of Identity: Whatever is, is. The Law of Contradiction: Nothing can both be and not be. The Law of Excluded Middle: Everything must either be or not be.

Kant’s version of the principle of (non)contradiction reinforces his commitment to the austerity and strict formality of general logic; as one might expect, it is formulated quite abstractly (while maintaining the syllogistic language of “belonging to” [zukommen]): the principle of contradiction asserts that “no predicate that contradicts a thing can belong to it” (A151=B190); the principle is, again, “without content and merely formal” (A152=B191). Kant also notes that this principle has on occasion been “carelessly formulated” by introducing a synthetic element. Conse10. See Tillmann Pinder, “Kants Begriff der Logik,” Archiv für Geschichte der Philosophie 61 (1979): § V. 11. “αυτη δη πασων εστι βεβαιοτατν των αρχων”; “ο δε γνωριζειν αναγκαιον τω οτιουν γνωριζοντι” Metaphysics IV (Γ), 1005b22–23, 1005b16–17, as translated in Christopher Kirwan, Aristotle’s Metaphysics, Books Γ, ∆, Ε (Oxford: Clarendon Press, 1971). Cf. 1005b11–12: “A principle about which it is impossible to be in error is the firmest of all.” 12. W. Stanley Jevons, Elementary Lessons in Logic (London: MacMillan, 1889), 117. Discussions of the “laws of thought” became commonplace in the nineteenth century; Jevons’s account is representative.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant’s Conception of General Logic   43 quently he objects to the version that states, “It is impossible that something should at one and the same time both be and not be.”

Copyright © 2008. Catholic University of America Press. All rights reserved.

Apart from the fact that the apodeictic certainty is superfluously added (through the word “impossible”), which must be understood from the proposition itself, the proposition is affected by the condition of time, and as it were says “A thing = A, which something = B, cannot at the same time be non-B, although it can easily be both (B as well as non-B) in succession.” ..... Now the principle [Satz] of contradiction, as a merely logical principle [Grundsatz], must not limit its claims to temporal relations. (A152=B192–193)

So clarified, Kant’s emphasis here is on identifying a principle that must hold for any and all expressions to be evaluated as meaningful, in the broadest sense of “meaning.” As we will see, Kant sharply differs from Aristotle on what logic tells us, specifically in terms of its ontological and metaphysical payoff. Yet, given that proviso, Kant would agree with Aristotle that one who consciously rejects this fundamental principle has forfeited his or her claim on rationality itself. As Aristotle famously—if not entirely clearly—puts the point, in speaking of one who putatively rejects the principle of non-contradiction: “But even this can be demonstrated to be impossible, in the matter of a refutation [ελεγκτικωσ], if only the disputant says something. If he says nothing, it is ridiculous to look for a statement in response to one who has a statement of nothing, in so far as he has not; such a person, in so far as he is such, is similar to a vegetable [φυτω]” (1006a11–15). For Kant, as well, one who attempts to maintain, consciously and explicitly, a contradiction, is reduced to thinking (or saying) nothing, or at least nothing that could be meaningful; at best what is asserted could be made meaningful only by explaining away what then becomes only an apparent contradiction. To be sure, explicit and obvious contradictions may be relatively rare, and frequently a great deal of work must be done to expose them.13 Nor is Kant denying that agents may 13. Indeed, one might well argue that this is both the strategy and the strength of the early Platonic dialogues, where Socrates, through his elenchos, makes clear that a given position implicitly contains a contradiction. More formally, when successful, the reductio ad absurdum of a claim “p,” with the resulting “p → ~p/~p” or “p → (q & ~q)/~p” can produce devastatingly powerful refutations.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

44   Kant’s Conception of General Logic

Copyright © 2008. Catholic University of America Press. All rights reserved.

in fact hold positions that, on inspection, violate the principle of contradiction; rather, he is claiming that when such a violation is exposed, the rational agent has an obligation to eliminate its source. Thus it is no surprise to see Kant repeatedly emphasize, particularly in his lectures and notes specifically devoted to general logic, the normativity of logic. General logic treats not “the rules according to which we do think, but should think” (Refl. 1627; Ak. XVI, 43); “Logic will be concerned with the laws of the understanding, not those in accordance with which one does proceed, rather those according to which one should proceed” (Logik Phillipi; Ak. XXIV.1, 339). In the lectures collected as the Wiener Logik, written during the time the First Critique was being prepared, Kant expands on the notion: “Some logicians presuppose psychology in their logic. Since this is an empirical science, there would arise from this a science of how we think under various hindrances, not a science of how we should think. There would be nothing but contingent and natural laws. But that is not what we are asking about. Logical rules must be taken from the necessary use of the understanding” (Ak. XXIV.2, 791–92).14 Kant here not only explicitly contrasts logic with psychology, but does so precisely by calling attention to the normativity of the laws of logic, just as does Frege, the most tireless critic of psychologism in logic: Any law asserting what is, can be conceived as prescribing that one ought to think in conformity with it, and is thus in that sense a law of thought. This holds for laws of geometry and physics no less than for laws of logic. The latter have a special title to the name “laws of thought” only if we mean to assert that they are the most general laws, which prescribe universally the way in which one ought to think if one is to think at all.15 14. As usual, one must be careful in utilizing the texts of Kant's lectures; for instance, here it is entirely unclear what is meant by "the necessary use of the understanding" (nothwendigen Verstandesgebrauch), where one expects Kant to speak instead of rules that necessarily hold for, or range over, the use of the understanding. Tillman Pinder argues for the earlier date of these lectures, against a later dating by Gerhard Lehmann, in “Zu Kants Logik-Vorlesung um 1780, anläßlich einer neu aufgefundenen Nachschrift,” Kant-Forschungen 1 (1987): 79–114. 15. Gottlob Frege, The Basic Laws of Arithmetic, ed. and trans. Montgomery Furth (Berkeley: University of California Press, 1967), 12.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Conception of General Logic   45 The fundamental conclusion one might draw about the various things Kant says about the principle of contradiction is that by reflecting on the possibility of thought, we can identify at least one specific claim that simply cannot be rejected, in that the very thought process that would lead to such a rejection would itself have to obey, at least implicitly, the principle itself. One is tempted here, perhaps, to invoke the notion of a “transcendental argument,” where (on one standard interpretation) a given premise can be challenged only by adopting the premise itself, thus showing the truth of the original premise. However, the very term “transcendental argument” is itself fraught with such controversy, and such arguments have been attributed to such a wide range of thinkers, from Plato and Descartes to Wittgenstein and Davidson, that I am hesitant to employ the notion here. Kant’s basic point, however, has been brought out well in a different way by Manley Thompson, in a paper discussing Hilary Putnam’s version of this principle, the “Minimal Principle of Contradiction,” which Putnam glosses as “Not every statement is true.”16 Thompson argues that this must be seen as an a priori truth, not in the sense of a claim within a logical system, but as a truth we must presuppose if we are to mean anything at all. “We thus accept it as an a priori truth, since we can offer no support of it; we can only acknowledge the necessity of always presupposing it.”17 As Thompson points out, systems of logic that have rejected similar such principles, such as Brouwer’s rejection of the law of excluded middle, have been taken seriously. In his Remarks on the Foundations of Mathematics, Wittgenstein raises an analogous situation, where double negations do not cancel out.18 It is not in this sense that Thompson presents the Minimal Principle of Contradiction as a principle of logic, but rather in a “neo-Tractarian or neoKantian way of speaking about logic.” In the sense, that is, not of a logic as the manipulation of variables, connectives, and quantifiers, but of 16. Hilary Putnam, “Analyticity and APriority,” in Realism and Reason: Philosophical Papers, vol. 3 (Cambridge: Cambridge University Press, 1983), 134. 17. Manley Thompson, “On A Priori Truth,” Journal of Philosophy 78 (1981): 463. 18. Ludwig Wittgenstein, Remarks on the Foundations of Mathematics, trans. G. E. M. Anscombe (Cambridge, Mass.: MIT Press, 1983), III.18; “Appendix I” (102–10). Of course, such views have been pursued and developed much further by intuitionist logicians; see Michael Dummett, Elements of Intuitionism (London: Duckworth, 1977).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

46   Kant’s Conception of General Logic logical principles that condition thinking and, again in a broad, nontechnical sense, meaning itself. As Thompson summarizes the point, a principle such as that of non-contradiction is without a “proper external subject matter” in that it doesn’t qualify as a rule of a metalanguage relative to a given object language; such a rule “ranges over all languages including their own.” In short, when we try to justify—or even explain—why the principle of non-contradiction is true, “the notion of explanation collapses because there is nothing properly external for them to be true of.”19 Cora Diamond has clearly captured this aspect of Wittgenstein’s early work, an aspect that I think clearly informs Kant’s own thinking about logic: “In the Tractatus, the idea of the illusory view from sideways on has a very particular form. When we philosophize we try as it were to occupy a position in which we are outside logic, where logic is that through which we say all the things we ordinarily say, all the things that can be said.”20 It is crucial in this context to remember, as Tonelli has pointed out, “that our time’s view of logic, as formal logic, or as symbolic logic, was completely foreign to philosophy until the beginning of the nineteenth century,”21 and we will later see other problems that arise with the sometimes too cavalier identification between logic and formal logic that Kant’s critics have assumed. At this point, we are simply trying to identify a single principle that can be revealed as absolutely necessary and strictly universal, or a priori, ranging over all possible thought qua judgment, and that would thus qualify as a principle of general logic. With the Kantian conception of general logic in mind, we can see that he would agree with Putnam’s insistence that “the truths of logic we are speaking of are so basic that the notion of explanation collapses when we try to ‘explain’ why they are true....... there is simply no room for an explanation of what is presupposed by every explanatory activity.”22 It 19. Thompson, “On A Priori Truth,” 474. 20. Cora Diamond, The Realistic Spirit (Cambridge, Mass.: MIT Press, 1995), 185; the phrase “from sideways on” is John McDowell’s. 21. Tonelli, Kant’s Critique, 2–3. Similarly, Richard Velkley (“Kant on the Primacy and the Limits of Logic,” Graduate Faculty Philosophy Journal 11 [1986]: 152): “recall that formal or general logic for Kant is not a ‘formal system’ as in later understandings of logic, but is primarily the study of thinking with concepts (conceiving).” 22. Putnam, “Analyticity and APriority,” 138.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant’s Conception of General Logic   47 is in this sense of a principle of thought that is presupposed as indispensable that Kant can call the principle of contradiction the “supreme principle of all analytic judgements” and the “universal and completely sufficient principle of all analytic cognition” (A151=B191).

Copyright © 2008. Catholic University of America Press. All rights reserved.

Criticisms of a Kantian Conception of General Logic This may well seem like a lot of effort to identify and, at least in some sense, justify a principle of logic that appears to be both innocuous and historically non-controversial, as well as being a principle Kant himself sees no need to spend time defending. Yet there are two good reasons for going into the detail I have above, relative to the principle of non-contradiction. First, if we can identify such a principle as qualifying under Kant’s criteria of absolute universality and strict necessity, then there may be some plausibility to the notion that we can identify a larger set of such principles that also so qualifies. Second, there are criticisms worth considering that reject any such notion of necessity of the kind Kant requires.23 I want now to turn briefly to those criticisms to provide a somewhat fuller picture of what Kant seems to think general logic can accomplish. W. V. Quine is perhaps the best-known critic not only of the analytic-synthetic distinction, but of the more general philosophical approach that embraces a strong sense of necessity within a context of emphasizing consciousness and the capacities of a knowing subject. For Quine, we must view the judging subject as an object for scientific scrutiny, and treat it as such.24 The subject is seen as a biological 23. I don’t take up here the criticism of the laws of logic registered by Pyrrhonism, for my concern at this point is not the issue of Kant’s response to skepticism, whether ancient or modern. Michael Forster offers some provocative comments about this very issue in his Hegel and Skepticism (Cambridge, Mass.: Harvard University Press, 1989), particularly 193–97n49. Nor do I here address concerns, often attributed to “postmodernism,” about the relationship between logic and power; I have argued elsewhere that, at the level of analysis Kant provides, those who reject all notions of logical rules, truth, and objectivity are committed to a relativism that is so radical as to be either meaningless or self-refuting. See my “Kant and Feminism” in Kant Studien 90, no. 3 (1999): 322–53. 24. To be fair to Quine, he has well-documented complaints that would call into question the very idea of assigning some kind of propositional content to a given judg-

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

48   Kant’s Conception of General Logic organism; we therefore use the appropriate tools to understand it. Psychology, chemistry, neuro-physiology to explore the present state of human knowledge functions; these disciplines in tandem with evolutionary theory and physical anthropology to understand the development of this same phenomenon. We place human knowledge in a scientific context and establish its properties by applying, revising, and developing theories relating to our observations. On this view, not only are the most abstract and highly developed phenomena of human knowledge to be understood as “output,” generated from the combination of sensory input and our conceptual framework; the same approach holds for notions as fundamental as the principle of cause and effect and the laws of logic. In the context of Quine’s naturalized epistemology, the latter merely represent an overwhelming survival value. An organism not complying with, for example, the principle of non-contradiction would be the first to be eliminated; the principle simply has an extraordinarily strong selective bias. Similarly, “though for the archaic and unconscious hypothesis of ordinary physical objects we can no more speak of a motive than of motives for being human or mammalian, yet in point of function and survival value it and the hypothesis of molecules are alike.”25 Thus we see established a biological continuum; at a relatively central position are possibly such quasi-analytic principles as the “laws of thought,” including the principle of non-contradiction; further out, theoretical statements pertaining to tables and chairs, the music of John Cage, shopping centers, and grand unified field theory; at the edge, observation sentences such as “red patch now.” For Quine, it is merely a difference of how far such different notions lie from the experimental or observational end of a system.26 As Kenton Machina has remarked, vis-à-vis the relatively central disciplines of logic and arithmetic, “the characteristic in virtue of which the statements ment. There are a number of technical points here, particularly concerning Quine’s views on reference, and his notion of theory as the unit of meaning, that I think I can safely avoid in this context. 25. W. V. Quine, Word and Object (Cambridge, Mass.: MIT Press, 1960), 20; 22. 26. W. V. Quine, “Necessary Truth,” in The Ways of Paradox and Other Essays (New York: Random House, 1966), 56.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Conception of General Logic   49 in these disciplines obtain centrality is simply this: we are not likely to give up these statements, even in the face of experience which does not accord with our theory.”27 Relative to the specific issue under consideration, Quine notes, “I see no higher or more austere necessity than natural necessity; and in natural necessity, or our attributions of it, I see only Hume’s regularities.”28 In short, Quine insists, “No statement is immune from revision. Revision even of the logical law of the excluded middle has been proposed as a means of simplifying quantum mechanics; and what difference is there in principle between such a shift and the shift whereby Kepler superseded Ptolemy, or Einstein Newton, or Darwin Aristotle?”29 One might be tempted here to respond to Quine by employing what Plato self-deprecatingly derides in Theaetetus (164c) as “logic chopping”: for if no statement is immune to revision, then that clearly leaves room for this statement itself to be revised. One obvious revision one might suggest is its negation, that it is not the case that no statement is immune to revision, or that at least one statement is immune to revision. Certainly the logic of Quine’s position leaves open this possibility, although Quine would argue that its acceptance would be made only within the entire “tribunal of experience,” which would still not establish the necessity of such a principle in a way that would satisfy Kantian requirements. Yet one can approach the same kind of a strategy from a slightly different direction, namely that suggested earlier by Thompson and Putnam. On Putnam’s revision of the principle of non-contradiction, it reads, “Not every statement could be true.” Presumably Quine wishes to deny the requisite necessity to this version of the statement, and simply regard it as one of the last principles we would be willing to give up. Yet, as Thompson develops the point, one who “thinks a contradiction” can do so only by implicitly conforming to the principle of noncontradiction: “Without this conformity, we would have to think not only both p and not-p, but also its negation, neither p nor not-p. With 27. Kenton Machina, “Kant, Quine, and Human Experience,” Philosophical Review 81 (1979): 484. 28. Quine, “Necessary Truth,” 56. 29. Quine, “Two Dogmas of Empiricism,” 43.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

50   Kant’s Conception of General Logic absolute nonconformity, with strictly illogical thought, we get endless iteration of this process. We do not think merely that both, both p and not-p and neither p nor not-p, but also that neither, neither both p and not-p nor neither p nor not-p. We thus think nothing at all.”30 The very possibility, that is, of our being able to think—and, importantly, being able to attempt to understand one another—presupposes that we, and they, conform to the principle of non-contradiction, and that the exposure of contradictory beliefs imposes the normative demand to eliminate the source of the contradiction. Failing to judge in implicit accord with these requirements would allow all thoughts to be possible; with everything up for grabs, there would then be no way of eliminating inconsistencies. Certainly rejecting the classical assumption that any statement is implied by a contradiction (ex falso sequitur quodlibet) would be seen within Kant’s logical tradition as having devastating and pervasive effects on constructing even the most minimal account of rationality. Without that presupposition, there could be no expectation or attribution of rationality. Thus we see that at least one principle must be regarded as absolutely universal and strictly necessary, ranging over the possibility of thought, that is, yielding an a priori constraint on rationality.31 After Frege’s development of predicate logic in the Begriffsschrift, a great deal of work came to focus on the development of axiomatic techniques in logic. Motivated largely by the challenge posed by nonEuclidean geometries and questions in the foundations of mathematics, logicians and mathematicians (among them Pasch, Peano, and Hilbert) sought to develop rigorous axiomatized systems where undefined terms and undemonstrated propositions are made explicit; the propositions of a given system can then be constructed in accordance with a given set of definitions and postulates, and theorems are then proved within that system. A consequence of this history has been the development of a number of “non-classical” logics, which operate 30. Thompson, “On A Priori Truth,” 471n8. 31. As we will see in discussing the Second Analogy, this will again be at issue in Kant’s critique of Hume; if we are to rely solely on the natural necessity that arises from what Quine calls “Hume’s regularities,” we discover that assembling those regularities presupposes a sense of necessity sufficiently strong to identify the regularities themselves.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Conception of General Logic   51 analogously to axiomatized non-Euclidean geometries by introducing alternate postulates or operators; new theorems are then derived within the parameters laid down by the particular axiomatic structure. A wide variety of “deviant” or non-classical logics have been outlined, including quantum logic, erotetic logic, deontic logic, intuitionist logic, temporal logic, relevance logic, and fuzzy logic; some would add modal logic to this list.32 There are also lengthy debates about whether any particular such logic constitutes an extension of classical logic or is a genuine competitor that should replace classical logic. Of particular interest in this context is paraconsistent logic, developed by logicians who counsel that we get “past the consistency hang-up.”33 Drawing on semantic paradoxes (e.g., “Everything I say is false”), debates in the history of science (Leibniz’s account of infinitesimals, Bohr’s theory of the atom), and more “attenuated” theories within which inconsistencies may arise, paraconsistent logicians argue that the wide range of inconsistent yet non-trivial theories requires eliminating the principle of non-contradiction, or at least weakening it to the extent that contradictions are not only permissible within a given theoretical account but even required. For reasons that we will see shortly, I will not here go into the details of how a given paraconsistent logic can be axiomatized and formally presented, but will rather try to show that whatever its merits, the formal systems proposed by paraconsistent logicians are not so much in conflict with Kant’s conception of general logic as addressing fundamentally distinct questions. I want to look first at a specific example Routley and Priest offer, a contradiction arising from the application of a legal statute.

32. For some discussion of these, see Susan Haack, Deviant Logics (Cambridge: Cambridge University Press, 1974). 33. G. Priest, R. Routley, and J. Norman, eds., Paraconsistent Logic: Essays on the Inconsistent (Munich: Philosophia Verlag, 1989), 154. Most of the discussion here is based on an essay by Priest and Routley in this volume, “Systems of Paraconsistent Logic,” 151–86; it should be noted that these authors favor a “relevant paraconsistent” system, in contrast to Newton C. A. da Costa’s system of paraconsistent logics, C1, in large part due to problems they see arising from da Costa’s treatment of the law of non-contradiction and ex falso quodlibet.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

52   Kant’s Conception of General Logic

Copyright © 2008. Catholic University of America Press. All rights reserved.

The constitution of a certain country contains the clauses (a) “No person of the female sex shall have the right to vote,” (b) “All property holders shall have the right to vote.” We may also suppose that it is part of the common law that women may not legally be property holders. As enlightenment creeps over the country this part of common law is changed to allow women to hold property. Inevitably, eventually, a woman, call her Jan, turns up at a polling booth claiming the right to vote. A test case ensues. Patently the law is inconsistent. According to the law Jan both does and does not have the right to vote.

Priest and Routley conclude “it seems that ‘Jan has the legal right to vote’ and ‘Jan does not have the legal right to vote’ are both true.”34 From the perspective of classical logic—and, of course, Kant—there is no particular demand here to embrace the conjunction of sentences (p, ~p) as true; rather, what is in conflict is the latter statement with the changes that common law in this example has undergone. To be sure, without an updated codification of those changes in statutory law, the two sentences are clearly in conflict; it is considerably less clear that “inferences are made, and made commonly, from the information” contained in them.35 Rather, from Kant’s perspective, this is precisely the kind of situation that exposes a contradiction to be eliminated, and the conflict between common law and statutory law thus imposes an obligation upon legislators to remove it. On Kant’s view, there need be little discussion over the alternatives of calling into question the principle of non-contradiction or making the relatively slight adjustment (itself presupposing the principle) to eliminate the inconsistency. To be sure, paraconsistent logicians might well reject such an approach as ignoring the more systematic difficulties raised by both the semantic paradoxes and the profound problems in applying classical logic to issues that arise within quantum mechanics. But what must be emphasized here is that Kant does not conceive of general logic along the lines of a system to be formalized along the lines of an axiomatized model; in particular, the principle of non-contradiction is rather to be seen as the most general rule that ranges over all expressions of 34. Ibid., 153–54. 35. Ibid., 152.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Conception of General Logic   53 meaningful thought, qua judgment. I cited Tonelli earlier to the effect that while symbolic approaches to logic had always been available, it was not until the nineteenth century that logic and its formal expression were closely identified. As he further remarks, “logic was never disassociated from the methodological consideration of the substance of thought.”36 This point is worth emphasizing. While Kant’s conception of logic is undeniably restricted by the dominance of Aristotelian logic, with the admixture of a number of other developments— from the Stoics, from Medieval and Renaissance philosophy, from the Port-Royal Logic, from Leibniz and Wolff, and from such contemporaries as J. H. Lambert—it is clear that that conception does not, or at least need not, reduce to the limited formal capacities of Aristotelian syllogistic. Thompson has articulated the difference by contrasting logic viewed as having “a definite subject matter” with a “neoKantian or neo-Tractarian sense” of logic.37 The former is, of course, closely associated with the post-Fregean axiomatic developments in formal and mathematical logic briefly discussed above; the latter is the sense of general logic we find in the First Critique, where its universal and necessary principles serve to range over all possible expressions of thought. Thus, while a paraconsistent axiomatization of a fragment of logic may urge abandoning, or at least bracketing, the principle of non-contradiction, Kant’s general logic approaches the issue in an entirely different way. Rather than trying to construct a consistent model that allows us, given certain fundamental parameters, to derive specific theorems, Kant seeks a principle that holds a priori for any and all expressions of thought: a minimal condition for rationality simpliciter. So one might imagine a logician presenting to an audience a particular model, one that embraces the paraconsistent rejection—in this case— of the principle of non-contradiction. After an elaborate and elegant argument, she concludes, “We must, therefore, reject the principle of non-contradiction.” From the Kantian perspective, it is legitimate to ask how the audience is to understand this conclusion. That it is true that we must reject it, and that it is true that we must not reject it? Or, 36. Tonelli, Kant’s Critique, 3. 37. Thompson, “On A Priori Truth,” 472.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

54   Kant’s Conception of General Logic worse, that it is true that we must both reject and accept it, as well as that we must not reject it and must not accept it? We must, obviously, interpret her claim relative to the model she has outlined—as Thompson puts it, relative to an “external subject matter”—rather than as a universal and necessary, or a priori, truth about the ability to judge. In sum, although I discover the unique logical features, qua presupposition, of a principle like the principle of non-contradiction through a radically subjective process of philosophical reflection, I discover it to be a universal and necessary condition for meaningful thought to be possible, and as such to be an objective condition. As Wittgenstein puts it in the Tractatus, the “truth is that we could not say what an ‘illogical’ world would look like.”38 It is along these lines that I think Kant—in spite of his myriad complaints against him—would be happy to accept Locke’s remark that “God has not been so sparing to Men to make them barely two-legged Creatures, and left it to Aristotle to make them Rational.”39 General logic, as presented by Kant, is the articulation of those concepts and rules that are indispensable for thought itself; insofar as they serve as the necessary conditions for the possibility of thought, they range over any particular presentation that seeks to provide a particular model, as one might find in a contemporary account of formal or mathematical logic. In this sense, then, Kant’s conception of general logic, properly understood, resonates with Frege’s view that “[t]hought is in essentials the same everywhere: it is not true that there are different kinds of laws of thought to suit the different kinds of objects thought about.”40 General logic seeks to identify these essentials. On Kant’s view, there is a set of logical concepts and principles (or rules) that are indispensable to characterizing rational thought expressed in judgment, and that then serve as necessary conditions for the 38. Ludwig Wittgenstein, Tractatus Logico-Philosophicus, trans. D. Pears and B. F. McGuiness (London: Routledge and Kegan Paul, 1961), 3.031. 39. John Locke, An Essay Concerning Human Understanding, ed. with an introduction by P. Nidditch (Oxford: Clarendon Press, 1975), 671 (Book IV, chap. 17, §4). Cf. Kant’s remark in the Blomberg Logik: “Locke’s book de intellectu humano is the ground of all true logica,” Ak. XXIV.1, 37. 40. G. Frege, The Foundations of Arithmetic, trans. J. Austin (Oxford: Blackwell, 1978), iii.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Conception of General Logic   55 possibility of thought. Such conditions range over all possible thought, and without them we abandon all hopes of making clear the very notion of rationality itself; the principle of non-contradiction serves as the clearest example. At the same time, this does not entail that general logic provides a set of sufficient conditions for rationality, and nowhere does Kant indicate that such provision is within philosophy’s purview, let alone that of general logic. The task of general logic, then, is to identify and articulate the universal and necessary conditions for judgment and reveal how they can be “justified.” Such a justification, however, as Kant makes clear, can be given only reflectively; as we will see below, the reflection involved is the radically subjective philosophical consideration of precisely those conditions that one comes to recognize as necessary for the possibility of thought. In short, by thinking about the conditions for the possibility of thought, we discover those conditions that must hold for thought itself to be possible, precisely the means by which Kant distinguishes a “principle” from a “theorem”; only the former “has the peculiar character that it makes possible the very experience which is its own ground of proof, and that in this experience it must always be presupposed” (A737=B765). Through this reflective procedure, we recognize that the principle of non-contradiction is necessary for the possibility of thought and that thought itself reveals the necessity that makes thought itself possible. This is in essence the nervus probandi of Kant’s method, and we see Wittgenstein characterize the point succinctly by noting that “logic is transcendental.”41 41. Wittgenstein, Tractatus Logico-Philosophicus, 6.13.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Ch a pter Thr ee

The Historical Background of Kant’s General Logic e

Copyright © 2008. Catholic University of America Press. All rights reserved.

General Logic and Grammar I have tried to show up to this point that Kant conceives of general logic as a set of universal and necessary rules for the possibility of thought, or as a set of minimal necessary conditions for ascribing rationality to an agent (focusing, up to this point, on the principle of noncontradiction). Such a conception contrasts with contemporary notions of formal, mathematical, or symbolic logic: rather as an attempt to identify those conditions that must hold for the possibility of thought, such conditions must hold a fortiori for any specific model of thought, including axiomatic treatments of logic and standard Gentzen-Fitch natural deduction models of first-order predicate logic. Kant’s general logic seeks to isolate those conditions by thinking through, or reflecting on, those conditions that themselves make thought possible. To clarify Kant’s reflective strategy, relative to general logic, it is of some help to examine the historical background within which his conception of logic is embedded. To begin to do so, I want to look at his comments about a discipline that has a longstanding historical con56

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



Background of Kant's General Logic   57

nection with logic, namely grammar. Traditionally, grammar, rhetoric, and logic had composed the “trivium” of university study, and along with the “quadrivium” (arithmetic, music, geometry, and astronomy) constituted the seven liberal arts. The relationship between logic and grammar has, of course, been long recognized; Aristotle’s Organon begins with the Categories and On Interpretation, both of which treat grammatical at least as much as logical issues.1 With Frege’s work and the renewal of interest in the philosophy of logic, investigations into the connections between grammar and logic have flourished. As Bertrand Russell makes the point: “The study of grammar, in my opinion, is capable of throwing far more light on philosophical questions than is commonly supposed by philosophers....... On the whole, grammar seems to me to bring us much nearer to a correct logic than the current opinions of philosophers.”2 To be sure, there would be the sharpest disagreements between Russell and Kant, not the least of which would involve the notion of a “correct logic.” But it is clear that Kant is in agreement with Russell’s observation that the study of grammar can reveal much about logic, and, not surprisingly, Kant begins almost all of his many lectures on logic with some comments about grammar. I want to look at some of these comments, as well as some of Kant’s Reflexionen, not just to shed some light on the strategy Kant pursues in the Critique of Pure Reason, but also to consider why Kant spent so little time talking about natural language outside its grammatical context. 1. One might even be tempted to say that here the distinction between the two breaks down. Not surprisingly, one can find suggestions of the importance of grammar in Plato; in addition to the obvious Cratylos, see Theaetetus 183a and the “Seventh Letter.” T. Benfey argues for the idea that Plato is suggesting the idea of a universal language scheme in “Über die Aufgabe des platonischen Dialogs Kratylos,” Abhandlungen Göttingen, Philologische-Historische Klasse 12 (Göttingen, 1866): 189–330 (I owe this reference to W. K. C. Guthrie, A History of Greek Philosophy, vol. 5 [Cambridge: Cambridge University Press, 1986], 31). Sextus Empiricus gives a long discussion of grammar in Against the Professors [Προσ Μαϕεµακους], chapter 3. Such examples can be easily multiplied from the history of Classical, Medieval, and Renaissance philosophy. See also Michael Frede, "The Origins of Traditional Grammar," in Essays on Ancient Philosophy (Minneapolis: University of Minnesota Press, 1987), 338–59, which emphasizes the Stoic contribution to this history. 2. Bertrand Russell, The Principles of Mathematics (London: Allen and Unwin, 1903), 42.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

58   Background of Kant's General Logic Kant does not mention grammar in the First Critique—indeed, he says virtually nothing at all about language there—but he does draw a connection in the Prolegomena between grammar and the method employed for discovering the categories. To find the categorial concepts “presupposes no greater reflection nor more insight than to pick out in a language the rules for the actual overall use of words so to collect the elements for a grammar (in fact both investigations are very closely related to one another)” (Ak. IV, 323). Some seven to ten years later, in the unfinished “Prize Essay on the Progress of Metaphysics,” Kant returns to the comparison, now incorporating the view that a critical ontology is to be “given the modest title of a mere Analytic of pure understanding” (A247=B303): “For just as a grammar is the analysis of a form of language [Sprachform] into its elementary rules or logic is the analysis of the form of thought, ontology is the analysis of cognition into concepts that lie a priori in the understanding and which have their employment in experience” (Ak. XX, 260). For the most part, however, the references Kant makes to “grammar” can be found in unpublished lectures and notes, where he tends to repeat the same basic points:

Copyright © 2008. Catholic University of America Press. All rights reserved.

1. Grammar and logic, at their most abstract, develop rules relative to form: “Grammar is the form of a language in general” (Ak. XXIV.2, 502; Ak. XXIX, 1045). “Logic is a rational science, a canon for the understanding. Just as grammar is for evaluating language as to form” (Ak. XXIV.2, 792). There are “universal rules of language. Such a grammar does not contain words, not copia vocabularum, but only the form of language” (Ak. XXIV.2, 693).

2. Grammar and logic are employed unconsciously: “One speaks but without having learned grammar, that is, one speaks in accordance with rules of which one is not conscious” (Ak. XXIV.2, 502); “We speak according to rules without being originally conscious of them” (Ak. XXIX, 1045); “one speaks German and afterwards brings in the rules of German grammar” (Ak. XXIV.2, 609); “One speaks without knowing [kennen] grammar, however; and he who speaks without knowing it does actually

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of



Background of Kant's General Logic   59 have a grammar and speaks according to rules of which he is not himself conscious” (Ak. IX, 11).

3. Grammar and logic, while originally employed unconsciously, can be systematically developed through reflection: “Logic is a rational science, a canon for the understanding. Just as grammar is for passing judgement on language as to form” (Ak. XXIV.2, 792); “All languages, in accordance with their first principles, can be reduced to a grammar. Moreover, grammar is a doctrine of the understanding, of course. For as our soul combines concepts, so must words also be combined” (Ak. XXIV.2, 790);3 “Were we to compare transcendental concepts, this would be a transcendental grammar, which would contain the ground of human language....... Were one to reflect on this, one would have a transcendental grammar. Logic would contain the formal employment of the understanding. Then transcendental philosophy, the doctrine of all universal a priori concepts, would be able to follow” (Ak. XXVIII.2.1, 576–77).

Copyright © 2008. Catholic University of America Press. All rights reserved.

4. Logic and grammar range universally over their respective domains: “[W]e can form for ourselves an idea of the possibility of such a science [i.e., logic], just as we can of a universal grammar, which contains nothing more than the mere form of language in general, without words, which belong to the matter of language” (Ak. IX, 12–13); “In a grammar we consider the universal rules, without which no language at all can exist”; “A universal doctrine of thought is thus possible, and from that follows also a universal doctrine of language, Grammatica universalis” (R 1620; Ak. XVI, 40).

5. Logic and grammar diverge in terms of their potential for becoming a science: “Grammar is only a discipline, logic a science; doctrine” (Ak. XXIV.2, 609); “Logic must contain principles a priori. Thus logic is a science, and grammar 3. One might recall that in at A78=B103, Kant originally wrote that “Synthesis ..... is a blind but indispensable function of the soul,” but changed “soul” in his own copy (Nachträge XLI) to “understanding.” This is probably how this claim from the Wiener Logik should be read, as well.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

60   Background of Kant's General Logic

Copyright © 2008. Catholic University of America Press. All rights reserved.

not, because its rules are accidental [zufällig]” (Ak. XXIV.2, 694); “neither grammar nor aesthetics is a science” (Ak. XXIV.1, 25).

We will see below some of Kant’s reasons for drawing this last distinction. In sum, both grammar and logic constitute rigorous reflective enterprises, identifying rules that range universally and necessarily over their relative domain. Just as one can think in accordance with logical rules without being acquainted with, or having studied, logic, one can speak grammatically without being acquainted with, or having studied, grammar. As we have seen, one who rejects the minimal constraints on rationality, such as the principle of non-contradiction, ultimately fails to make meaningful utterances; one who attempts to abandon all constraints imposed by grammar would similarly fail. Kant is committed to the idea that all thought is linguistically mediated: as he asks in the Wiener Logik, “how will they [the logici] be able to think judgements without words?” (Ak. XXIV.2, 964); similarly in the Lectures on Anthropology: “Language is the indicator of thought and, on the other hand, the most excellent mode of indicating thought is that of language, the greatest means to understand ourselves and others. Thinking is talking with ourselves” (Ak. VII, 192).4 So it is not surprising to see Kant draw a strong analogy between grammar and logic, going so far as to claim that “the grammarians were the first logicians” (R 1622; Ak. XVI, 42). More important for our purposes here, however, is that this analogy allows us to begin to see the connection between general and transcendental logic, a connection Kant fully exploits in the strategy he follows in the Critique of Pure Reason. A very simplified picture of grammar, one that doesn’t concern itself with morphology or phonology, presents a syntax by describing a set of conditions that any proposition must meet to be well-formed, as well as the formal rules that govern the relations between and among 4. As Beatrice Longuenesse makes the point, for Kant “no judgment [as psychological activity] can take place without linguistic expression.” Kant and the Capacity to Judge, trans. Charles T. Wolfe (Princeton: Princeton University Press, 1998), 100n47. See also Brandt, Die Urteilstafel: “For Kant there is no thought that would not be grasped [gefaßt] linguistically,” 77n54.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



Background of Kant's General Logic   61

propositions.5 A semantic element is introduced when meaning—the relation between a proposition and what it is a proposition about—is brought into consideration. If Kant is right and the respective functions of a grammarian and a logician are closely analogous, then this syntactic/semantic division should also be applicable to logic. Furthermore, if, as I will argue below, there is a strong connection between general and transcendental logic, this distinction should also help clarify this connection as well. General logic, as a systematic set of formal rules for thought in general, could be said to provide a pure syntax for the formation of judgments. Just as the grammarian gives formal syntactic rules for the legitimate construction of linguistic expressions, the (general) logician puts forth a minimal set of rules for thought in general. A proposition must satisfy the grammatical condition of being well-formed to be able to make sense; a judgment, as we have seen, must meet the (negative) logical condition of not being self-contradictory to be thinkable. Additionally, just as a well-formed proposition may be nonsensical—for example, Chomsky’s famous “colorless green ideas sleep furiously”—a judgment that is not self-contradictory can still fail to make objective reference, or fulfill the conditions of objective validity (A222–23=B270). But such questions do not fall within the domain of general logic; as we have seen, it abstracts from all content, and does not consider the relation between a judgment and its object. Continuing the analogy, general logic abstracts from any semantic component, and as such can provide only a negative criterion of truth. Given that the central claim I hope to defend here is that Kant’s analogy between general and transcendental logic provides the key to 5. This naive characterization is put forth to emphasize the similarities in the procedures shared by linguists and logicians, rather than a precise taxonomy of the subdisciplines of linguistics, and in particular is designed to focus on the universal features that (Kant sees) shared by grammar and logic. To be sure, a linguist treats morphological and syntactic issues as involving separate but related questions, while the logician would treat both as syntactic, in contrast to semantics. (Not infrequently, the close tie between morphology and syntax in linguistics is expressed by the hybrid “morphosyntax.”) I also do not take up the important differences the linguist might describe in terms of “competence” and “performance,” nor those complications introduced by Chomsky’s conception of “degrees of grammaticality.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

62   Background of Kant's General Logic understanding correctly both the scope and limits of reason, it is perhaps worthwhile at this point to introduce this analogy in the context of Kant’s view of the relationship between grammar and logic. As a logic, transcendental logic has a purely syntactic aspect, constituted by the pure concepts of the understanding. But in contrast to general logic, it also has an essentially semantic function in virtue of its application (and restriction) to possible experience. The truth-conditions of a judgment of possible experience relate fundamentally to the object about which the judgment is made. In this way, a pure set of syntactic conditions for making well-formed judgments about objects of possible experience would yield a transcendental logic. As Kant characterizes it in the Critique, it would be a logic “which should contain solely the rules of the pure thought of an object,” and “would exclude only those cognitions which would have empirical content” (A55=B80). Such a logic, limited to its appropriate domain, would yield a transcendental analytic that can serve as a “logic of truth.” Transcendental logic (specifically transcendental analytic) would then provide a syntax for possible experience, as in a similar way grammar provides a syntax for linguistic expression. In either case, a syntax is empty without something that it is a syntax for; in the case of grammar, without words (“the material for language”), in the case of transcendental analytic, without the manifold of sensible intuition. For in “the absence of intuition all our cognition is without objects [Objekten], and therefore remains entirely empty” (A62=B86). But given this sensible manifold, the necessary and universal rules of transcendental analytic put us in a position to be able “to spell out [buchstabieren] appearances, so that we may be able to read them as appearances” (Ak. IV, 312). Given, say, a declarative sentence, the grammarian proceeds to give a syntactic analysis by labeling its various elements with the relevant syntactic categories, for example, Noun, Verb, Determiner, and bring them under the more general labels Noun Phrase, Verb Phrase, Prepositional Phrase, which are then taken together to reconstitute the Sentence. Such an analysis would be equally applicable to the sense “the cat sat on the refrigerator” and “the refrigerator sat on the cat.” The peculiarity of the latter is not grammatical (syntactic), but is due to a semantic aspect, what is sometimes called “referential obstruction.” It

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



Background of Kant's General Logic   63

is only when the relation between the sentence and what it is about is brought into consideration that these two well-formed sentences would be distinguished. The analysis provided by the general logician is similar in method and scope. Given a judgment “No S are P,” a syntactic analysis is provided by characterizing it as a universal negative proposition with a subject term “S” and a predicate term “P”; as the contradictory of the particular affirmative judgment “Some S are P,” the contrary of the universal affirmative “All S are P,” and the super-altern of the particular negative “Some S are not P”; as distributing both its terms; as implying its (strict) converse “No P are S” and obverse “All S are non-P.” This string of formal properties holds equally well for the judgments “No Philosophers are Prevaricators” and “No Philosophers are Millionaires.” The veracity and financial status of those involved introduces a semantic question; while equally well-formed judgments, they are distinguished by bringing into consideration the relation between the judgment and what the judgment is about. It is this connection between syntax and semantics (the provenance of transcendental logic) that Kant exploits in his strategic move from the table of judgments to the table of categories. Thus, while the judgment “All bodies are divisible” can be converted (by limitation, or per accidens) to “Something divisible is a body,” with respect to general logic “it remains undetermined to which of the two concepts the function of the subject, and to which the function of the predicate, is to be assigned.” This assignment can be made only within the domain of transcendental logic: “[W]hen the concept of body is brought under the category of substance, it is thereby determined that its empirical intuition in experience must always be considered as subject and never as mere predicate. Similarly with all other categories” (B129). The central claim in the Metaphysical Deduction is one of Kant’s most difficult, as well as one of the most frequently discussed in the secondary literature: “The same understanding, through the same operations by which in concepts, by means of analytic unity, it produced the logical form of a judgement, also introduces a transcendental content into its representations, by means of the synthetic unity of the man-

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

64   Background of Kant's General Logic ifold in intuition in general” (A79=B105). The obvious question that arises is how does the logical form of a judgment (what I have been calling the syntactic element) introduce a “transcendental content” (which, whatever else it may be, is in some sense a semantic element)? The clearest response, which I largely follow here, has been provided by Arthur Melnick. The point is that when we move from general logic to transcendental logic, we introduce this semantic aspect by specifying the domain—possible experience—over which its rules range. Melnick takes quantification to represent the syntactic characteristic of a judgment described under the heading “Of Quantity” in Kant’s table of judgments. Various things can be said about the formal moves in quantificational structure—for example, quantifier elimination— but when we take a quantified judgment in relation to what is judged about, we introduce a “transcendental content”: “Now the notion of an individual (of one thing) is not a notion that applies to judgements; rather it is a notion that applies to what is judged about. Thus, the syntactical concepts under the heading ‘Of Quantity’ are intimately connected to the notion of an individual, which is not a syntactical notion at all.”6 Without trying to strain the analogy, a linguistic analysis might be looked at in similar fashion. A sentence possesses certain structural features, and a morpheme is amenable to transformation (e.g., adding “-ly” to an adjective to produce an adverb) in virtue of syntax, in the simplified way I have been using this term. A Noun Phrase would then take on a semantic dimension—would introduce a “transcendental content”—when this phrase is considered in connection with what the noun refers to or represents. Although he doesn’t develop it, this may be what Kant is driving at in the isolated comment given above, from the Lectures on Metaphysics (L2), that an analysis of transcendental concepts would result in a “transcendental grammar.” It is also in this context that one may see why Kant spends so little time discussing natural language, for his interest is in identifying those conditions that range over any natural language. As Lars Fr. Svendsen has made the point, “Kant generally seems to have very little belief ” in constructing a Leibnizian ars charac6. Melnick, Kant’s Analogies of Experience, 40.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



Background of Kant's General Logic   65

teristica; rather, he is “merely interested in universal grammar, not also in a universal language.”7 This may also provide a way of understanding Kant’s suggestion (at Metaphysik L2) that transcendental philosophy would follow upon the development of a transcendental grammar. Such a grammar, limited to the purely formal or syntactic analysis of linguistic expression, would simply be the formal analysis of thought; as such, by yielding the universal and necessary conditions for the possibility of linguistically mediated judgments, it would thus be identical to, or ultimately reduce to, those universal and necessary conditions that constitute general, or universal, logic. It should start to become clearer from the above account that Kant’s conception of general logic is in a fundamental sense distinct from the standard view reducing that conception to the syllogistic of Aristotle, Theophrastus, Galen, and others, which in turn criticizes Kant on precisely this score. This is not to say that, when Kant characterizes general logic in terms of its formal apparatus, he did not have syllogistic in mind, but simply that what he means by “logic” is considerably more complex than such a reduction allows, and provides some conceptual space for an alternative picture.8 I would like to fill out some of this picture by suggesting that Kant comes very close, in his characterization of both logic and grammar, to embracing a view strikingly similar to that put forth, in various ways, by those arguing from a different perspective for a “universal language” or a “universal grammar,” although this connection has remained, as far as I can tell, almost entirely unexamined in the literature on Kant.9 That quest for a universal grammar, of course, has a long, rich, and complex history, the many nuances of which I cannot attempt to do justice to here; at the same time, Kant is fundamentally at odds with its basic metaphysi7. Lars Fr. Händler Svendsen, “Kant’s Critical Hermeneutics: On Schematization and Interpretation” (diss., Faculty of Arts, Oslo, 1999), 244n170. 8. As we will see, this interpretation of Kant’s conception of logic has systematic repercussions for interpreting the metaphysical deduction, as well as in evaluating the more general project of the Transcendental Analytic. 9. The only exception to this I’m familiar with is Lars Fr. H. Svendsen, in his dissertation (see note 7) and “Språk og logikk i Kants kritiske filosofi.” Jere Paul Surber briefly considers the possibility in his Language and German Idealism: Fichte’s Linguistic Philosophy (Atlantic Highlands, N.J.: Humanities Press, 1996), 97, but quickly rejects it.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

66   Background of Kant's General Logic cal assumption. Yet both these differences and the surprising similarities between the views of Kant and at least some of the universal grammarians helps put Kant’s project into a broader historical perspective, as well as clarifying further his strategy within the Critique of Pure Reason. Kant adds a motto, in the second edition of the Critique, in the form of a relatively lengthy citation from Francis Bacon’s Instauratio Magna. I want to suggest that a claim by another Bacon encapsulates much of Kant’s thinking on logic, and its relation to language, namely Roger Bacon’s famous claim that “in substance grammar is one and the same in all languages, although it allows variation in terms of accident[s].”10 The search for such a grammar was already old by the thirteenth century, when Bacon wrote this; we have seen at least hints of it as early as Plato’s Cratylus, and one continues to find various indications of the view in, for instance, the standard Latin grammars of Priscian and Donatus. The history of the view is, indeed, a remarkable example of how a conceptual approach can evolve: beginning as a novel idea, in turn being embraced widely and enthusiastically, followed by this interest waning and becoming dormant, only to be revived within a different intellectual era and context. Thus we find the search for a universal grammar, often expressed as the search for, or construction of, an ideal language, never quite dying away entirely, and wielding varying degrees of influence; we can see such a revival, for instance, in the seventeenth century.11 Obviously enough, the most prominent thinker of relevance for Kant in this regard was Leibniz—though probably through the works of Christian Wolff—and his attempt to provide a characteristica universalis. At the same time, it is clear that Kant could not have found the Leibnizian approach at all plausible, given the radically different conception each had of analyticity and the role of the intellect in discerning truth.12

Copyright © 2008. Catholic University of America Press. All rights reserved.

.

10. Roger Bacon, Oxford Greek Grammar, ed. E. Nolan and S. A. Hirsch (Cambridge: Cambridge University Press, 1902), 27 [part II.1.2], dated by the editors as written around 1252; the original reads “grammatica una et eadem est secundum substanciam in omnibus linguis, licet accidentaliter varietur.” 11. See, e.g., M. M. Slaughter, Universal Languages and Scientific Taxonomy in the Seventeenth Century (Cambridge: Cambridge University Press, 1982). 12. I argue above, in some detail, that Kant rejects the Leibnizian model (as well

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



Background of Kant's General Logic   67

Instead of attempting to rehearse this history, it will better suit the purpose here to focus on one prominent school of universal grammarians, the so-called “Modistae” or “modists” who flourished from approximately 1250 to 1350. I want to argue, in a necessarily abbreviated way, that the general conclusions of the Modistae, with two crucial exceptions, are surprising similar to those reached by Kant. As we will see, those two exceptions—the denial of the primacy of Latin (or any natural language), and the perspectival shift now often referred to as Kant’s “Copernican Revolution”—are, clearly, of overwhelming importance. However, both the similarities and the disagreements between Kant and the Modistae will, again, reveal something about Kant’s conception of logic, language, and reason. The Modistae constituted, roughly, a school of grammarians concerned with providing a foundational account of reality by constructing a universal grammar that would capture the invariant features of reality. The various writers who embraced this approach—Boethius of Dacia, Martin of Dacia, Radulphus Brito, Thomas of Erfurt, and Siger of Courtrai—focused on the structural aspects of language to determine those invariable features of the world that language described. As Jan Pinborg characterizes their view, which he takes to be fundamentally Aristotelian in origin, “The claim—read out of Aristotle’s Posterior Analytics—that any discipline worthy to be called a science would have to treat features of reality which are universal and immutable had especially far-ranging repercussions.”13 In general, the Modistae emphasized the grammatical analysis of the parts of speech, the account of which was generally adopted from the ancient grammatical tradition, and the role they played in determining meaning, in that, on their view, language mirrored the world, and thus the articulation of those features of language that are invariant provide a fundamental insight into that world. As Bursill-Hall describes the view: as that of Locke) on the grounds that it leads to both antinomies and skepticism, thus providing indirect support for Kant’s own conception of the intellect. 13. Jan Pinborg, “Speculative Grammar,” in The Cambridge History of Later Medieval Philosophy, ed. N. Kretzmann, A. Kenny, and J. Pinborg (Cambridge: Cambridge University Press, 1982), 255. As Pinborg notes (ibid.), “The most important factor for the development of modistic theory is the recovery of the whole Aristotelian corpus, especially the Posterior Analytics, the Metaphysics, and the De anima.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

68   Background of Kant's General Logic

Copyright © 2008. Catholic University of America Press. All rights reserved.

The philosophical background of the Modistae led them to believe that grammar had its basis outside language itself; they claimed therefore that there was one universal grammar dependent on the structure of reality and that the rules of grammar were quite independent of the language in which they were expressed. There was one grammatical system fixed and valid for all languages but which the philosopher alone is able to discover.14

Hence we find the following picture common to the Modistae: reflecting on the use of language, we can discover those structural aspects— invariant, universal, immutable, and necessary—that range over all meaningful uses of language. Because language mirrors reality, those aspects thus articulated provide us with the fundamental characteristics of reality. And as ranging over all languages, they are (or should be) in no way dependent on any particular natural language; how we conceive the world, and express its conception linguistically, will provide the structural outline of the characteristics of that world. A universal grammar will thus offer both a general picture and many of the details of that structural outline, and thus in turn serve as a foundation for the determination of signification, or a general semantics. In actually trying to provide such a universal grammar, the Modistae of course went to elaborate lengths to substantiate their view, specifics that I will not go into here. At the same time, the structure of what I have been calling Kant’s “grammar” of experience is so minimal and sparse that it may well be unacceptable today—and would certainly not have been acceptable to the speculative grammarians— under that name. But in outline, we can see the common thread that runs through their approach and Kant’s. Both seek, through reflection, the presuppositions that must be made for meaningful statements— in Kant’s language, “judgments”—to be possible. Both seek, as well, those structural features that range over all uses of natural language, or grammatical universals. And both argue that those universals in turn tell us much about the world and our understanding of it. Yet two crucial features obviously distinguish the Modistic approach from that of Kant. First, from our present perspective, is the 14. G. L. Bursill-Hall, Speculative Grammars of the Middle Ages: The Doctrine of the Partes Orationis of the Modistae (The Hague: Mouton, 1971), 35.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



Background of Kant's General Logic   69

rather naïve and parochial—if not contradictory—commitment the Modistae made to Latin as constituting the basis for a universal speculative grammar which, as Bursill-Hall notes, the Modistae raised to the level of a metalanguage.15 Of course, they were not the first—nor the last—to claim primacy for a specific natural language, and the notion that Latin could serve such a role, given its hegemony in the intellectual, religious, cultural, political, and commercial world of the West when the Modistae were writing may have made it seem considerably more plausible then than now. In any case, Kant rejects any natural language as a candidate for a model of a universal grammar, but focuses on the logical features that any such language must satisfy. Indeed, Kant doesn’t, strictly speaking, limit these features to human beings, but rather views them as conditions for any possible judgment made by any agent for whom we can claim rationality: minimally, an agent who can refer to itself successfully with the first-person pronoun, who uses concepts in forming judgments, and who passively receives sensible intuition.16 Second, and philosophically of much greater significance, is the assumption the Modistae adopt that language mirrors reality, and thus that the investigation of language allows us to “read off ” the structural features of that reality.17 This, as mentioned above, is the fundamental metaphysical assumption Kant rejects, and, indeed, holds as largely responsible for the history of metaphysics, a history Kant regards as an unending battle of fruitless speculation and failed expectations. In place of this assumption, of course, is that perspectival shift now com15. Ibid., 38. 16. In short, such an agent is simply distinguished by Kant, on this account, from non-human animals, whose intellect he views as wholly passive, and from a being such as God, whose intellect he views as wholly active. 17. To be sure, this view of the Modistae is not itself without controversy: Pinborg notes that the Modistae “were not so naïve as to think that every sentence corresponded to the structure of reality” and that “What mattered to the [speculative] grammarian was the way reality was described, not reality itself ” (“speculative Grammar,” 261), while Trentman claims that for Siger of Courtrai, and by extension the other Modistae, “the general structure of our thinking mirrors the most general structure of the world.” J. A. Trentman, “Speculative Grammar and Transformational Grammar: A Comparison of Philosophical Perspectives,” in History of Linguistic Thought and Contemporary Linguistics, ed. H. Parret (Berlin: De Gruyter, 1975). Trentman’s view, at least in this case, seems to shared by Bursill-Hall (Speculative Grammar of the Middle Ages, 31).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

70   Background of Kant's General Logic monly referred to as the “Copernican Revolution,” where those universal and necessary—a priori—features are contributed by the subject. It need simply be noted that for those who regard the Kantian project with some sympathy, it is only within the context of this perspectival shift that Kant’s claims make sense.18 Given Kant’s conception of cognition, necessity is thus a feature only of judgments, requiring, on the part of the judging subject, an active contribution to cognition. In sum, then, the necessary and universal features—the two complementary, but independent, criteria of the a priori (B4)—cannot be read off the world, as the Modistae claimed, but are features the judging subject contributes, a contribution that subject discovers only on the basis of what Kant calls “transcendental reflection.” That contribution, articulated in terms of the employment of a priori concepts and rules, as delineated in the Critique, would then constitute a universal grammar, or logic, of thought and reason. We see here a typical feature of Kant’s two-front strategy in action here: a fundamental challenge to the skeptic to make meaningful claims (including the capacity to make claims that assert skeptical views or conclusions) without presupposing (and satisfying) certain universal and necessary conditions, and a powerful critical weapon against the inevitable temptation to proceed cognitively beyond the legitimate boundaries of possible experience. A variety of fascinating, though strictly historical, questions arise from looking at Kant’s concerns with the structure of language in the context of the ongoing search for a universal grammar. To be sure, it may already be seen as a considerably broadening of the very term “grammar” to include Kant’s remarks within this tradition. In spite of the remarks cited earlier, nowhere does Kant develop his own conception of a universal or transcendental grammar with any of the specificity one finds in more developed treatments. To be sure, Kant is as hampered by the lack of sophistication of what would qualify as the 18. I need only mention here as arguing for this view, among many others, Henry Allison, Graham Bird, Anthony Genova, Arthur Melnick, and Manley Thompson. An indication of the problems that arise in trying to make sense of Kant’s theoretical philosophy without treating it within this context can be seen in John McDowell’s Mind and World (Cambridge, Mass.: Harvard University Press, 1994), as pointed out at some length in Bird’s review (Philosophy 71 [1996]: 219–43).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



Background of Kant's General Logic   71

“linguistics” of his era, and his own relative ignorance of languages beyond German, French, and Latin, as he is by the formal apparatus of the logic of his day, at best an uneasy mix of syllogistic and Stoic logic.19 At the same time, his repeated insistence, throughout the Critical phase of his career, on the strong analogy between logic and what he refers to as “grammar” indicates a continued concern with the structure and foundations of language, and at least suggests that Kant believes his account makes a contribution to our understanding of the fundamental aspects of human language. A further question is that of determining—and this would require a historical investigation that, to my knowledge, has not been carried out—which, if any, of the central texts examining language from a “philosophical” perspective Kant was familiar with. He certainly would have been cognizant of the central themes, relative to language, in Locke’s Essay, and Leibniz’s response in the New Essays, as well as the views of Tetens, Rousseau, Süssmilch, and, of course, Hamann and Herder. It is, however, much more difficult to establish Kant’s familiarity with such texts as Harris’s influential Hermes, Condillac’s Essai sur l’origine des connoissances humaines, Wilkins’s Essay towards a Real Character and a Philosophical Language, or even the Grammaire Générale of Port-Royal.20 It seems likely that Kant knew of this tradition at best through secondary or even tertiary sources. I have been unable to find any indication that Kant had any familiarity whatsoever with the doctrine of the Modistae although, as we have seen, his views are surprisingly similar to theirs in significant ways. It seems at least a plausible historical conjecture that Kant was familiar, perhaps through Herder, with the general contours of these 19. Kant knew some Greek, although it is not clear how much; the general scholarly consensus is that he had virtually no ability to read English, his Latin was excellent, and he was quite at home in reading French. 20. Harris’s text was, however, known to Herder and Hamann as early as 1768 (see E. Cassirer, The Philosophy of Symbolic Forms, trans. R. Mannheim [New Haven: Yale University Press, 1953], 1:145), and thus some of his views may have been transmitted to Kant indirectly. Warda’s standard catalog of Kant’s library (Immanuel Kants Bücher (Berlin: Martin Breslauer, 1922) doesn’t indicate any copies of the kinds of texts on language discussed here, but is not a wholly reliable source in determining what texts Kant had access to or had read.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

72   Background of Kant's General Logic inquiries into language; had he been more cognizant of its details, it also seems plausible that Kant would have been able to find a place for himself in this tradition. There is no doubt that Kant’s remarks on grammar are scattered, unsystematic, and at best programmatic; we have seen why Kant chooses in the Critique to speak of “logic” rather than “grammar.” Yet Kant would have seen the Critical philosophy as making a crucial contribution to that tradition, in its recognition that any inquiry at the level of natural language introduces an ineliminable empirical, or contingent element, and thus that a universal grammar must provide those universal and necessary conditions making possible the employment of any natural language. As Svendsen helpfully reminds us, “any natural language has a semantic wealth which transcends these minimal conditions, but Kant has never denied that.”21 In any case, Kant’s “grammar” would be considerably more restricted and austere, its modesty no doubt dictated by the threat not just of skepticism, as mentioned above, and, interestingly enough, idealism, but the other more general threats Kant notes in the second-edition preface to the Critique: materialism, fatalism, atheism, free-thinking, fanaticism, and superstition (Bxxxiv). Such a grammar, of course, Kant would argue could succeed only within the perspective introduced by the “Copernican Revolution,” where the a priori is identified with the contribution made by the judging subject. Without pursuing any further here the knotty problems involved in the relation between thought and language, I think it is obvious that Kant recognizes the close connection between the two, and the formal elements of each. For a linguistic expression to be well-formed, it must satisfy certain syntactic conditions; for it to make an objective reference, it must introduce a semantic component by bringing content into consideration. For a judgment to be a judgment of possible experience, it must as well satisfy syntactic criteria, and also take into consideration the semantic dimension of relating the judgment to a manifold of sensible intuition. Finally, as Kant makes explicit, such syntactic rules are the product of reflection, whether a reflection on language or on possible experience. If such a set of rules can be isolat21. Svendsen, “Kant’s Critical Hermeneutics,” 264.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of



Background of Kant's General Logic   73

ed and justified, we would be in possession of a “transcendental grammar” qua a grammar of possible experience.

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Historical Background of Kant’s General Logic In the preface to her Language and Logic in the Post-Medieval Period, E. J. Ashworth notes that one of her goals there is “to describe just what logic a well-read man of the sixteenth or seventeenth century would have been acquainted with.”22 To do the same thing for Kant would be of no little interest, particularly once one recognizes that the idea that Kant simply identifies logic with Aristotelian syllogistic is inaccurate and oversimplified to the point of being misleading. The difficulties involved, however, are manifold: while we have Warda’s inventory of Kant’s library,23 we also know that Kant sold off parts of an extensive book collection when young due to financial hardship; while Pozzo and Oberhausen have provided extremely valuable information on course offerings at the Collegium Fredericanium,24 it remains unclear what precise information Kant himself was exposed to, particularly in terms of “formal” logic, and even more difficult to establish with confidence that Kant had not read, or was unfamiliar with, a given text. I do not want to go into this extraordinarily complicated developmental story of Kant’s conception of logic, in part because of the historiographical difficulties mentioned, in part because this is not a historical examination of Kant’s logic, but, most important, because ultimately this development plays a relatively small role in what I hope to establish about how Kant approaches logic within the Critique of Pure Reason. This is hardly to say that the research of Tonelli, Pozzo, Oberhausen, and others is not without enormous interest for gaining a fuller conception of Kant’s development, and certainly informs the language Kant employs in the Critique.25 Rather, the task here is to fo22. E. J. Ashworth, Language and Logic in the Post-Medieval Period (Dordrecht: D. Reidel, 1974), ix–x. 23. Warda, Immanuel Kants Bücher. 24. Pozzo and Oberhausen, Vorlesungsverzeichnisse der Universität Königsberg 1720– 1804. 25. In addition to the work already cited, I have profited from other work exploring this history: Terry Boswell, Quellenkritische Untersuchungen zum Kantischen Logikhand

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

74   Background of Kant's General Logic cus on Kant’s strategy in the Critique; where the history of logic can shed light on that strategy, I will utilize it. Initially, it is worth pointing out that Kant’s philosophical era was one of extreme ferment: as Giorgio Tonelli notes, “Köngisberg, and its university in particular, the Albertina, were the scene of some acrimonious struggles of religious-philosophical parties, which cannot have left no trace on Kant’s first philosophical perspective, nurtured in that environment.”26 In addition to the controversies Tonelli documents between the Pietists and the Aristotelians in Germany, and among the Wolffians and anti-Wolffians, Königsberg felt the impact of a wide variety of other influences, particularly from Great Britain and France, as well as the influential Romantic views of, for example, J. G. Hamann. Then, as now, one’s philosophical orientation in logic revealed itself throughout one’s philosophy (and vice versa); given the close connection many in Kant’s day recognized between logic and metaphysics, this is not a terribly surprising result. Thus one sees battles fought throughout the eighteenth century (and before) over a number of questions: whether logic was an art (ars or Kunst) or a science (scientiae or Wissenschaft), or both; what was the relationship, if any, between the analytic and dialectic within logic; whether logic was better characterized as logica docens or logica utens (or both, or even neither); whether logic was solely critical or could serve as a doctrine; similarly, whether logic provided a canon for thought, or an organon, as well as what these Aristotelian-inspired terms implied about the relationship between judgments and what those judgments were about, thus raising a number of questions about nominalism, realism, and other metaphysical positions, controversies that, of course, brought with them their own long histories. While, again, this history is fascinating in its own right, the point can be summarized rather quickly: in spite of the dominance of Aristotle’s logic as a formal method (syllogistic), there was no similar consensus about the role, use, and status buch (Frankfurt a.M.: Lang, 1991); Conrad, Kants Logikvorlesungen als neuer Schlüssel zur Architektonik der Kritik der reinen Vernunft; María Jesús Vázquez Lobeiras, Die Logik und ihr Spiegelbild (Frankfurt a.M.: Lang, 1998). 26. Tonelli, “Conditions in Königsberg and the Making of Kant’s Philosophy,” in bewußt sein: Gerhard Funke zu eigen, ed. A. Bucher, et al. (Bouvier: Bonn, 1975), 127.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of



Background of Kant's General Logic   75

Copyright © 2008. Catholic University of America Press. All rights reserved.

of logic; nor were these controversies new, but had been topics for discussion among logicians and, more generally, philosophers for centuries. Hence, if for this alone, it is naive to identify Kant’s conception of general logic—particularly on questions of the status and role of logic—with the Aristotelian syllogistic, although one certainly must recognize that he conceived of its formal apparatus in that fashion. What logic texts are we certain Kant was familiar with? We can, for example, be confident that he would have known the work—at least the logic—of Arnauld and Nicole (the “Port-Royal” logic), and significant similarities between the Port-Royal and Kant’s characterization of judgment are quite apparent.27 Furthermore, the Port-Royal logic adopts the traditional four-part structure of logical investigations (construing “logic” in the broad sense of the era): concepts, judgments, inferences, and method, a structure reflected in the Critique.28 He would obviously have been familiar with the texts adopted as standard texts for course lectures, such as the text he lectured from, G. F. Meier’s Vernunftlehre,29 as well as other standard or easily accessible eighteenth-century treatments of logic, including those by Crusius, Darjes, Lambert, Knutzen, Reimarus, and Rabe.30 Reinhardt Brandt approaches the question from a different perspective, asking with what logic texts and treatments the reader of the Critique would have been familiar; such a list, for Brandt, includes the Port-Royal logic, Jungius’s Instutitiones Logicae (1681), Bayle, Wolff, Reusch, Reimarus, and Lam27. See Longuenesse, Kant and the Capacity to Judge, 5: Kant “understands logic in much the same way as the Port-Royal logicians did, as the ‘reflection that men have made on the operations of their mind’”; also see 74ff. It is less clear that Kant would have known the Port-Royal Grammar (the Grammaire générale et raisonnée of Arnauld and Lancelot). 28. Brandt, Die Urteilstafel, 53. 29. Reprinted in vol. 16 of the Akademie edition, along with Kant’s own notes on this text. 30. C. A. Crusius, Logica (1753); J. G. Darjes, Die Lehrende Vernunft-Kunst (1737) and Via ad Veritatem (1755); J. H. Lambert, Neues Organon (1764); M. Knutzen, Elementa philosophiae rationalis seu logicae cum generalis tum specialioris mathematica methodo in ausum auditorum suorum demonstrata (1747); H. S. Reimarus, Vernunftlehre als eine Anweisung zum richtigen Gebrauche der Vernunft in dem Erkenntnis der Wahrheit (1756); P. Rabe, Dialectica et Analytica (1703). I rely on Pozzo’s work for much of this information; in addition to that already cited, see his “Catalogus Praelectionum Acadamiae Regiomontanae 1709–1804,” Studi Kantiani 4 (1991): 163–87 .

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

76   Background of Kant's General Logic

Copyright © 2008. Catholic University of America Press. All rights reserved.

bert.31 While interesting parallels can be noted between other writers and Kant’s conception of logic, both in the development of his views and in the view stated in the Critique itself, it is less clear that he would have had firsthand familiarity with such writers as Jacobus Zabarella, Richard Burthogge, Franco Burgersdicius, or John of St. Thomas, or even such a specific text as Jungius’s important Logica Hamburgensis.32 At the same time, there is no reason to think that Kant would not have gained at least some familiarity with these views from his teachers (e.g. Martin Knutzen) and his other philosophical contacts. Nor should one discount Kant’s reliance on Jacob Brucker’s history of philosophy as a source—and sometimes not a tremendously reliable one—for historical information, particularly for the views of Epicurus.33 This is of some significance, given Kant’s kind words for Epicurus in the brief “History of Pure Reason” (A854=B882; see A471=B499n, Ak. XXIV.1, 185); as Pozzo notes, a crucial aspect of Kant’s logic is its negative role in preventing error (logic as emendatio cogitationum), and, at least in this, it doesn’t go much beyond Epicurus.34 Kant, of course, says very little about all of this in the Critique, and is satisfied to summarize a picture of general logic as a science that “contains the absolutely necessary rules of thought, without which no employment whatsoever of the understanding takes place” (A52=B76). “A general but pure logic thus has to do with pure prin31. Brandt, Die Urteilstafel, 53–54. Kant mentions Reusch by name (Ak. XXIV.2, 701, 776, 796). S. Carboncini and R. Finster, in “Das Begriffspaar Kanon-Organon,” Archiv für Begriffsgeschichte 26 (1982): 53, note that “Kant knew well” Darjes’s Via ad veritatem. 32. Thus Tonelli notes (Kant’s Critique, 141) “Much more important, albeit presumably ignored by Kant, was Richard Burthogge’s Organum Vetus and Novum [1678]. It is well known that Burthogge is in a certain way a precursor of Kant’s methodology.” 33. Jacob Brucker [Iacobi Bruckeri], Historia critica philosophiae a mundi incunabulis ad nostram usqve aetatem deducta [1742–44]. Kant may have also consulted J. H. S. Formey’s Histoire abrégée de la philosophie (1760), or its German translation. See Young’s edition and translation of Kant’s Lectures on Logic (Cambridge: Cambridge University Press, 1992), 675n3. 34. Pozzo, Einleitung, 150; cf. Tonelli (Kant’s Critique, 183): “[W]e find in Brucker all characteristics attributed by Kant to Epicurus’ logic, including Kant’s explicit sympathy for his approach.” Also see Carboncini and Finster, “Das Begriffspaar,” 52–53. Indeed, as the latter suggest, this role may be the crucial aspect of Kant’s conception of logic.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



Background of Kant's General Logic   77

ciples a priori, and is a canon of the understanding and of reason, but only with respect to what is formal in their employment” (A53=B77). We will see some of the technical distinctions this characterization is based on below, but it should already be clear that Kant draws on a broad range of logical views (including, if not especially, Aristotle’s) to develop his own. This again should give pause to one who wishes simply to reduce Kant’s conception of general logic to Aristotle’s. As Brandt has put the point, “Historically, Kant proceeds eclectically and developed his thoughts, on the foundation of a number of logical doctrines themselves, about what general logic and its logicians have to say.”35 The fundamental result here is that general logic provides a set of rules for thought; as we have earlier seen, these rules have a peculiar modal status, in that they are necessary for the possibility of thought, and that possibility in turn leads to the (reflective) recognition of that necessity. While we can clarify this characterization by introducing some of the technical terms Kant employs, we may already be able to see that such rules provide necessary conditions ranging over a given domain, and thus may well provide only a framework within which specific and appropriate questions can be fruitfully examined. It is on this basis that we can start to see the modesty of Kant’s results, relative to transcendental logic, by emphasizing the fundamental importance of its construal as a logic. As we have seen, Kant (and his readers) tends to speak in terms of the judging subject as a human being. More precisely, however, the subject Kant characterizes is one that meets certain minimal conditions: the ability to use the first-person pronoun, that is, to correctly ascribe to oneself an “I,” the ability to regard oneself as free, and partic35. Brandt, Die Urteilstafel, 73n52. Pozzo (Einleitung, 81) uses the same term in summarizing Kant’s solution to the problem of general logic; a long developmental history leads to Kant’s “rather eclectic” solution in the Critique. See also Charles Nussbaum’s “Critical and Pre-Critical Phases in Kant’s Philosophy of Logic” (Kant Studien 83, no. 3 [1992]), where he argues that Kant was not entirely successful in departing entirely from his earlier Leibnizian orientation in logic; Pozzo seems to agree (Einleitung, 214) to the extent that Kant was never able to rid himself of the tradition—particularly the local Königsberg tradition—within which he was educated; Alan Shamoon (“Kant’s Logic” [diss., Columbia University, 1979], 169) argues that an ambiguity in Kant’s treatment of singular and universal judgments leads to the result that “there can be no definitive answer as to what consitutes (general) logic for Kant.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

78   Background of Kant's General Logic ularly important in this context, the ability to make judgments using concepts. While human beings presumably satisfy these criteria, these are not necessarily criteria that are descriptive exclusively of human beings. In any case, such an agent, to satisfy the most minimal condition of rationality, must think consistently, or at least must recognize the normative demand that it ought to think consistently. This is, as we have seen, the kind of agent Locke has in mind when he derides the notion that while God made human beings two-legged, they had to wait for Aristotle to attain rationality. Rather, such agents must think and communicate in conformity with—if not necessarily in recognition of—a set of rules, or constraints, that make thought (and communication) possible. On the basis of such an approach, one can attempt to identify and articulate that set of rules systematically. The distinction is captured in a 1677 British text by Thomas Good:

Copyright © 2008. Catholic University of America Press. All rights reserved.

Logick hath it’s name from λογας, ratio, because it's an art which teacheth Reason and Discourse. Ther is a twofold Logick, One Natural, which is nothing but natural reason, or of which every man (which is not an Idiot) doth in some measure partake: The other Artificial, which is the perfection of the Natural, nam Ars perficit naturam: this belongeth only to Schollars.36

Thus we have, using Good’s terms, a natural logic and an artificial logic; the former is that employed, often unconsciously, by any rational agent, while the latter is the systematic and rigorous development of that natural logic. This, of course, recalls Kant’s similar remarks on grammar: one may speak a natural language in perfect conformity with grammar while not conscious of its rules;37 the former would constitute, analogously a “natural grammar.” In the Wiener Logik, Kant applies similar considerations to logic: “Every man observes the rule before he can produce it as a formula [ehe er sie in Formeln bringen kann]. 36. Thomas Good, A Brief English Tract of Logick (Oxford: L. Lichfield, 1677). 37. The distinction between acting in conformity with a rule and consciously following a rule is much more of a philosophical commonplace now than in Kant’s time. “Behaviour fits a rule whenever it conforms to it; whenever the rule truly describes the behaviour. But the behaviour is not guided by the rule unless the behaver knows the rule and can state it. This behaver observes the rule.” W. V. Quine, “Methodological Reflections on Current Linguistic Theory,” in Semantics of Natural Languages, ed. D. Davidson and G. Harman (Dordrecht: Reidel, 1972), 442.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



Background of Kant's General Logic   79

But, slowly, he attends to that which he does. The epitome of all these rules is called logica naturalis. The science that presents these rules systematically is called [logica] artificialis” (Ak. XXIV.2, 791). Kant notes a few lines below this that by “logic” he means only logica artificialis. In short, reflection on “natural” logic provides a way of developing a systematic and rigorous “artificial” logic, which in the Critique forms the basis of—or better, provides a clue for—general or universal logic; a logic that can come forth as a science.38 The distinction Kant indicates here is an old one, developed as an Aristotelian-inspired distinction between logica utens and logica docens. This terminology can be found, among other places, in Aquinas, Zabarella, Wolff, and Meier’s Vernunftlehre; later, it is employed by Peirce: “Every reasoner, then, has some general idea of what good reasoning is. This constitutes a theory of logic: the scholastics called it the reasoner’s logica utens,” in contrast with the “scientific study” of logic. (It is worth noting that Peirce also remarks “There are certain parts of your logica utens which nobody really doubts.”)39 A logica docens presents a systematically developed, rigorous, and in some sense complete account of the rules for possible judgment, a logica artificialis that can yield a logic as a science. At the level of general logic, we are concerned only with the formal elements of judgment, and do not take into consideration content; in terms introduced earlier, the domain of general logic is the syntax of possible thought. It is when we to take into consideration that content that we move from the province of general logic to what Kant refers to as a the “logic of the special [besonderen] employment of the understanding” (A52=B76). It is within this transition 38. This is, however, not to say that natural logic, in any deep fashion, “grounds” (artificial) logic. As Pozzo (Einleitung, 162–63) points out in a discussion of F. Barone’s Logica formale e logica transcendentale (Torino: Cuneo, 1964), Kant rejects the notion of a “natural logic,” and thus rejects Christian Wolff ’s notion that a natural logic grounds a more formal and sophisticated account of reasoning. I think it is more precise to speak in terms of “natural logic” not being in a position to provide requisite systematicity and development in terms of principle (see A81=B107), and in that sense cannot ground general logic qua science. 39. Peirce, Elements of Logic, §§ 186, 188, 204–205; §192. Pozzo notes that “[t]his traditional distinction is found seldom [actually, only once, R. 1671, Ak. XVI, 72] in Kant, probably because he held it to be superfluous. Yet it is fundamental, implicitly, in his conception of logic”; Einleitung, 170.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

80   Background of Kant's General Logic that we introduce the notion of a transcendental logic and begin to see how Kant fully exploits the rich analogy between general and transcendental logic. While there are a number of technical distinctions one can explore in characterizing Kant’s conception of general logic,40 I want only to look briefly at one, Kant’s emphasis on logic as a canon, a notion that he seems to take largely from Epicurus, and that Kant standardly contrasts with “organon.” Kant takes a canon to be the “sumtotal [Inbegriff] of the a priori principles of the correct employment of certain faculties of cognition in general [gewisser Erkenntnisvermögen überhaupt]” (A796=B824). A canon yields a system of rules, and Kant here exploits the etymology of κανων (as well as its association with "criterion," specifically in relation to judgment), which means both a literal rule (e.g., a measuring stick used by a mason or carpenter), and "rule" in the normative sense of "standard." The contrast is with the employment of a logic as an organon: "By organon ..... we understand an instruction for bringing about a certain cognition. This implies, however, that I already know the object of the cognition that is to be produced according to certain rules. An organon of the sciences is therefore not mere logic, because it presupposes the exact cognition of the sciences, of their objects and sources" (Ak. IX, 13). To treat general logic as an organon is to use its formal rules as if they yield material content about an object; that is, to fail to abstract from all content of knowledge. When a logic is misemployed as organon, rather than in its appropriate role as canon, it gives rise to dialectic and systematic error and illusion. Paton has remarked that Kant here is "using terms that refer to a traditional distinction in formal logic; but the interpretation he gives to Dialectic seems to be a preparation for the Dialectic of Transcendental Logic."41 This is correct, I think, but it also precisely misses the point, a point often obscured by the fact that both 40. Here Tonelli is particularly informative; see, for instance, his extended summary of Kant’s use of the terms “organon,” “canon,” “discipline,” and “doctrine” (Kant’s Critique, 108–23), as well as “Das Wiederaufleben der deutsch-aristotelischen Terminologie bei Kant während der Enstehung der ‘Kritik der reinen Vernunft,’” Archiv für Begriffsgeschichte 9 (1964): 233–42. 41. Paton, Kant’s Metaphysic of Experience, 1:190.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



Background of Kant's General Logic   81

general and transcendental logic are, in important ways, formal; after all, Kant characterizes the pure modes of cognition given in transcendental logic as “pure and merely formal principles” (A62=B88; cf. A136=B174, A44=B62). The distinction between the two logics, as we will see, is made not with respect to their formality, but with respect to their application and employment—by distinguishing and specifying the domains over which their respective rules range. Thus Kant’s use of technical terms from the history of logic is to underscore what the systems of logic have in common as logics. Thus, Kant mentions the analytic of general logic at A60=B84; dialectical illusion in general logic seems to be associated with traditional formal fallacies, for example, the fallacies of four terms and illicit process, at A296=B353 (see A33=B390).42 This raises an important distinction, however, between “logical illusion” and “transcendental illusion”: when an error in reasoning in general logic is made perspicuous, the illusion disappears, while transcendental illusion persists, is “inseparable from human reason,” and continues to call for correction (hence the need for the “discipline” of pure reason) (A298=B354–55). Without delving too far into the details of Kant’s philosophy of mathematics, the contrast he makes between logic and mathematics, relative to whether either can function as an organon, is instructive in clarifying the function of a logic, and again points to important connections between general and transcendental logic. As we have seen, logic can serve only as a canon; used as organon, it gives rise to dialectical error. In contrast, Kant claims in the Jäsche Logik that “mathematics ..... is an excellent organon as a science containing the ground of the extension [Erweiterung] of our cognition in regard to a certain use of reason” (Ak. IX, 13). It is also worth noting that Kant indicates at A46=B63 that Transcendental Aesthetic can also serve as an organon. According to Kant, “mathematical cognition is the cognition gained by reason from the construction of concepts” (A713=B741). Such construction must be able to be exhibited a priori in non-empirical intuition; and the single mathematical object so constructed is considered 42. As Pozzo (Einleitung, 27) notes, Darjes distinguished analytic as the “logic of truth” and dialectic as the “logic of illusion.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

82   Background of Kant's General Logic as universally determined. In this fashion, and in contrast to philosophy, mathematics “hastens at once to intuition, in which it considers the concept in concreto, though not empirically, but only in an intuition which it presents a priori, that is, which it has constructed, and in which whatever follows from the universal conditions of construction must be universally valid of the object thus constructed” (A715– 16=B743–44). In mathematics, reason operates in precisely the way we saw him reject as a possible human capacity—namely, as an archetypal intellect, creating its own objects. The difference, however, is that on the rationalist scheme of things, there is no specific distinction between the ordinary objects as thought by an archetypal intellect and mathematical objects so conceived, which forms the basis of Kant’s rejection of Leibnizian (and Cartesian) schemes of cognition. For the human being, in the Critical philosophy, the distinction between objects of possible experience and mathematical objects is sharply drawn, with respect to the ontological claims involved. For Kant, “in mathematical problems there is no question of this [relation of perception to possible experience], nor indeed of existence at all” (A719=B747). As Thompson succinctly captures the point, “What appear as existence question in mathematics are really questions of constructibility and not existence.”43 Why, then, is mathematics an “excellent organon,” while logic (general or transcendental) used as organon leads to dialectical illusion? It is simply that within mathematics, objects are “created” by construction in pure intuition, and the formal rules for use in construction yield content about the objects so constructed. Mathematics thus contains “the ground of the extension of our cognition in regard to a certain use of reason.” But, again, this takes place within mathematics; as soon as genuine existence claims are involved, empirical conditions are introduced, for we cannot “obtain an intuition corresponding to the concept of reality [Realität] otherwise than from experience” (A715=B743). General logic yields no material content for thought in general, for it abstracts from all content about experience while providing rules that are necessary (but clearly not sufficient) for experience to be possible. 43. Thompson, “Singular Terms and Intuitions in Kant’s Epistemology,” 333.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of



Background of Kant's General Logic   83

In this way, a logic provides principles for the correct use of judgment, relative to a specified domain; as such, it yields a science of the universal laws necessary for thought and thus constitutes a canon. To attempt to apply these principles materially, as organon, is to attempt to extend them beyond their legitimate province, generating dialectical illusion. Consequently, were we to follow the procedure of mathematics in general logic, we would immediately be ignoring the limits of general logic, operating dialectically, and embracing transcendental illusion.

Kant’s General Logic: Some Conclusions

Copyright © 2008. Catholic University of America Press. All rights reserved.

What, then, is the role of general logic? General logic, as the canon of necessary rules for the possibility of thought—the systematic presentation of the a priori principles of the correct use of the understanding überhaupt (A796=B824)—yields what Carboncini and Finster describe as the sine qua non, or what I have described as a set of constraints on rationality simpliciter.44 This can perhaps best be seen by looking at what general logic can—and, always important for Kant, what it cannot—establish, relative to determining the truth of a given judgment. General logic establishes minimal necessary conditions for the possibility of truth, yielding what Kant calls the “negative touchstone of truth” (A60=B84). Such a negative criterion ensures only that a 44. “Das Begriffspaar,” 29. Informally, one might think of general logic as a “filter,” through which any judgment must pass in order to be evaluated as true or false; at that stage, a syntactically ill-formed judgment would be eliminated as a candidate. Transcendental analytic would function as a second filter, in terms of evaluating a judgment of possible experience as true or false. In this sense, then, general logic is the logic of thinking in general; transcendental logic is the logic of thinking about objects. (Without arguing for it here, I believe that the strictures of the categorical imperative in Kant’s practical philosophy function in a similar fashion.) The fundamental point here is to see that while a given judgment may well “pass” through such a preliminary process, that leaves us quite far from determining whether that judgment is in fact true (or false), and that Kant’s conception of logic seeks to establish precisely this result. Similarly, a moral judgment that does not violate the categorical imperative is not ipso facto the thing to do. In short, we should not expect a set of rules—whether of general logic, transcendental logic, or practical philosophy—to establish more than it is in a position to do qua a set of rules. I owe this way of formulating this aspect of the relation between the theoretical and practical philosophy to Daniel Farrell.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

84   Background of Kant's General Logic given thought is not self-contradictory,45 for “I can think whatever I please, provided only that I do not contradict myself, that is, provided my concept is a possible thought” (Bxxvi n.; see A569n=B624n).46 Presumably, this eliminates any obviously self-contradictory judgments as possible thoughts, for example, “Some men are not men,” or “No bodies are extended.”47 It may also exclude judgments about Meinongian “impossible” objects, such as prime numbers divisible by six. The more important point for Kant, however, is that a well-formed judgment may still fail to make an objective reference and that this failure becomes clear only with the introduction of the sensible manifold (and thus lies beyond the ken of general logic):

Copyright © 2008. Catholic University of America Press. All rights reserved.

It is, indeed, a necessary logical condition that a concept of the possible must not contain any contradiction; but this is not by any means sufficient to determine the objective reality of the concept, that is, the possibility of such an object as is thought through the concept. Thus there is no contradiction in the concept of a figure [Figur] which is enclosed within two straight lines, since the concept of two straight lines and of their coming together contain no negation of a figure. The impossibility arises not from the 45. Keeping in mind, of course, that thought—one might say here “propositional content”—is expressed by means of judgment, and that judgment itself must be expressed linguistically. 46. As Manley Thompson observes, there is an important distinction to be drawn here, relative to “thinking a contradiction.” “We think a contradiction only when we think it as such, as thought that cancels itself. In thinking a contradiction without thinking it as such, we fail to think anything at all—we are illogical.” “On A Priori Truth,” 471n8. For an extended discussion of Kant’s notion of contradiction, from a much different perspective, see Michael Wolff, “Der Begriff des Widerspruches in der ‘Kritik der reinen Vernunft,’” in Probleme der “Kritik der reinen Vernunft,” ed. B. Tuschling (Berlin: De Gruyter, 1984), 178–202. 47. This claim rests on Kant’s contention that we can determine in an analytic judgment whether or not the predicate is “contained” in the subject. L. W. Beck, in “Can Kant’s Synthetic Judgements Be Made Analytic?” (in Kant: A Collection of Critical Essays, ed. R. P. Wolff (New York: Doubleday, 1967), 3–22) helpfully distinguishes between a “logical” and a “phenomenological” criterion for determining the analyticity of a judgment; only the former seems able to withstand the consistent criticism, first systematically treated by Eberhard, that the analytic-synthetic distinction is variable, dependent upon how much one knows at a given time. The phenomenological criterion does seem vulnerable to this charge; as Beck recognizes, “Kant was not free from a psychologizing introspective tendency in his decisions on what is analytic and what is synthetic” (9).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of



Background of Kant's General Logic   85

Copyright © 2008. Catholic University of America Press. All rights reserved.

concept in itself, but in connection with its construction in space, that is, from the conditions of space and time (A220–21=B268).

The point is that a well-formed judgment can be made even about a figure enclosed within two straight lines, for its terms are not selfcontradictory. The impossibility of this judgment making an objective reference—albeit here to a mathematical object—is shown by the attempt to construct the object in pure intuition. The point can be generalized, relative to truth-claims: the truth of a given judgment can be determined only with respect to an object judged about. Thus no judgment can violate the principles of transcendental analytic “without at once losing all content, that is, all relation to any object [Objekt], and therefore all truth” (A62–63=B87). Similarly, a concept in the absence of possible experience functions as a mere idea—without truth, that is, without relation to an object (Gegenstand) (A489=B517). Because general logic abstracts from all cognitive content, it has nothing to say about any reference to an object, and can provide only a negative criterion of truth. Consequently, a standard characterization of logic, such as that with which Jungius begins his Logica Hamburgensis—“Logic is the art of our mental operations directed to determining the true and the false”48—must be interpreted quite narrowly on Kant’s conception of general logic, providing only a negative criterion for any such determination. In short, after the analysis of a judgment given by general logic, we are still quite far from determining the truth of that judgment, or even whether—at that point—it is a viable candidate as a judgment of possible experience. Kant introduces the question “famed of old ..... What is truth?” in the context of general logic. As Prauss persuasively argues, the question in this context is absurd in itself, and can find no response in general logic, for it is a question that must be dealt with by transcendental logic.49 In a vivid image, probably taken from Lucian of Samosata, or perhaps Virgil, and repeated from Kant’s Inaugural Dissertation, the 48. J. Jungius, Logica Hamburgensis, ed. and trans. (into German) R. W. Meyer (Hamburg: Augustin, 1957), 1; the original reads “Logica est ars mentis nostrae operationes dirigens ad verum à falso discernendum.” Kant would challenge the notion that a scientific logic ultimately is an “ars.” 49. In “Zum Wahrheitsbegriff bei Kant,” 166–182, Prauss gives a detailed analy-

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

86   Background of Kant's General Logic attempt by general logic to provide such a response leads to the “ludicrous spectacle of one man milking a he-goat and the other holding a sieve underneath” (A58=B82–83). If “truth” is taken to be the agreement of knowledge with its object, for a determination of truth we must turn to a logic that exhibits rules for cognitive claims relative to objects. Thus Kant calls transcendental logic—more precisely, transcendental analytic—a “logic of truth” (A59=B84). In contrast to mathematics, which begins with definitions, definitions come only at the end of philosophy (A731=B759); thus, when Kant calls the truth the “agreement of cognition with its objects,” he is careful to describe it only as the “nominal” definition of truth (A58=B82), not as a definition in the strict sense. This characterization of “truth” is one Kant repeats throughout the Critique, and he presents only this one aspect of his considered view of this issue. The result then is that it is only when the concepts and principles of transcendental analytic have been secured that we can then go on to inquire about the applicability of this nominal definition. In this sense, transcendental analytic serves as a “logic of truth,” in contrast to the (solely) negative criterion provided by general logic. Even then, however, we are not in possession of a universal and sufficient criterion of truth, but only in possession of a structure within which truth can be meaningfully inquired about. I now want to conclude this chapter with a brief summary of the above results, before finally returning to the famous remark Kant makes about the “completeness” of Aristotle’s logic. With that in hand, we can start to see what how Kant seeks to effect his strategy in the Critique, relative to transcendental logic and the analogy between it and general logic he exploits. sis of A57ff.=B82ff., beginning with the ambiguity of Kant’s statement (translated by Kemp Smith as “For if a question is absurd in itself ”) “Denn, wenn die Frage an sich ungereimt ist.” The ambiguity is whether this statement should be taken as (in Prauss’s reformulation) “die Frage als solche ist ungereimt” or “die Frage ist eigentlich (an sich) ungereimt”; that is, whether the question in itself is absurd, or the question is absurd in itself. Prauss argues that the former paints a history of philosophy in which all philosophers—before, after, and including Kant—are posing a question that as such is absurd, and thus suggests the latter reading, in which the question is absurd as posed by, or to, general logic. It should also be noted that Kemp Smith translates here “die Frage” as “a question”; the correct “the question” shows more clearly what Kant is getting at, and would seem to support Prauss’s interpretation. Guyer and Wood, and Pluhar, both translate the claim as “the question.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



Background of Kant's General Logic   87

All agents—all but that, to cite Thomas Good, “which is not an Idiot”—maintain beliefs that result in the issuance of judgments, embedded in language. Those judgments must follow some minimal set of rules, such as a commitment to consistency, to yield a coherent set of beliefs (where that set has at least two members).50 Anyone reflective enough to consider what those rules would look like might provide a logica naturalis or logica utens to characterize them, just as anyone reflective enough to consider the rules of language might develop a similar set of rules to be construed as a grammatica naturalis. Indeed, at one point Kant suggests that grammarians gave us our first hint about developing a logic, in the sense of reflecting on the use of language to specify those rules that were necessary for language to be possible. The development of logic, in terms of a logica artificialis or logica docens, required a systematicity in the rigorous specification of the rules of logic that were unavailable within “natural” logic, and only such an artificial logic could be regarded as a science, a result that could not be attained by grammar (due, presumably, to the unavoidable ambiguity inherent in natural language). A science of logic—general logic—as exemplified by the principle of non-contradiction, could then provide that set of rules which must hold universally for thought to be expressed as judgment, and thus can be construed as the sine qua non of thought simpliciter, or as a set of universal and necessary, or a priori, constraints on rationality itself. In ranging over all possible expressions of thought, such constraints would, of course, have to be recognized as holding— that is, would have to be presupposed—for any formal presentation of a logic, in the contemporary sense of a formal model of mathematical logic or a natural deductive technique. 50. I take it as obvious that one who had zero beliefs would not qualify as an agent; while, technically, one who had exactly one belief would not be in jeopardy of being inconsistent, it is not clear that such a being would qualify as an “agent” (except, perhaps, in some metaphorical or idealized Kierkegaardian sense). Indeed, the more interesting constraint may well come from the other direction: Christopher Cherniak has pointed out that even for a physically idealized computer, the “supercycle” of which is approximately 2.9 x 10–23 seconds, to use a truth-table to evaluate a set of 138 distinct beliefs could not be done “during the entire history of the universe.” Minimal Rationality (Cambridge, Mass.: MIT Press, 1986), 143n3. As Cherniak observes, a set of 138 atomic beliefs might well be too “small-minded” to be possessed by “anything we could call a full-fledged agent,” 94.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

88   Background of Kant's General Logic I have tried to defend above, in a variety of ways, this picture as Kant’s conception of general logic. This conception, to be sure, contrasts sharply with those we have already seen that reduce all of Kant’s understanding of formal logic to the Aristotelian syllogistic. But I think if we look more carefully at what Kant says about Aristotle, in the Critique and in his logic lectures, not only will we see why Kant said what he did about Aristotle, but if we interpret it correctly, we can see that Kant’s remark is in fact defensible. Consequently, we will see further that the standard picture of Kant’s logic systematically misrepresents and misunderstands what Kant says about logic, and thus the criticism grounded in those inaccuracies can largely be deflected. Stefan Körner has given a succinct description of what one can call this standard picture, that Kant “believes himself to have discovered all the absolute synthetic presuppositions of our thinking. They are, in particular, the presuppositions of arithmetic and Euclidean geometry, of Newtonian physics, and, in a sense, of the traditional logic. Since he believed these fields of thought themselves to be ultimate and permanent achievements of the human mind, he naturally regarded their presuppositions as absolute.”51 Focusing here on “traditional logic,” we see that Kant, on this account, is constrained by the conceptual and structural limitations of Aristotelian logic; he is, in short, forced to think in terms of Aristotelian logic, particularly syllogistic, when he characterizes logic. When he says (Bviii) that Aristotle’s logic appears to “finished and complete,” and unable to develop any further, he is compelled—if only for his historical situation—to dismiss those developments that logic had undergone since Aristotle (e.g., by the Stoics, Leibniz, and others). The subsequent history of logic since Kant’s era reveals Kant’s claim to be the glaring blunder that it is, and his philosophical approach thus suffers from his willingness to have “remained content with the poor version of traditional logic which he used.”52 Before turning to the question of what Kant might mean in this context by the “completeness” of Aristotle’s logic, it is worth looking 51. S. Körner, Kant (Middlesex: Penguin, 1955), 26. Körner never really elaborates on his qualification of “in a sense” relative to traditional logic. 52. Kneale and Kneale, The Development of Logic, 358.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of



Background of Kant's General Logic   89

at some of the other things Kant says about Aristotle, particularly from his logic lectures. Taken as a whole, the picture that emerges is one of respect for Aristotle’s accomplishments, particularly in logic, along with recognition of the limitations of Aristotle, the artificial “subtleties” of his logic, and the anti-philosophical spirit among those who uncritically venerated him (on the latter, see Ak. XXIV.2, 740). “Aristotle developed a blind trust in himself, and he harmed philosophia more than he helped it” (Ak. XXIV.1, 36). It is perhaps surprising that one who provided philosophy with a complete and perfect doctrine could harm philosophy more than help it, but Kant is clear: Through frequent observations we have scouted out the rules of the understanding. Aristotle established some; but these were nothing but road signs toward errors. It took great effort to forget such false propositions, to give the understanding its natural perfection again, and to investigate its true rules. (Ak. XXIV.1, 27)

Copyright © 2008. Catholic University of America Press. All rights reserved.

Aristotle can be regarded as the father of logic. But his logic is too scholastic, full of subtleties, and fundamentally has not been of much value to the human understanding....... There is much acuity in his organon. All our logical terminology is from him. Otherwise it tends to mythology and subtlety and is banned from the schools. (Ak. XXIV.2, 796; see Ak. IX, 20)

In these lectures, of course, Kant (or his auditor) tends to speak much more freely and expansively than he does in his more formal writing. But it is difficult, in any case, to reconcile the standard picture described above, that sees Kant viewing the Aristotelian syllogistic as a human achievement without equal, with a philosopher Kant characterizes as providing “nothing but road signs toward errors,” and whose logic “fundamentally has not been of much value to the human understanding.” Thus it is appropriate to ask what precisely Kant saw as Aristotle’s achievement, and in what sense it is “complete.” I earlier cited Manley Thompson’s distinction between contemporary conceptions of logic that take logic as having an “external subject matter” and a “neo-Kantian or neo-Tractarian” conception of logic, and argued that it is clear that it is the latter one finds in the Critique of Pure Reason. I think one way of getting at Kant’s regard for Aristotle’s accomplishment is to see the latter’s Organon as a first attempt to

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

90   Background of Kant's General Logic provide, in the West, a scientific logic, an attempt to give a systematic account of the rules of thought in general as a logica artificialis. At the same time, as Kant himself points out in the Critique, Aristotle proceeded haphazardly, and without using a principle to provide the systematicity demanded of a science (A81=B106). The idea seems to be that within Aristotle’s logic, a set of minimal concepts (and rules) can be found that are indispensable to any logic worthy of the name of “science,” but by lacking the appropriate philosophical methodology, Aristotle’s logic remained “rhapsodic” and was overwhelmed by “subtleties.” At the same time, the content characterizing the necessary and universal rules of thought, the minimal a priori principles of judgment, were in place qua logica naturalis. These conditions cannot change, in that they are simply the conditions to which a judging subject must conform in order to think. Yet the formal presentation of these rules must be given in a systematic, in-principle fashion for logic to attain the status of science. And this is precisely what Kant says in the Jäsche Logik: “From Aristotle’s time on, logic has not gained much in content, by the way, nor can it by its nature do so. But it can surely gain in regard to exactness, determinateness, and distinctness” (Ak. IX, 20; Kant’s emphasis). The a priori constraints on judgment, insofar as they provide necessary rules that range over all possible thought, cannot change; their presentation, of course, can.53 In no way does it seem, on this view, that Kant would be limited to syllogistic or unwilling to recognize more elegant presentations of those a priori rules; nor would this eliminate an acceptance of contemporary formal techniques in mathematical logic. 53. As one example, note Michael Friedman’s suggestion that we take Kant’s conception of logic as amenable to “at most, monadic quantification theory plus identity” in his Kant and the Exact Sciences (Cambridge, Mass.: Harvard University Press, 1992), 63n9, while rejecting Thompson’s claim that “[t]he general logic required by Kant’s transcendental logic is thus at least first-order logic plus identity,” a claim Thompson subsequently revised in arguing that predicating in terms of “Fx” (in contrast to Aristotelian “S-P” judgment-forms) would be a general logic developed from Kant’s transcendental logic. See M. Thompson, “Singular Terms and Intuitions,” 334, and the postscript added in Kant’s Philosophy of Mathematics: Modern Essays, ed. C. Posy (Dordrecht: Kluwer, 1992). The point, here, is that if Kant’s general logic is as “flexible” as Friedman and Thompson (and others) indicate, then it makes little sense to restrict Kant’s conception of general logic to Aristotelian syllogistic.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



Background of Kant's General Logic   91

In what sense, then, is Aristotle’s logic “complete”? In employing this term, it is obvious that Kant does not mean “complete” in the contemporary sense when logicians distinguish formal systems as “weakly” or “strongly” complete;54 as we have seen, this relies on not only a conception of logic that is alien to Kant, but one that, if Tonelli is correct, is alien to Western philosophy before the beginning of the nineteenth century. A more plausible candidate in this context is the notion that one has a set of necessary and sufficient rules for thought in Aristotle’s logic; while not presented in a way to Kant’s liking, a complete but unsystematic list of the rules of thought is to be found in the Analytics, or more broadly, the Organon. In other words, for any given judgment, all the rules we need for that judgment to be syntactically well-formed are in place; not only is there no need for others, the sufficiency condition indicates that there can be no others. In that sense, a set of necessary and sufficient rules would be complete. Kant is standardly interpreted in this way, and he occasionally speaks in ways that make this interpretation attractive (e.g., A13=B27). However, I don’t think that his conception of the completeness of general logic must be regarded as committed to this strong sense of completeness, and for philosophical and textual reasons it is more plausible to regard Kant’s notion that general logic is “complete” in the sense of only providing a set of necessary conditions. Certainly, Kant provides “strictly necessary” and “absolutely universal” as independent, and each sufficient, criteria of the a priori; when we describe the rules of general logic as a priori—which they indisputably are on Kant’s account—they clearly are delineated as necessary in the strongest sense possible. Thus an inventory of the rules necessary for thought would include those Kant has identified; the demand that they also satisfy the criterion of sufficiency is not required by (much of) Kant’s characterization of general logic. And it is worth noting that this result is supported by a somewhat surprising source, P. F. Strawson: “The fundamental logical operations or forms of judgment recognized in Kant’s table are such as are, and must be, recognized in any general logic worthy of the 54. A formal system is “weakly complete” if and only if all logically true wellformed formulas are theorems of the system; a formal system is “strongly complete” if no contradiction results from the addition to the system of an independent axiom.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

92   Background of Kant's General Logic name.”55 Kant also seems to indicate that “complete,” in the sense of providing necessary and sufficient conditions, is better thought of in terms of a Transcendental Idea, where we posit “a certain collective unity as the goal of the activities of the understanding” (A644=B672). Along these lines, he indicates how completeness functions as a desideratum: “A correct logic is like a straight line, so that one must not deviate from it either to the right or to the left. It would be desirable [es wäre zu wünschen] that logic could be brought [gebracht wäre] to such correctness and perfection” (Ak. XXIV.1, 28). We will return to similar considerations when looking at Kant’s claims relative to the completeness of the table of categories. In short, it seems that Kant’s notorious comment about Aristotle’s logic tells us much more about Kant’s conception of logic, as a set of rules that are necessary and universal (a priori) for the possibility of thought, than it does about what he thinks of Aristotle.

Copyright © 2008. Catholic University of America Press. All rights reserved.

55. P. F. Strawson, “Sensibility, Understanding and the Doctrine of Synthesis: Comments on Henrich and Guyer,” in Kant’s Transcendental Deductions, ed. E. Förster (Stanford: Stanford University Press, 1989), 71. One might compare another surprising source—Quine—who notes that Kant’s conception of logic excludes any notion of set theory, and suggests that we reserve the very word “logic” for the kind of elementary logic Kant employs. See W. V. Quine, “Carnap and Logical Truth,” in The Ways of Paradox, 103–4.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Ch a pter Four

The Metaphysical Deduction e

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Goal of the Metaphysical Deduction Up to this point, I have tried to clarify Kant’s conception of general logic as a set of universal and necessary rules ranging over the possibility of thought, relative to a generic kind of thinking subject. This subject is able to refer to itself using “I,” employs concepts to make judgments, and can regard itself as free. Relative, again, to this kind of subject, general logic functions to identify a set of conditions that this subject can come to see, reflectively, as necessary for the possibility of its own thought. Furthermore, any agent one takes to satisfy the above criteria—that qualifies as this kind of subject—will also be able to discover, reflectively, this set of conditions for the possibility of thought.1 I claimed earlier that to understand Kant’s approach in the Cri1. To be sure, Kant recognizes that there is no guarantee that everyone will be successful in identifying a set of universal and necessary conditions for thought—that is, general logic—for there is no a priori guarantee that a person will be not be deficient in judgment; as Kant puts it, following Peter Ramus, there is always the possibility of one being deficit in secunda Petri (A134=B173n). It is unlikely, given some of Kant’s other comments (e.g., in the Lectures on Anthropology and the Blomberg Logik), that someone who qualifies as a thinking subject could not come to recognize some necessary conditions for the possibility of thought, but this is of course a contingent claim.

93

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

94   The Metaphysical Deduction tique of Pure Reason, and to evaluate its success, it is crucial to see that the vast majority of the book is structured around the analogy Kant puts forth between general logic and transcendental logic, that is, between (1) a set of universal and necessary conditions for the possibility of thought and (2) a set of universal and necessary conditions for the possibility of experience. The fundamental component of this analogy is Kant’s transition from the table of judgments to the table of categories; only if this transition can be seen as at least plausible, if not entirely defensible, does the analogy hold. As Reinhard Brandt has put the point even more strongly: “If there is a single foundation [Fels] on which the doctrines of the Kantian philosophy can be constructed, then it must be the table of judgements”; R. P. Horstmann urges that “the metaphysical deduction is an indispensable precondition for the possibility of the transcendental deduction.”2 In short, without establishing a clear relationship between the two tables, Kant loses his most compelling argument, that of the Transcendental Analytic of the Critique, for taking the categories to be the universal and necessary conditions for the possibility of experience. The argument for the establishment of this relationship has come to be known as the “metaphysical deduction,” and this chapter will seek to clarify and, in a basic sense, defend that argument. As noted earlier, what the very term “metaphysical deduction” refers to is both obscure and controversial, even for this text. It is not clear, for instance, where the argument is, how it proceeds, or what it establishes. Nor is it entirely obvious why—or if—it is characterized by Kant as “metaphysical,” or if it qualifies as a “deduction.” And even if we can identify the argument and its structure, and clarify why Kant calls it what he does, its success remains the focus of widely diverging interpretations and evaluations. The expression “metaphysical deduction” appears only once in the Critique, and only in the second edition. There Kant says, “In the 2. Brandt, Die Urteilstafel, 1. Fels can also mean “cliff,” and the ambiguity of this imagery is not inappropriate here. Rolf Horstmann, “The Metaphysical Deduction in Kant’s Critique of Pure Reason,” Philosophical Forum 13, no. 1 (Fall 1981): 44. For the original German text, B. Tuschling, ed., Probleme der “Kritik der reinen Vernunft” (Berlin: De Gruyter, 1984), 33.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



The Metaphysical Deduction   95

metaphysical deduction the a priori origin of the categories has been proved through their complete agreement with the general logical functions of thought” (159). Given the importance of this sentence, and the fact that “metaphysical deduction” has virtually a canonical use in the literature on Kant, it is perhaps worth looking at it a bit more closely. Here Kemp Smith, as he often does, takes “a priori” to be adjectival, when it is standardly employed by Kant as an adverb; he also employs a “strong” interpretation of “wurde ..... dargetan” by using “has been proved.” However, this sentence could also be rendered “In the metaphysical deduction, the origin of the categories has been exhibited a priori through their complete coincidence [Zusammentreffung] with the universal logical functions of thought.” Thus a question arises, whether Kant intends, in the argument to which he refers, to show a priori the origin of the categories, or to show the a priori origin of the categories. Yet given that Kant wants to exhibit the origin of the categories, which are to be employed a priori, this may well be a distinction without a difference. Furthermore, the manner in which Kant structures the entire sentence in which the reference to the “metaphysical deduction” is embedded indicates some kind of parallel between what is demonstrated in the Metaphysical Deduction and in the Transcendental Deduction, and there is little doubt that the latter involves some strong sense of “proof.” One other parallel Kant no doubt has in mind should at least be noted. In the Transcendental Aesthetic, Kant provides both a metaphysical exposition [Eröterung] and a transcendental exposition of both space and time. An exposition provides “a clear (if not necessarily exhaustive) representation of that which belongs to a concept”; it is metaphysical when it “exhibits [darstellt] the concept as given a priori” (B38; this terminology is, significantly, introduced in the secondedition Critique). In contrast, an exposition is transcendental when it explains a concept “as a principle from which the possibility of other a priori synthetic cognition can be understood” (B40). As Horstmann glosses this contrast, “the metaphysical aspect of the exposition of a presentation has got something to do with securing the status of a presentation, whereas the transcendental aspect of such an exposition shows the types of possible modes of knowledge [Erkenntnissen]

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

96   The Metaphysical Deduction

Copyright © 2008. Catholic University of America Press. All rights reserved.

that such a presentation provides.”3 In short, a metaphysical exposition has to do with the origin, presented a priori, of a concept; if Kant introduces this terminology in the second edition with the new term “metaphysical deduction” in mind, presumably this indicates a deduction that seeks to justify a priori the origins of the concepts to which he refers, namely, the table of categories. We will see below whether we should place much hermeneutical significance on the fact that Kant also characterizes the argument as a “deduction.” It should now at least be relatively clear what Kant wants to do in the argument that he refers to as the “metaphysical deduction”: demonstrate that a set of concepts—specifically concepts to be employed necessarily and universally, or a priori—can be isolated by exploiting the fact that they coincide with the functions of thought. Those functions of thought, universal and necessary, have been displayed in a table of judgments. The goal, then, of the metaphysical deduction is to justify a transition from that table to the table of categories, and to show that the latter can apply a priori to objects of possible experience. If this is correct, one could then say that the metaphysical deduction justifies the claim that some set of universal and necessary conditions can be applied to objects; in turn, if one follows Dieter Henrich’s reconstruction, the second-edition Transcendental Deduction justifies that they do so in its first half, and how they do so in its second half.4

The Strategy of the Metaphysical Deduction If the goal of the metaphysical deduction is to justify a transition from the table of judgments to the table of categories, and to show that the latter can be applied a priori in making judgments about objects of possible experience, the obvious question arises: how does Kant argue for this transition? After looking briefly at what must be regarded as the “received” view of Kant’s argument, I will suggest an alternate 3. Horstmann, “The Metaphysical Deduction,” 38. 4. I need not go into the details here of Henrich’s well-known, influential, and controversial argument about the structure of the second-edition Transcendental Deduction. I mention it here solely to clarify why Kant viewed the two arguments as fulfilling complementary tasks, which would help explain some of the parallel indicated between the two texts at B159.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of



The Metaphysical Deduction   97

Copyright © 2008. Catholic University of America Press. All rights reserved.

way of looking at it, which has the advantages of both greater textual fidelity and philosophical plausibility. The received view imposes upon Kant’s text a position that is, at best, quite difficult to defend; it then rejects his solution as being unable to provide that defense. The alternate interpretation I will argue for takes Kant’s claims to be considerably more modest, and in turn considerably easier to defend. As noted earlier, this will also help demonstrate that Kant hopes—in his own tortuous way—to appeal to a pre-philosophical (or at least pre-reflective) conception of the rules of logic resonant with “common sense.” In Kant’s Transcendental Logic, T. K. Swing complains that “Kant never explains how the Table of Categories is derived from the Table of Judgments. He simply presents the two tables one after the other, apparently assuming that the derivation of one from the other is obvious.”5 In what I have called the “received” view, most commentators on the metaphysical deduction have also employed the term “derive” to characterize Kant’s argument. Tonelli characterizes the argument as consisting in the “derivation of the categories from the table of judgments,” Strawson refers to the “derivation of the categories from the Table of Judgements,” Ralph Walker describes the metaphysical deduction as “the argument which derives the categories from the twelve forms of judgment,” and J. Michael Young claims that the categories, on Kant’s view, “derive from the functions of thought.”6 5. T. K. Swing, Kant’s Transcendental Logic (New Haven: Yale University Press, 1969), 19. 6. Tonelli, Kant’s Critique of Pure Reason, 233; Strawson, The Bounds of Sense, 82; Ralph C. S. Walker, Kant (London: Routledge and Kegan Paul, 1978), 74; Young, “Translator’s Introduction” to Kant, Lectures on Logic, xvi. In his generally very accurate translation of Brandt’s The Table of Judgments, Eric Watkins renders “Die Verstandesbegiffe oder Kategorien lassen sich nach Kant aus der Urteilstafel gewinnen” as “According to Kant, the concepts of the understanding or categories can be derived from the table of judgements.” Even Henry Allison, whose interpretation of Kant’s argument here is extremely careful, and sensitive to Kant’s language, seems to view the argument as a derivation, as when he suggests that “it seems highly likely that the list of logical functions ..... actually presupposes the list of categories which is supposed to be derived from it.” Kant’s Transcendental Idealism, 128. Reich’s Die Vollständigkeit uses Ableitung and ableiten throughout his text, seeking thus to “derive” the table of judgments from the original principle of the transcendental unity of apperception, an ambitious but, I believe, fundamentally misguided project for the reasons discussed here and in L. Krüger, “Wollte Kant die Vollständigkeit seiner Urteilstafel beweisen?” Kant Studien

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

98   The Metaphysical Deduction Standardly, of course, a “derivation,” particularly in the context of a system of formal logic, is said to produce a given claim (proposition, sentence) from another; if q can be shown to follow from p, by a series of inferential steps, then one can say that q is “derived” from p. In a more restricted sense of deductive validity, a given claim (proposition, sentence) is said to be derived from a premise or set of premises by employing a set of rules and showing that, on the basis of those rules, the claim follows through a series of inferential steps in accordance with those rules and that set of premises.7 While much about the metaphysical deduction is admittedly unclear, it does seem obvious that Kant does not claim that the table of categories can be derived from the table of judgments on any standard conception of “derivation.” To claim that Kant does attempt to effect such a derivation, and then criticize him for having failed to do so, (a) imposes upon the text an obligation it need not meet and (b) ignores what Kant himself says about this transition by introducing the technical term “derivation” into a discussion where that term is neither used nor, again, called for. Rather, the term Kant employs is Leitfaden, or “guiding thread.” Kant employs the term Leitfaden as early as 1747, in his “Thoughts on the True Estimation of Living Forces,” and continues to use it, if sparingly, throughout his career, as in the Phillipi Logik in discussing the subjective and intersubjective grounds of truth, and in a variety of ways in the Critique of Judgment. The Grimms’ Wörterbuch notes that Leitfaden was used as an ordinary term by Lessing in 1751, as well as by another writer Kant was familiar with, Christoph Wieland;8 it was also used by both Francis Bacon and Leibniz. Its common context was in describing the thread of Ariadne, used in helping Theseus escape 59 (1968): 333–56. See also Julius Ebbinghaus’s review of Reich in Deutsche Literaturzeitung 44 (1933): 2074–77, particularly 2075. To be sure, Kant writes in Reflection 2233 (Ak. XVI, 279) that a system can be produced only by deriving it [“nur durch ableitung gemacht werden”]; but he distinguishes quite explicitly at A83=B109 a “system of pure reason” from his project in the Critique. 7. Thus Alfred Tarski: “[I]f within logic or mathematics we establish one statement on the basis of others, we refer to this process as a derivation or deduction.” Introduction to Logic (New York: Oxford University Press, 1965), 119. 8. J. Grimm and W. Grimm, Deutsches Wörterbuch (Leipzig: S. Hirzel, 1885), 6:736–37.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



The Metaphysical Deduction   99

the Labyrinth after slaying the Minotaur. While it may be ironic that Kant’s Leitfaden seems rather to lead us into a labyrinth, it was a common enough expression. Philological considerations aside, the point is that a Leitfaden provides only a hint—a guiding-thread—for the move from the table of judgments to the table of categories. Robert Pippin recognizes this when he calls the transition between the two tables “a derivation of sorts,” made possible by means of this Leitfaden, remarking further that Kant uses the term in an “odd, almost Pickwickian fashion.”9 Kant’s Leitfaden does not provide the grounds for a derivation of the table of categories, and certainly not in the sense many have seen as the burden of the metaphysical deduction. For example, on this more modest reading, a categorical judgment of subject-predicate form could provide some kind of clue for bringing in the concept of “substance” in making a judgment about an object. This is a considerably weaker claim than trying to derive the concept of substance from the categoreal form of judgment. We will see below what precisely this Leitfaden does establish, but it is worth noting that two authors, whose general interpretations of the Critique could hardly be farther apart from one another, at least agree on this point. Thus Martin Heidegger insists that we take Kant’s language seriously here, and regard his comments as providing nothing more than a clue; H. J. Paton also comments that the “more modest description” of the metaphysical deduction as “’The Clue’ is in some ways preferable.”10 Even if we see Kant’s burden in the metaphysical deduction as the 9. R. Pippin, Kant’s Theory of Form (New Haven: Yale University Press, 1982), 90. 10. Martin Heidegger, Kant und das Problem der Metaphysik, 4th ed. (Frankfurt a.M.: Klostermann, 1973), 54 (§12); Paton, Kant’s Metaphysic of Experience, 1:239n3. I employ here the term “metaphysical deduction” to refer to sections 9–12 of the Critique, simply for ease of reference, and as a term not doing much, if any, philosophical work. Indeed, I think, with Paton, that this may be about all we need, or can, do with it. I’ve already tried to indicate, albeit briefly, why Kant may have employed the term “metaphysical,” and I think it is also relatively clear why the term “deduction” should be regarded with some suspicion. On this latter point, Horstman observes that “C. C. E. Schmind, in his Wörterbuch zum leichteren Gebrauch der Kantischen Schriften ( Jena: 1798), recorded the transcendental as well as the empirical deduction as a key-word belonging to ‘deduction,’ but not the metaphysical deduction”; Horstmann, “The Metaphysical Deduction,” 17n4. In short, I think that—paradoxically enough—a number of Kant’s readers have been led astray by putting too much emphasis on the “metaphysical deduction” as a metaphysical deduction.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

100   The Metaphysical Deduction more defensible claim that the table of judgments offers a “clue” for the development of the table of categories, his other conception of the table of judgments—that it is systematic and complete—is, if anything, more controversial. It is Kant’s belief that all acts of the understanding can be reduced to or expressed in judgment, and that the various functions of thought can be systematically displayed, as in the table he gives at A70=B95. This view of Kant’s has come under severe criticism; in his typically colorful fashion, Jonathan Bennett thus writes that Kant’s “favoured dozen” judgment-forms “serve throughout the Critique only as a Procrustean bed on which he hacks and wrenches his philosophical insights into a grotesque ‘system.’”11 A common complaint takes the position that Kant had in mind a specific set of categories, and went about constructing a table of judgments in such a way that the latter could serve as a “clue” for the derivation of the former. There are problems with this charge, however, beyond attributing to Kant a rather unseemly tendentiousness. First, as we have seen, Kant thinks that general logic by its nature is complete, as far as its content is involved. Kant’s lectures on logic, where the general logic is much more fully presented than in the Critique, are based on G. F. Meier’s compendium Auszug aus der Vernunftlehre, itself based on Baumgarten’s text, which in turn was a condensation of Wolff ’s general logic, called by Kant “the best we have.”12 It would then seem entirely natural to take over the content of these earlier treatments; as Kant points out, while his table of judgments seems (scheint) to depart “from technical distinctions ordinarily recognized by logicians,” it does not do so in any essential way (A70–71=B96). Second, and more important, is Kant’s insistence that the table of judgments is developed from a specific principle, by means of which the understanding and its functions could be systematically displayed. It is this claim that distinguishes Kant’s approach from all earlier efforts, including Aristotle’s, and that he thinks guarantees the completeness of the table of judgments (see A81=B107–8). The point is summarized in the Prolegomena: 11. Bennett, Kant’s Analytic, 89. The image is from Schopenhauer’s “Criticism of the Kantian Philosophy,” the appendix to vol. 1 of Die Welt as Wille und Vorstellung. 12. Jäsche Logik, Ak. IX, 21.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of



The Metaphysical Deduction   101

Copyright © 2008. Catholic University of America Press. All rights reserved.

[I]n order to discover such a principle, I looked about for an act of the understanding which contains all the rest and is differentiated only by various modifications or moments, in bringing the manifold of representation under the unity of thinking in general. I found this act of the understanding to consist in judging. Here, then, the labors of the logicians are ready at hand, though not quite free from defects; and in this I was enabled to present [darzustellen] a complete table of the pure functions of the understanding, which were however undetermined in regard to any object [in Ansehung alles Objektes].13

The central claim is that there is a specific act of the understanding, judgment, that allows the manifold of representation to be brought under the unity of thinking in general. (Here Kant speaks in terms of “representations,” the broadest term in his philosophical vocabulary.) If such an act can be isolated and articulated, we would then have a set of rules—serving as syntactic conditions for the formation of judgments—that exhibit the various moments of this act. But all this raises a fundamental problem: Why did Kant think general logic, as presented in the table of judgments, was complete? What support does he offer for this position? Certainly, he never offers an argument in the Transcendental Analytic, where one might suppose such an argument would be found. Rather, he seems content to rely on such dogmatic assertions as that made in the firstedition preface: “What reason produces entirely out of itself cannot be hidden, but is brought to light by reason itself immediately the common principle has been discovered” (Axx; cf. Axiv, A79=B105). Occasionally, he is more reticent, admitting in the “B” Transcendental Deduction that no “further explanation” can be given “why we have only these and no other functions of judgement” (B146).14 I think 13. Prolegomena §39, Ak. IV 323–24 (Ellington’s translation, modified). The last clause repeats a remark Kant makes in the Critique where he introduces the table of judgments, which is central to the distinction between general and transcendental logic: “concepts, as predicates of possible judgements, relate to some representations of a not yet determined object” (A69=B94; my emphasis). See also the Metaphysical Foundations of Natural Science (Ak. IV, 475n), where Kant stresses the importance of an adequate and precisely determined conception of judgment as the key to solving the problem posed by the Transcendental Deduction. 14. See Prolegomena §39 (Ak. IV, 323), for a similar remark.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

102   The Metaphysical Deduction two points need to be made with regard to this claim of completeness. First, Kant provides a valuable—if generally overlooked—hint about how such completeness is to be viewed, although he does not pursue the point. The completeness of a given science “is possible only by means of an idea of the totality [einer Idee des Ganzen] of the cognition yielded a priori by the understanding; such an idea can furnish an exact classification of the concepts which compose that totality, exhibiting their interconnection in a system” (A64–65=B89–90). The sum of cognition produced by the understanding “constitutes a system, comprehended and determined by one idea [einer Idee]” (A65=B90). Kant, of course, has a great deal to say about ideas, particularly in the Transcendental Dialectic, where ideas are described as pure concepts of reason that systematically organize and unify the understanding. Ideas order the concepts of the understanding, and have only a regulative, not a constitutive, use. This regulative employment allows us to posit “a certain collective unity as the goal of the activities of the understanding” (A644=B672). Kant’s introduction to the Analytic of Concepts would seem to suggest that the completeness of the transcendental analytic is grounded in a maxim of reason, and thus involves a regulative employment of the idea of totality. It may be, then, that Kant intends the claim that the understanding can exhaustively survey its own capacities to be taken in this same regulative sense.15 Second, it is not clear how much relies on this question.16 There is a difference here between determining that a given set of judgmentforms necessarily conditions all thought and determining that this set is complete. By isolating the intellectual act of judgment, it seems rea15. In the Blomberg Logik, while discussing Lambert, Kant points out that it is a mistake among scholars to demand completeness of the means by which the conditions of cognition are given: “We can proceed in no other way in logic. The exercise of rules aren’t to be taught, since I can do nothing other than by means of these rules” [“Die Ausübung der Regeln geht nicht an zu lehren, denn ich kann es wieder nicht anders als durch Regeln thun.”]; Ak. XXIV.1., 339. 16. The locus classicus for a discussion of these issues is Reich’s Die Vollständigkeit der kantischen Urteilstafel. Krüger offers a much different perspective on this question, and I think a generally persuasive set of criticisms of Reich, in “Wollte Kant die Vollständigkeit seiner Urteilstafel beweisen?” A remarkable survey of 18th-century logic texts, with much relevance for the present discussion, is Giorgio Tonelli’s “Die Voraussetzungen zur kantischen Urteilstafel in der Logik des 18.Jahrhunderts,” in Kritik und Metaphysik (Berlin: De Gruyter, 1966), 134–58.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



The Metaphysical Deduction   103

sonable that various components of this act can then be methodically articulated. Whether or not this systematic classification is complete is certainly more problematic, but perhaps less crucial to Kant’s strategy; as Graham Bird notes, “It may well be that while Kant’s programme of classification is systematic, his execution of the programme is not complete.”17 It is perhaps enough, given the subject-predicate logic of Kant’s day, that his table of judgments is sufficiently broad in scope to capture the central aspects of judgment, while not omitting anything essential, in order to provide an adequately rich “clue” for the table of categories. It is worth quoting Strawson here again: the “fundamental logical operations or forms of judgement recognized in Kant’s table are such as are, and must be, recognized in any general logic worthy of the name.”18 Before we move to the argument itself, one last point should be made about Kant’s procedure in the metaphysical deduction, namely that the universal and necessary rules Kant seeks to identify, qua logic, tell only half the epistemological story. It is “just as necessary to make our concepts sensible, that is, to add the object to them in intuition, as to make our intuitions intelligible, that is, to bring them under concepts.” Both elements are necessary for cognition, and neither alone is sufficient; as we have seen, “only through being unified can cognition arise” (A51=B75–76). A full consideration of this unification of concepts and intuitions must take into account the notoriously obscure argument of the “Schematism,” however. As Kant insists, in writing to K. L. Reinhold, his principle for synthetic judgments—“all synthetic 17. G. Bird, Kant’s Theory of Knowledge (New York: Humanities Press, 1962), 93n. 18. Strawson, “Sensibility, Understanding, and the Doctrine of Synthesis,” 71. A much more difficult issue, which I cannot go into here, is raised by the development of quantificational logic. Thompson argues, for example (“Singular Terms and Intuitions,” 334), that the “general logic required by transcendental logic is thus at least firstorder quantificational logic plus identity”; he later emended this, in a postscript to this paper: “Instead of saying that in the general logic required by Kant’s transcendental logic the form of predication is that represented by ‘Fx,’ I would now say that this becomes the form when general logic is reconsidered and developed from Kant’s transcendental logic rather than taken as developed in abstraction from any relation of knowledge to object,” in C. Posy, Kant’s Philosophy of Mathematics (Dordrecht: Kluwer, 1992), 102. Michael Friedman suggests we take Kant’s conception of logic as amenable to “at most, monadic quantification theory plus identity”; “Kant’s Theory of Geometry,” Philosophical Review 94 (1985): 466n14.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

104   The Metaphysical Deduction judgements of theoretical cognition are possible only by the relating of a given concept to an intuition”—is presented “completely unambiguously” in the Critique, “from the Schematism of the faculty of judgement on.”19 Assuming this is a reference to the chapter entitled the “Schematism of the Pure Concepts of Understanding,” only at that stage is the connection between intuitions and concepts fully established. One of Kant’s most able early critics, his student J. S. Beck, criticized the order in which Kant presented these two elements of cognition in the Critique of Pure Reason. Beck felt that by Kant’s beginning the Critique with the Transcendental Aesthetic, and its discussion of the forms of intuition within which representations are received, readers were misled into identifying the sensible manifold with things in themselves, which is exactly what Kant most wanted to avoid. As Beck wrote Kant in 1794, “[A] proof that even the friends of the Critique don’t know what they [the Critique’s difficulties] are about is that they don’t know where they ought to locate the object that produced sensation.”20 Beck suggested beginning with the Transcendental Analytic and Schematism, thus establishing as the highest point of transcendental philosophy the transcendental unity of apperception, before relating it to the sensible manifold. Fichte, of course, took the point to its extreme, by grounding his Wissenschaftslehre in an absolute self, beginning with the positing of “A = A,” or “I = I.” It is not necessary to go into a detailed account of Beck’s and Fichte’s proposals to “improve” Kant’s method, but it is worth nothing that Kant did consider alternate approaches, as he mentions in a letter to J. H. Tieftrunk, such as beginning with the spontaneous awareness expressed by the transcendental unity of apperception and thus passing from the categories to appearances (as a priori intuitions).21 But Kant particularly wanted to avoid the conclusion that the mind somehow 19. Letter of 5.12.89, Ak. XI, 38, Kant’s emphasis. 20. Letter of 6.17.94, Ak. XI, 511. The point is made in detail in Beck’s Einzig-mögliche Standpunckt aus welchem die critische Philosophie beurrtheilt werden muß, first published in 1796 (Riga: Johann Friedrich Hartnoch), and now available in the Aetas Kantiana series (Brussells: Culture et Civilisation, 1968). See especially part IV, “Commentary,” 345–483, in particular 366ff. 21. Letter of 12.11.97, Ak. XII, 221–22. Henrich discusses this point in “The Proof Structure of Kant’s Transcendental Deduction,” Review of Metaphysics 22 (1969): 72ff.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.



The Metaphysical Deduction   105

simply constructs its objects of (non-mathematical) cognition, and thus he always emphasizes that the understanding functions in conjunction with a passively received sensible manifold—only in that way can cognition arise.22 While the “synthetic unity of apperception is ..... that highest point, to which we must ascribe all employment of the understanding” (B134n), this understanding’s only legitimate employment is in its application to the sensible manifold of intuition. Thus, rather than beginning with the formal apparatus of the understanding, Kant concludes that “[i]t is more natural to begin with the given, that is, with intuitions insofar as these are possible a priori, furnishing us with synthetic a priori propositions that disclose only the appearances of objects. For then the claim that objects are intuited only in accordance with the form in which the subject is affected by them is seen to be certain and necessary.”23 As Kant maintains late in his life, this is precisely the lesson Fichte failed to observe in proposing his Wissenschaftslehre: “[T]he principles of logic cannot lead to the material of cognition. Since logic, that is to say, pure logic, abstracts from the content of cognition, the attempt to cull a real object out of logic is a vain effort and therefore a thing that no one has ever done.”24 While Kant is speaking here specifically of general logic, I think the above considerations, and the argument of the Transcendental Analytic as a whole, reveal just how relevant this remark is for Kant’s characterization of transcendental logic, as well. Kant then seems to have the following in mind, relative to the table of judgments. For the human being—or in general, any agent that judges using concepts, in the manner described earlier—all thinking is expressed in judgment, and the faculty of judgment is the same as the faculty of thought (A69=B94; A81=B106). A given judgment must 22. In “Stoff and Nonsense in Kant’s First Critique,” 21–36, I argue that this line of thought informs Kant’s firm commitment to realism, albeit empirical realism embedded within the larger context of transcendental idealism. 23. Letter of 10.1.97, also to Tieftrunk, Ak. XIII, 463. See also Kant’s letter to Beck of 1.20.92, where Kant envisions a “System of Metaphysics” to overcome some of the difficulties of the Critique. “I should begin with the categories, in their proper order (after first expounding, without investigating their possibility, just the pure intuitions of space and time, in which alone objects can be given to them).” Ak. XI, 300; my emphasis. 24. “Open Letter on Fichte’s Wissenschaftslehre,” Ak. XII, 396 (Zweig, 253).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

106   The Metaphysical Deduction conform to certain necessary syntactic conditions, or rules, to be wellformed, and such rules are provided by general logic. In general logic, we abstract from all content of cognition, and “consider only the mere form of understanding” (A70=B95). Only the formal relations between representations are involved, regardless of their source or what they represent. As Paton reminds us, it is not that general logic has no object, but that “it ignores differences in objects.”25 Kant’s claim then is that the act of judging reveals the capacity to unite the manifold of representations under the unity of thinking in general, and that the various syntactic moments of this act can be displayed in a systematic table of judgments that provides a set of general or universal (allgemeine) rules for the possibility of thought. He also indicates, at least twice in introducing the metaphysical deduction, that the completeness of this table may be viewed as a regulative, not a constitutive, claim, insofar as its totality is referred to as an “Idea.” The goal, then, of the metaphysical deduction is to identify a set of judgment-forms that can be reflectively isolated and presented systematically; this set of forms, Kant believes, is in some sense complete, either because of his commitment to the claim that reason is able to provide an exhaustive inventory of its capacities, or to the claim that it functions heuristically as a regulative idea for reason’s self-investigation. The strategy Kant employs takes this table of judgment-forms to offer a clue, or hint, about how to proceed in isolating and articulating a systematic and, again in some sense complete, table of universal and necessary concepts for the possibility of experience, that is, for making judgments about objects of possible experience. More informally, if we can discover a rule—or set of rules—that thought itself must presuppose for thought to take place, we can say we have identified at least one rule that is necessary for the possibility of thought. When we turn from possible thought to possible experience, we should be able, on Kant’s view, to similarly identify a rule (or set of rules) that is necessary for the possibility of experience, albeit now in the context of applying those rules to sensible content that we do not ourselves gener25. Paton, Kant’s Metaphysic of Experience, 1:191n1. General logic, Paton continues, “always assumes that there are objects of thought, [but] it is under no obligation to explain what such objects are. It merely supposes that objects are given.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

The Metaphysical Deduction   107 ate.26 It is Kant’s general argument for this view to which I now wish to turn.

The Argument of the Metaphysical Deduction

Copyright © 2008. Catholic University of America Press. All rights reserved.

The central line of argument I have been attempting to develop here is straightforward: a logic is the systematic presentation of rules that range, universally and necessarily, over a given domain. Kant argues that just as the rules of general logic universally and necessarily range over the domain of possible thought, the rules of transcendental logic—specifically transcendental analytic27—range over the domain of possible experience. It is crucial in this context to see both what logic can provide and what logic is in no position to provide; namely, a logic can identify and clarify the concepts that constitute its rules, but cannot provide content beyond that, and thus offers only a negative criterion for determining the truth of a given claim. As we have seen, a sentence that violates a rule of general logic—for example, of the principle of non-contradiction—would reveal that the sentence in question could not be true. However, a sentence that satisfies those rules cannot thereby said to be true, for truth involves a reference to an object, and general logic abstracts from all such reference. In contrast, transcendental analytic provides a “logic of truth” (A59=B84), but only within the domain of possible experience; that is, a sentence 26. I simplify here to make a general point, ignoring such complications as judgments of pure mathematics, and reflective judgments about mental content that, in a complex way, may be said to generate objects of a specific type. 27. Kant could have made the analogy between the two logics much clearer had he referred consistently to either “general analytic” or “transcendental analytic,” and I will for the most part follow Kant’s looser usage and refer to “transcendental analytic” as “transcendental logic.” But it should be clear that both logics contain an analytical and a dialectical moment; Kant mentions the analytic of general logic at A60=B84; dialectical illusion in general logic is associated with traditional formal fallacies, e.g., the fallacies of four terms and illicit process, at A296=B353; cf. A333=B390. This raises an important distinction between “logical illusion” (here in the sense of general logic), and “transcendental illusion”: when an error in reasoning in general logic is made perspicuous, the illusion disappears (or at least should disappear), while transcendental illusion persists, is “inseparable from human reason,” and continues to call for correction (A298=B354–55). I once heard Paul Ricoeur characterize the latter quite nicely as “the tragedy” of human reason.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

108   The Metaphysical Deduction that fails to satisfy the rules of transcendental analytic cannot be a true judgment of possible experience, but a sentence that does satisfy those rules is not thereby determined to be true, but rather is revealed as truth-evaluable. This by itself should make clear how Kant conceives of the relationship between the two logics, although this conception has been disputed. Thus some philosophers have argued that general logic is somehow “prior” to transcendental logic; others take transcendental logic as prior, or as more fundamental.28 But in a well-known passage, which I will examine more carefully below, Kant is unambiguous and explicit:

Copyright © 2008. Catholic University of America Press. All rights reserved.

The same function which gives unity to the various representations in a judgement also gives unity to the mere synthesis of various representations in an intuition; and this unity, expressed most generally, we call the pure concept of the understanding. The same understanding, through the same operations by which it produces the logical form of judgment by means of the analytic unity, also introduces a transcendental content into its representations, by means of the synthetic unity of the manifold in intuition (A79=B104–5; my emphasis).

Rather than one logic being seen as prior to, or more fundamental than, the other, the two logics are complementary, and are distinguished in terms of their scope and application; they are distinct, in other words, due to the domain over which they range. Both logics play an essential role in the grounding of the possibility of cognition; as Gilead notes, “the two logics are but different uses or employments of the same understanding.”29 Consequently, the question of which logic is prior or 28. In chapter 1 of his Experience and Its Systematization (The Hague: Martinus Nijhoff, 1972), Nathan Rotenstreich gives the historical background to this controversy, presenting Reinhold’s and Maimon’s positions, and then extending the analysis to Hegel’s critique of the assumption on which the distinction between general and transcendental logic would rest. The controversy was renewed by Smart and Paton; see Harold Smart, “Two Views on Kant and Formal Logic,” Philosophy and Phenomenological Research 14 (1955): 155–71, and H. J. Paton, “Formal and Transcendental Logic,” Kant Studien 49 (1957): 245–63. 29. Amihud Gilead, “The Relationship between Formal and TranscendentalMetaphysical Logic According to Kant,” Monist 65 (1982): 435. Similar views are found in F. Grayeff, “The Relationship of Transcendental and Formal Logic,” Kant Studien 51 (1959–60): 49–352, and Allison, Kant’s Transcendental Idealism, 124.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Metaphysical Deduction   109 more basic is ill-posed, neglecting the fact that they serve distinct, ineliminable, and complementary roles. Ultimately, as we have seen, the two logics come together under the “highest point” of transcendental philosophy, the transcendental unity of apperception (see B134n). As I have tried to indicate, I think it is a mistake to regard Kant as deriving in any strict sense anything—whether the table of judgments from the principle of the transcendental unity of apperception, or the table of categories from the table of judgments—in that section of the Critique he refers to as the “metaphysical deduction.” At the same time, he does seek to establish a crucial point in those pages that culminate in the table of categories at A80=B106. Here I want to give a general overview of that argument, before looking at some of the specific, and controversial, issues it raises. After doing that, I want to try to give a more informal, and perhaps less echt-Kantian, picture, of what that argument fundamentally attempts to establish. For these purposes, I will regard the beginning of the metaphysical deduction as that point in the text where Kant introduces the “Clue” to the discovery of the categories (A66=B92), and concluding with the tabular display of those categories; the central part of the argument then comes in §§9 and 10, with the crucial move being introduced in the two paragraphs at A78– 79=B104–5. Kant begins by noting that concepts arise when we begin to reflect on how our cognitive capacities (Erkenntnisvermögen) are “put into play”; but a mere “mechanical” enquiry into these concepts fails to provide any sort of systematicity, and can offer only what Kant calls elsewhere an “assumed and comparative universality” (B3). In contrast, transcendental philosophy offers a principle that yields a rule for identifying those concepts that arise from the absolute unity of the understanding.30 Kant then goes on to identify the understanding with 30. As is probably obvious, I have not focused here on the absolutely crucial role played in Kant’s argument by the transcendental unity of apperception (cf., e.g., B131–32), for a number of reasons, among them that (1) it has already received an enormous amount of attention, given its role in the transcendental deduction (and, by Reich, in the metaphysical deduction); (2) I have tried to indicate aspects of this role in discussing earlier Kant’s conception of the judging subject; (3) I wish to keep this account within a reasonable scope; and (4) I place emphasis here on the analogy between general and transcendental logic.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

110   The Metaphysical Deduction the faculty of judgment and with the faculty of thought. These have earlier been characterized negatively, as non-sensible; they are now characterized affirmatively as discursive, and all thus essentially employ concepts. All concepts are grounded in what Kant calls, redundantly, “functions of unity,” where a function is said to provide “the unity of the act by which distinct [verschiedene] representations are brought under one common representation,” the latter being, evidently, a concept, although at the level of general logic, thought (or judgment) relates to undetermined objects. In any case, judgments always refer mediately to objects. At this point (§9), Kant displays the table of judgments, which is said to present the function of thought in judgment in various moments, yielding twelve judgment-forms, abstracted from all content. He then goes on to make four brief, relatively technical remarks about these forms, already foreshadowing (or surreptitiously importing) claims he will make about the role of the pure concepts of the understanding. Thus singular judgments can be treated as universals; traditionally, the judgment “Socrates is mortal,” while a singular judgment, is treated as a universal affirmative (“A”) judgment in that the reference of the term “Socrates” exhausts the entire class represented by the subject term. Similarly, infinite judgments are distinct from affirmative judgments, although Kant’s distinction in this case seems relevant only to transcendental logic.31 Judgments of relation are divided among those that relate the predicate to the subject in categorial judgments; hypothetical judgments, which relate two judgments; and disjunctive judgments, which relate the parts of cognition to a whole, distinct from the logical sequence of the hypothetical judgment-form by introducing the reciprocal relationship of community. Finally, the modality of a judgment, in contrast to the (presumably purely conceptual content provided by quantity, quality, and relation) other moments of the table of judgments, treats only “the value of the copula in relation to thought in general”; we thus get problematic judgments (the truth of which is “optional [beliebig]”), true judgments, and 31. As Paton remarks (Kant’s Metaphysic of Experience, 1:205), the distinction Kant points to here has to do with pure a priori cognition; “For Formal Logic, such a reason would be irrelevant.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Metaphysical Deduction   111 necessary judgments. These criteria, however, concern only logical, not objective, possibility. (It is worth noting here just how weak a claim a problematic judgment makes, the truth of which both is optional and involves only the minimal conditions imposed by logical possibility, which for Kant generally indicates simply not being self-contradictory.) Kant begins §10 by repeating the point that general logic abstracts from all content, now to distinguish it specifically from transcendental logic, which has lying a priori before it a manifold of sensibility. For this manifold to yield cognition, it must be taken up and connected through a spontaneous synthesis (that always precedes any analysis); that is, it must be grasped [begreifen] in a unified cognition. Kant further characterizes synthesis as a blind but indispensable function of the understanding,32 which brings the manifold to concepts and thus, by virtue of that function, produces cognition “properly so called [in eigentlicher Bedeutung].” It is synthesis in its most general aspect that yields the pure concept of the understanding. Kant then contrasts general logic and transcendental logic by noting that for cognition—the domain of transcendental logic—the manifold of pure intuition must first be given, but even by also introducing the imagination, one must further possess the unity given by the concepts of the understanding in order to have the set of necessary conditions satisfied for the cognition of objects. He then identifies the conceptual, unifying function of general logic with that of transcendental logic, with the crucial difference that the latter requires the introduction of transcendental content. This allows us to apply a priori the unifying concepts, which are necessary for synthesis to be possible, to the sensible manifold, to objects of possible experience. It is the latter that is the fundamental role played by transcendental logic and that distinguishes it specifically from general logic. The categories are essentially pure concepts that apply a priori to objects, and because general logic has nothing to say about such an application, it is silent.33 With this, and a quick refer32. As Schmidt notes in his edition of the Critique, Kant changes “einer Funktion des Verstandes” to “Funktion der Seele” in his own marginal notes on this passage, and I adopt this change here, given that Kant speaks in terms of the understanding in the very next sentence in continuing his point. 33. Here Kemp Smith’s somewhat loose translation is entirely appropriate—gener-

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

112   The Metaphysical Deduction

Copyright © 2008. Catholic University of America Press. All rights reserved.

ence to Aristotle, Kant introduces his table of categories, concluding the argument of the metaphysical deduction.34 As noted earlier, Hegel criticizes Kant’s “subjective idealism” as a philosophy of reflection.35 Without trying to respond to this criticism here, I think it is clear that what Kant calls “transcendental reflection” is, in fact, absolutely crucial to his philosophical strategy. While Kant seems to rely on the rather dogmatic presumption that the subject, by reflecting on reason, can somehow provide an exhaustive inventory of its capacities, and its limits, I think the process involved is considerably more defensible. Below I will give a somewhat more informal account of this procedure, but here I will look briefly at Kant’s own conception of this approach. Kant characterizes reflection (reflexio or Überlegung) as a procedure that has nothing to do with objects themselves, but is “the state of mind [Zustand des Gemüts] in which we first set out to discover the subjective conditions under which we can arrive at concepts” (A260=B316). By reflecting on, specifically, possible experience (and, of course, if it is done correctly), we discover conditions that make that experience possible, and further isolate those specific conditions that are necessary for that experience to be possible, or the transcendental conditions for experience, the pure concepts of the understanding. These conditions are said to be “transcendental,” in that all necessity lies in a transcendental condition (A106); if experience is possible, it al logic isn’t failing to say something about the relation between concepts and objects; rather, it “isn’t in a position” to say anything [kann nicht leisten]. That is, again, the task of transcendental logic. 34. He does add some significant comments about the systematic method followed to develop the categories (in contrast to Aristotle’s “rhapsodic” approach), characterizes the table of judgments as giving an exhaustive inventory of the understanding’s capacities, notes that the categories are primary, in contrast to what he calls “the predicables of the pure understanding,” and eschews defining the categories, which would be required in a system. He adds two sections (§§11 and 12) in the second-edition, which I also don’t look at here, in order to focus on the central claim in the metaphysical deduction. 35. See G. W. F. Hegel, Glauben und Wissen (Frankfurt am Main: Suhrkamp, 1970); as Ernst Cassirer argues against Hegel, however, “we miss the real meaning of the philosophy of the Enlightenment if we regard it and dispose of it merely as a ‘philosophy of reflection,’” The Philosophy of the Enlightenment, trans. F. Koelln and J. Pettegrove (Princeton: Princeton University Press, 1951), viii.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Metaphysical Deduction   113 is that possibility that reveals the necessity involved. It is this complex modal structure that seems to capture best Kant’s strategic conception of “transcendental,” and seems to be the point of his relatively obscure distinction of a “theorem” from a “principle” at A737=B765: the latter “has the peculiar character that it makes possible the very experience which is its own ground of proof.” What, then, if we choose to reflect (in the technical sense of “transcendental” reflection) on the possibility of thought itself? As we have seen, Kant focuses on the principle of non-contradiction as capturing the sense of a rule that is necessary for thought to be possible. Yet before we return to that perspective on the role of general logic, it needs to clarified what the table of judgments asserts, relative to judgments or possible thoughts. For as we have also seen, Kant identifies the two logics as providing the necessary and universal conditions for judgment: general logic ranging over the domain of possible thought, transcendental logic ranging over the domain of possible experience, where the latter introduces a priori the sensible manifold of intuition, or what Kant calls “transcendental content.” To see what is at stake in the metaphysical deduction, it must be determined what is, allegedly, established by this identity. The central claim seems to be that both judgment-forms and the categories are functions of unity; the former for making possible the unity of judgment simpliciter, the latter for making possible the unity of “synthesis,” by which Kant seems to mean in this paragraph (A79=B105) a judgment relative to the manifold of pure intuition.36 The question, relative to general logic, becomes simply this: how does general logic provide “functional unity” to judgments? While Kant could be a good bit more explicit, the point is that in any judgment, the thinking or judging subject employs concepts, qua marks, to bring diverse representations under a concept, insofar as a concept collects under a general term a variety of representations that share in that concept. Two important points should be kept in mind in this con36. It seems clear that Kant is being quite careful here with his language, for at this stage of the argument he has a great deal of work to do to distinguish ordinary judgments about external objects (including applied mathematics) from judgments of pure mathematics.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

114   The Metaphysical Deduction text, in characterizing general logic. First, Kant quite explicitly distinguishes an a priori intuition from a concept at B40; an intuition can be regarded as containing “an infinite number of representations within it,” whereas a concept contains its representations under it, roughly analogous to the relationship between genus and species. Second, as Kant repeatedly insists, general logic abstracts from all content, and its “sole task” is to clarify concepts. Thus, in a given judgment, for example, “roses are red,” we can give an exposition of the subject term “rose” and of the predicate term “red”; we can express various formal properties of the sentence (universal, affirmative, categorical, assertoric); we can apply various formal operations to the sentence (conversion per accidens, obversion, and other immediate inferences employing the square of opposition).37 It is sometimes difficult—particularly because Kant moves so frequently back and forth between the syntactic features of general logic and the non-syntactic elements introduced as the domain of transcendental logic—to maintain the austere level of formality of general logic. Thus in the judgment “roses are red,” we may be able to clarify what might count as a rose, and characterize the structure of the judgment (and its relationship to other judgments), but we do not have the apparatus to refer to an object as a rose, or even identify if something counts as a rose. As Melnick points out, while general logic can (now) symbolize such a judgment as “(∀x)(Fx → Gx),” and show its truth-functional equivalence to “(∀x)(∼Fx v Gx),” “the logician is not concerned with what domain the variables range over in the sentences.”38 As soon as reference comes into play, we introduce Kant’s “transcendental content,” and move to the domain of transcendental logic. And it is here that I think Melnick has precisely identified the crucial aspect of the “clue” provided by the table of judgments: judgments of possible experience must presuppose the minimal structure provided by general logic. As he puts it, relative to judgments of quantity, “the 37. With later developments in logic, the modern square of opposition eliminated a number of these inferences, such as contraries, sub-contraries, and alternation, due to existential assumptions made by the traditional square and rejected by the modern version. Given some of the things Kant says about the ontological argument in the Transcendental Dialectic, I believe his account is not in any significant conflict with this development. 38. Melnick, Kant’s Analogies of Experience, 39.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Metaphysical Deduction   115 very question of what is to count as an individual makes no sense for a subject who does not judge under quantificational (or some functionally equivalent) form.”39 In sum, when we turn to non-syntactic notions such as reference and still richer semantic notions, such as what Kant calls “subsuming”—“determining whether something does or does not stand under a given rule” (A132=B170)—we are no longer in the domain of general logic, although its rules still hold as presuppositions for making judgments relative to transcendental content, and hold for the rules for making those judgments, namely, transcendental logic. One can look at the hypothetical judgment form in a similar manner. Take the judgment Kant uses as an example in §20 of the Prolegomena: when the sun shines on the stone, it grows warm. Here we have, as we saw Kant observe, a connection between two judgments, “the sun shines on the stone” and “the stone grows warm.” At the level of general logic, we simply have a sequence of two judgments, where one precedes the other, and can denote that sequence as “p → q.”40 At this point, we are not in a position to determine a temporal or causal order, for to do so requires, in addition to “transcendental content,” the relational concept of causality, and, ultimately, the schematized version of the pure concept of causality, as presented in the principle of the Second Analogy. At the same time, we can see that a sequence of “if p then q” gives a hint or clue about applying a sequential rule to a given sensible manifold, in that if p precedes q, then we have at least a minimal conception of a sequence that, when schematized, would indicate that p must precede q.41 One way of using the latter as a rule of possible (causally oriented) experience is to identify p as a cause of q; that is, to make a truth-evaluable causal claim presupposes a sequence of two distinct events; the hypothetical judgment form allows us to 39. Ibid., 40. For a much more elaborate discussion of some of these aspects of the metaphysical deduction and Kant’s conception of judgment, see Longuenesse, Kant and the Capacity to Judge. 40. If Melnick is right, and the hypothetical judgment-form must be regarded as a non-material conditional, roughly that which captures the sense of a counterfactual or subjunctive conditional, then Kant’s claim may be seen as even less ambitious; see Melnick, 52–53. 41. I owe this way of looking at the argument to L. W. Beck’s “Once More unto the Breach: Kant’s Answer to Hume, Again,” in Essays on Kant and Hume, 130–35.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

116   The Metaphysical Deduction structure that sequence formally, and in that sense provides a clue to the conditional form employed in a causal judgment. Ultimately, of course, the conceptual unity provided by the understanding allows Kant to characterize cognitions in contrast to simple strings of associations, which grounds his response to Hume’s account of causal claims as mere “habits of the mind.” It is this that has allowed some to see Kant’s distinction between “judgments of perception” and “judgments of experience” as a reductio of Hume’s argument. That is, Kant presents this distinction as a way of indicating what a judgment would look like, were we to adopt an analysis based solely on association; if, as Kant argues, Hume’s account can provide only such an analysis, which on Hume’s own view is not sufficient to establish what he himself wishes to demonstrate, this indicates the inadequacy of Hume’s approach. This, in turn, shows that an adequate treatment of judgment requires the unity imposed by the conceptual underpinnings, yielded a priori through the transcendental unity of apperception, of the categories.42 In understanding the analogy Kant exploits between general and transcendental logic, it is crucial to see not only that the two logics are the same, insofar as they provide concepts (and, in turn, rules) for the employment of the understanding, relative to their respective domains, but also to see that they are unified in the judging subject. In the Transcendental Deduction, Kant notes that we must affix [heften] “all employment of the understanding, even the whole of logic” to the transcendental unity of apperception (B134n) While this could certainly lead (and has led) to the kind of interpretations amenable to a Fichtean and Hegelian metaphysical conception, Kant’s point here is more in tune with the realism of the Critique, particularly in its second edition.43 The point is made by Kant relative to the conceptual 42. In my dissertation, I noted that in this context, Kant could be seen as arguing polemically, in that on Hume’s treatment, we are left only with subjectively valid associations, not the objectively valid conceptual synthesis Kant argues for, more directly, elsewhere, in his comments on Hume. Longuenesse makes a similar point in “Kant et les jugements empiriques,” Kant Studien 86 (1995): 278–307, revised as chapter 7 of Kant and the Capacity to Judge; as she writes there, one way of looking at Kant’s distinction is “what would result if an associationist account of empirical judgement were the right one,” 187. 43. I have argued elsewhere (“Stoff and Nonsense in Kant’s First Critique”), that

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Metaphysical Deduction   117 role of the understanding in possible experience (i.e., transcendental logic), when he writes, “[O]nly in so far as I can grasp [begreifen] the manifold of the representations in one consciousness do I call them all my representations. For otherwise I should have as many-colored and diverse a self as I have representations of which I am conscious to myself ” (B134). It is clear that Kant takes it as obvious that the judging subject does not—cannot—have a self as diverse as its representations; therefore I must be able to grasp those representations conceptually in judgment. Consequently, in judgments of possible experience, I apply concepts to the manifold of intuition, and by doing so introduce unity by taking diverse representations under those concepts. While, again, it would help Kant’s own case in the metaphysical deduction had he been clearer, or at least more explicit, in those passages, the same argument would have to be seen as applicable to the unifying role played by general logic. General logic brings diverse representations under concepts, albeit abstracted from all semantic reference. If it did not, our thoughts would be as dissimilar as the representations over which our thoughts range, and the thinking subject would again be as “many-colored and diverse” as its representations. It is quite unclear what such a self would actually look like, and, indeed, it might be irresponsible even to refer to such an entity as a “self.” But it should be clear why he identified the “whole of logic” with that self represented by the “I think” of the transcendental unity of apperception. As we have seen, on Kant’s view the judging subject contributes concepts, rather than abstracting them from experience, but relative to an independent manifold of intuition. Kant’s focus, of course, is on those pure concepts of the understanding, the categories, which are necessary for possible experience, where the mark of that necessity is the indispensable role they play in making experience possible. At the same time, one can see this same transcendental strategy in the metaphysical deduction, relative to the domain of possible thought; without some set of universal logical concepts, thought itself would not be possible, and thought itself reveals the necessity of those concepts. Kant is misunderstood if he is not seen as urging a robust realism as a fundamental aspect of transcendental idealism.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

118   The Metaphysical Deduction There are difficult questions about whether Kant’s table of judgmentforms is in any sense “complete,” and as I have tried to indicate, less pressing questions about whether the categories can in any sense be “derived” from the table of judgments. But it seems considerably more plausible to suggest that thought itself must satisfy certain conditions to be possible, given the alternatives. I would now like to turn to a less formal description of Kant’s approach in the metaphysical deduction, in order to reinforce the idea that that approach is plausible, defensible, and relevant to a number of still current philosophical controversies.

Copyright © 2008. Catholic University of America Press. All rights reserved.

Logic and Common Sense Earlier, it was noted that Kant’s philosophical strategy was grounded in a particular form of reflection, transcendental reflection; in short, the idea is that by reflecting on the conditions that must hold for judgments relative to a given domain to be possible, one can identify a set of universal and necessary conditions that range over the judgments within that domain. By reflecting on experience, we can, that is, identify the universal and necessary (a priori) conditions of possible experience.44 What, then, if we choose to reflect (in this technical sense of “transcendental” reflection) on the possibility of thought itself? As we also saw, this immediately raises the question of “metacritique”: if we have to think in accordance with rules in order to determine the rules for thought, what justifies the rules with which we attempt to make that determination? A variety of responses are available, beyond embracing one of the many varieties of skepticism available: we might specify a logica utens or the way “we in fact reason in ordinary life”;45 we might 44. The assumption that we have possible experience distinguishes the analytic method of the Prolegomena from the synthetic method of the First Critique. Arguing for this assumption raises a number of technical points; however, here the question is whether one can safely assume not possible experience, but possible thought, and, as is familiar from arguments in Plato, Augustine, and Descartes, one might be forced to attribute either a rather extreme form of skepticism or simply an incoherent position to one who denied this assumption. 45. Pascal Engel, The Norm of Truth (Toronto: University of Toronto Press, 1991), 6; as noted earlier, the language that distinguishes between a logica utens and a logica docens can be traced back at least as far as Thomas Aquinas.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Metaphysical Deduction   119 simply say that many, most, or all of us adopt some set of practices, on the basis of a combination of history and convention; we might, say, with Wittgenstein, that we have exhausted the idea of “justification [Begründungen]” and “our spade is turned.”46 Kant never explicitly addresses the question of what it would mean to “justify” the principles or rules of general logic, yet it is clear from his remarks on Aristotle that to call into question those rules one would have to adopt a position that Kant would regard as incoherent. First, the very act of calling such rules into question will itself presuppose at least some fragment of the rules being challenged, for such a challenge to be meaningful.47 Second, as we will see, Kant is not so much interested in what we do in following rules, but what we have to do, or at least what we should do, in making a judgment. From a more contemporary perspective, however, the a priori status of any such rules has been challenged, in different ways by—among others—those advocating a “naturalized” epistemology as well as those labeled, simplistically, as “postmodernists.” By looking somewhat informally at Kant’s general position, vis-à-vis general logic, I think we can see how he might respond to these contemporary responses, and provide some insight into his own view that the rules of general logic can be justified as universal and necessary conditions for the possibility of thought. I will conclude this section by looking, once again, albeit briefly, at the analogy Kant seeks to exploit between general and transcendental logic. A student informs me, “I may or may not be in class this afternoon.” A sign outside a fast-food restaurant states, “Parking for drivethru only.” The host of a television cooking show tells her audience, “Don’t do as I say, do as I tell you.” An academic department adver46. Wittgenstein, Philosophical Investigations §217. Nietzsche, of course, gives us yet a different response; see my “Should the Skeptic Live His Skepticism? Nietzsche and Classical Skepticism,” Manuscrito 21, no. 1 (1998): 47–84, for some discussion of the issues involved. 47. While I have generally tried here to use the term “presupposition” in a nontechnical sense, it should be noted that the logic of presupposition has generated a good bit of controversy, as in P. F. Strawson, An Introduction to Logical Theory (London: Methuen, 1963), 175–79; Genova, “Kant’s Notion of Transcendental Presupposition in the First Critique”; Brittan, Kant’s Theory of Science, 28–42, which follows Bas van Frassen’s influential “Presupposition, Implication, and Self-Reference,” Journal of Philosophy 65 (1968): 136–52.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

120   The Metaphysical Deduction tises, “99 Things You Can Do with an English Major.” A highway sign reads, “Take Notice: When This Sign Is under Water, This Road Is Impassable.” If we choose to interpret these various sentences using only the tools of general logic, we do not get very far; the first, assuming bivalence, is the analytically true or vacuous “p v ~p”; the second seems to be contradictory, while the third—plausibly assuming that “say” and “tell” here are roughly synonymous—clearly is the contradictory (imperative) “~p & p.” The humor of the fourth arises from its ambiguity: is it referring to a person, or an academic program a person can follow? The fifth, interestingly enough, raises few problems whatsoever at the level of general logic, for its peculiarity rests on an odd semantic assumption.48 I think it is safe to assume here that all of these statements are, in their own ways, informative, but that general logic by itself simply doesn’t allow us to make that determination. To be sure, a statement that is explicitly contradictory cannot be true as stated; but by appealing to both semantics and pragmatics, we can interpret the statements given above in order to identify what is being communicated, or at least what one is attempting to communicate. This reveals a number of things about general logic, specifically that general logic provides a minimal set of parameters within which judgments can be made, that understanding a judgment goes well beyond the analysis provided by general logic, and that intersubjective communication will draw on a number of resources that work together. Thus, when I tell my student—who has informed that she may be in class, or she may not be in class—that I was already aware that that statement was true, she and I both are cognizant of the fact that the pragmatic implication of her remark is to indicate a relatively high probability that she will, in fact, not be in class. General logic is not in a position to provide that result. The point becomes somewhat clearer if we take as an example another student, who informs me that he will be in class and will not be in class. Again assuming bivalence, this statement (“p & ~p”) could not possibly be true, and it would, in fact, be an odd thing to hear. The 48. One might interpret this final example as assuming that a sign would be both visible (in order to be read) and not visible (in that it would be underwater), and thus contradictory. These examples, it should be noted, are all genuine.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Metaphysical Deduction   121 point to be made here about general logic is twofold: first, we know that on virtually all classical treatments of sentential logic this statement is false; second, we know that an agent who maintains that the statement is, or could be, true, either doesn’t understand the notion of contradiction or will have, when it is pointed out, an intellectual obligation to eliminate one of those beliefs from his set of beliefs. Here, as in a number of other places, we find Kant and Frege in agreement. In the various lectures on logic Kant contrasts the way we think with how, ideally, we ought to think, thus imposing a normative obligation on the agent to judge in accordance with the minimal rules imposed by general logic; as he puts it in a Reflection, logic provides the rules not “according to which we do think, but rather should think.”49 Frege insists on making the same distinction, particularly in the context of his anti-psychologism. Frege writes in The Foundations of Mathematics that “the question of how we arrive at the content of a judgement should be kept distinct from the other question, from where do we get the justification for our claim?” and is quite explicit that his phrase “the laws of thought” is to be interpreted normatively: the laws of logic prescribe what “ought” to be thought, and “the laws of logic [can] be called ‘laws of thought’ ..... so far as they stipulate the way which one ought to think.”50 Indeed, one could precisely characterize, without controversy, Kant’s conception of general logic in Frege’s own words: the laws of logic “are the most general laws, which prescribe universally the way in which one ought to think if one is to think at all.”51 It is here that two fundamental aspects of Kant’s transcendental strategy, relative to rules, converge: his suggestion that the highest point of transcendental philosophy is the transcendental unity of apperception (B134n), and the crucial distinction between the quaestio juris and the quaestio facti (A84=B116). Without attempting to unravel here the horrifying complexities of Kant’s account of the transcendental unity of apperception, we should note again that Kant’s strategy in 49. Ak. XVI, 43 (Refl. 1627); in Refl. 1612 (Ak. XVI, 36), he contrasts subjective from objective laws; the former characterize how we do think, while the latter characterize how we should think. 50. Frege, The Foundations of Arithmetic, 3; the translation here, by J. L. Austin, has been modified and emphasis added; Frege, The Basic Laws of Arithmetic, 12. 51. Frege, The Basic Laws of Arithmetic, 12.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

122   The Metaphysical Deduction

Copyright © 2008. Catholic University of America Press. All rights reserved.

many ways plays off the ambiguity of the terms “subjective” and “objective.” Thus a condition may be characterized as subjective insofar as it is imposed by the subject, yet be objective insofar as it is identified as holding for, or necessarily presupposed by, any and all such subjects; in short, the terms “subjective” and “objective” are not, in this context, opposed to one another, but rather complementary. Thus an agent who can refer to itself using “I,” and who judges using concepts, is seen by Kant as having the capacity to reflect on the content of its possible thought and experience and, by so doing, identify a set of rules that is necessary for that thought and experience to be possible. While this process of reflection is radically subjective, I discover it to be a universal and necessary (a priori) condition, and as such to be an objective condition.52 Thus an agent employing (explicitly or implicitly) the “I think” of the transcendental unity of apperception can be seen as committed necessarily to a set of rules that are identified subjectively but that function as objective conditions, in that the set holds for any such agent. As we saw, Kant imposes very minimal conditions to be satisfied by a judging subject; thus on this argument, anyone who so satisfies those conditions can discover, reflectively, those necessary and universal conditions, or rules, that hold for judgment to be possible.53 52. The term “objective” can also be seen as ambiguous; on the one hand, it can indicate a condition for the possibility of experience, or for the objects of experience (A158=B197); on the other hand, it can indicate a condition that we must regard as one “we cannot make otherwise,” as Thompson puts it (“On A Priori Truth,” 464). For Kant, the universal and necessary conditions for possible thought (general logic) and for possible experience (transcendental logic) satisfy both senses of “objective,” and yet, as contributed by the subject, are subjective. It is this ambiguity (and Kant’s notorious account of the thing-in-itself) that has led so many to regard Transcendental Idealism as either repackaging Berkeley or foreshadowing the kind of Absolute Idealism of Hegel, and especially Fichte; these approaches to Kant fail to take into consideration the robust Empirical Realism of the First Critique, particularly in its second edition, the modesty of Kant’s positive claims about what logic can demonstrate, and his insistence on the negative results of the Critique’s account. 53. Looked at in a different way, it is not clear that an entity that failed to satisfy these conditions (use of the first-person pronoun, judging by means of concepts, and ability to regard itself as free) would be an entity we would be willing to characterize as an agent. Again, I think the modal structure of such a claim needs to be clear: Kant is not saying that a given set of entities can be identified a priori as agents and thus satisfy these conditions; rather, he is saying if we attribute agency to an entity, then it must satisfy those conditions. The attribution itself, then, cannot be made a priori; rather, what follows that attribution follows a priori.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Metaphysical Deduction   123 This radically subjective approach also allows us to see why Kant would regard either as inadequate or as addressing a distinct question those who see such rules as, at best, mere historical conventions or, along the lines of Quine, as simply those things we would be least willing to give up. Kant seeks to respond to a question of warrant, or justification (the quaestio juris), and insists on distinguishing that inquiry from one into historical or conventional accounts, which he refers to as the quaestio facti (the question of fact) and, on occasion, derides as a “physiological” investigation, as when he criticizes Locke. The question of fact is addressed by marshalling empirical data and adducing generalizations, and in so doing identifying relatively widespread—or divergent—usages and behavior that can be seen as conforming to (or failing to conform to) a given rule. But insofar as such a procedure is inductive, it cannot establish the universality and necessity Kant takes as both the criteria of the a priori and fundamental to an adequate response to the question of justification. Kant’s various arguments, in attempting to address the question of justification for the pure concepts of the understanding, necessarily involve the introduction of a specified domain, the sensible manifold, or what he calls in the metaphysical deduction “transcendental content.” Yet that strategy becomes much clearer, and I think more defensible, when we see it as analogous to Kant’s conception of general logic, although the difficulty here is a bit different. To justify the pure concepts of the understanding a deduction is required; a “deduction” of the principles and rules for general logic is conceptually difficult to undertake—hence the problem of “metacritique”—in that even understanding whether or not such a deduction would be required presupposes some of the concepts that would somehow require justification. I think by looking at an image Wittgenstein employs we can see how Kant would have regarded the question of justification vis-à-vis general logic, and in turn see why the rules of general logic can be seen as absolutely universal and strictly necessary, for, in this context, there is simply no room for a coherent alternative. In On Certainty, Wittgenstein writes, “It might be imagined that some proposition, of the form of empirical propositions, were hardened and functioned as channels for such empirical propositions as

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

124   The Metaphysical Deduction were not hardened but fluid, and that this relation altered with time, in that fluid propositions hardened, and hard ones became fluid.”54 Kant, in isolating and justifying a set of rules said to constitute a “logic,” might be said to present this as such a channel or riverbed, but with two related, and fundamental, differences from the Wittgensteinian image. First, the channel itself, for Kant, does not consist of empirical propositions, and only in an attenuated sense can it consist of propositions of the form of empirical propositions. Rather, it consists of those kinds of claims that for Kant are transcendental—necessary for the possibility of making judgments relative to a given domain. If we take as one set of claims the principles of general logic, such a riverbed cannot change—its hard rock, that is, could not get washed away. Jonathan Lear, in commenting on this image of Wittgenstein’s, remarks, “Perhaps as we explore the river bed shifts will occur: certain things will begin to make more sense, other things less.”55 The point for Kant is that it is not clear—or, perhaps more accurately, it cannot be clear— how certain kinds of notions could make “more or less” sense. To return to a point made earlier: what would be involved in exploring the principle of contradiction? How, that is, could we effect such an exploration, without presupposing the very principle we seek to “explore”? If the very notion of “making sense” requires conformity to the principle of contradiction, then it hardly seems clear how the principle itself could make more or less sense. The rules of general logic, then, are already in play when we try to do any sort of investigation or critical examination; the alternative is to accept, at this level of analysis, any and all claims, which is intolerable.56 Consequently, we really have no alternative but to adopt some minimal principle of consistency among 54. Wittgenstein, On Certainty, trans. Denis Paul and G. E. M. Anscombe (New York: Harper and Row, (1969), §96. Also §99: “And the bank of that river consists partly of hard rock, subject to no alteration or only to an imperceptible one, partly of sand, which now in one place now in another gets washed away, or deposited.” 55. Jonathan Lear, “Leaving the World Alone,” Journal of Philosophy 79 (1982): 390. 56. Yet another way of putting the point can be found in Ross Harrison, On What There Must Be (Oxford: Clarendon Press, 1974), 56: “A protagonist for which both ‘a’ and ‘not a’ had exactly the same standing would not be a protagonist who was able to distinguish between his true and his false judgements, and would not be a protagonist who comprehended his world.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

The Metaphysical Deduction   125

Copyright © 2008. Catholic University of America Press. All rights reserved.

thoughts as a universal and necessary constraint on the possibility of thought; while it might be said that any further attempt at justification “collapses”—or that there is no further room for explanation—Kant’s conclusion also holds for anyone seeking to reject such a principle of consistency as necessary. For there is simply no coherent conceptual space or room within which that move can be made. It is currently much more fashionable to reject this kind of result, and identify such a principle of logic as, at best, a generally agreed upon, generally followed rule, whether as a practical constraint on communication or as the result of the long cognitive development of an organism that would be quickly eliminated in virtually any competitive environmental niche in which it didn’t adopt some set of rules, including consistency.57 Others reject the very notion of logical necessity, as well as other “modernist” conceptions of truth and/or objectivity. Later I will argue that the latter have misinterpreted either Kant’s strategy, or his results, or both; here I wish to look briefly at two versions of the former view, those who accept some kind of necessity but solely on the basis of historical and inductive arguments, to see that they respond solely to the question of fact, not the question of justification. The anthropologist Norman Brown has identified hundreds of 57. There is no in-principle reason to see a conflict between the kinds of results for which some evolutionary epistemologists have argued and Kant’s results vis-à-vis the status of logical rules. Indeed, given some of the remarks Kant makes in the Critique of Judgment, he might welcome a neo-Darwinian account as providing some degree of empirical or inductive support for the theoretical results of the First Critique. The difference, of course, returns to Kant’s distinction between the quaestio juris and the quaestio facti; an evolutionary account for the a priori status of certain concepts and rules doesn’t respond to the question of justification, and thus Konrad Lorenz simply misses the point in his interesting attempt to “ground” the a priori status of space, time, and the pure concepts of the pure understanding in an organism’s biological constitution and competitive strategies. See K. Lorenz, “Kant’s Doctrine of the A Priori in the Light of Contemporary Biology,” in General Systems Yearbook, ed. L. Bertalanffy and A. Rapoport, vol. 7 (Ann Arbor, Mich.: The System, 1962), 23–35. Similarly, Donald Campbell (“Evolutionary Epistemology,” in The Philosophy of Karl Popper, ed. P. Schilpp [LaSalle, Ill.: Open Court Publishing, 1974], 442) writes that Kant “proposes a theory which had the unavoidable consequence that our quest for knowledge must necessarily succeed.” This fails to recognize that for Kant any empirical cognitive judgment (i.e., not a judgment of pure mathematics) is contingent and revisable; Kant’s point is rather that a set of rules must be in place—presupposed—in order to make any possible judgment, both those that succeed and those that fail. Indeed, without those rules, we would not be able to recognize a success or a failure.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

126   The Metaphysical Deduction “universals,” a “trait or complex present in all individuals ..... all societies, all cultures, or all languages.” As Brown himself notes, claims asserting the existence of universals “are hypotheses or arguments based on various limited kinds of evidence.”58 Brown’s remark here, as well as his reliance on a theoretical framework deeply indebted to biology and evolutionary psychology, makes clear that his attempt can result only in what we have seen Kant call “comparative universality.” While Brown’s arguments and results may buttress support for Kant’s conclusions, or at least make the very attribution of universals to homo sapiens more palatable, we can see quite clearly that he responds solely to the question of fact. Consistent confirmation of universal traits among human beings at best provides a strong inductive argument, but cannot provide the universality and necessity required by Kant’s conception of logic. Rather, a conceptual analysis must be provided, in order to show that thought itself is not possible without conformity to some rule or set of rules, as I tried to indicate, in various ways, above. Another writer who has made the attribution of such universals more respectable is the linguist Noam Chomsky. In addition to doing technical work in linguistics, as well as his still more controversial work in political criticism, Chomsky has explored some of the philosophical consequences that he takes his theoretical linguistic work to entail. In very general terms, these consequences include a commitment to innatism, a sharp critique of behaviorism, and a rationalist model of the human mind that opens up conceptual space for a strong sense of freedom, something Chomsky argues cannot be satisfied on any traditional empiricist model.59 It is a rationalist model that Chomsky largely identifies with Descartes and the Port-Royal philosophers, and he has explored the historical roots of his own linguistic work most extensively in his Cartesian Linguistics. 58. Donald Brown, Human Universals (Philadelphia: Temple University Press, 1991), 42, 51. To be sure, Brown distinguishes a number of ways of providing a characterization of what a “universal” is and how one should distinguish among them, such as Chomsky’s distinction between “substantive” and “formal” universals, and Greenberg’s distinction between “conditional” and “non-conditional” universals. 59. As Stephen Pinker puts it, “radical empiricism is not necessarily a progressive, humanitarian doctrine. A blank slate is a dictator’s dream.” The Language Instinct (New York: Morrow, 1994), 427.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Metaphysical Deduction   127 Here again we see the distinction between a (roughly) empirical approach to the question compared to the transcendental strategy Kant employs. For instance, Chomsky notes that “[w]e attain knowledge when the ‘inward ideas of the mind itself ’ and the structures it creates conform to the nature of things.”60 He thus resists the transcendentally idealistic dimension of Kant’s thought, the so-called Copernican Revolution. For Kant, in speaking of the conformity of things and judgments about those things, those rules that are imposed by such conformity are contributed by the mind—indeed, this is how Kant argues for their status as necessary. In this way, Chomsky is committed to the same general metaphysical picture we found in the Modistae, whose work, it should be pointed out, Chomsky does not include in his account of the historical background of linguistics. Thus Chomsky blurs, or even ignores, Kant’s important distinction between the question of legitimacy and the question of fact by talking of origins and giving, in principle, a biological response to a philosophical question. Kant seeks the justification of the necessary conditions of possible thought and possible experience (as well as duty), and Chomsky’s tendency to talk in terms of hard-wiring and neuroanatomy would be considered by Kant to provide a physiological answer that does not address the distinct philosophical issue of legitimacy and warrant. For related reasons, Kant is suspicious of accounts, such as Descartes’s, that rely heavily on innate ideas, although it must be added that Kant’s own contrasting account is extraordinarily difficult, relying on an obscure application of the biological doctrine of epigenesis (cf. B167). Chomsky himself has agreed that his work in linguistics can, in some basic sense, be seen as doing what Kant was doing, noting that “I haven’t myself referred specifically to Kant very often, but rather to the seventeenth-century tradition of the continental Cartesians and the British Neoplatonists, who developed many ideas that are now much more familiar in the writings of Kant.”61 That Chomsky does not regard the fundamental and relevant dif60. N. Chomsky, Reflections on Language (New York: Pantheon, 1975), 8. 61. Brian Magee, Men of Ideas (London: BBC, 1978), 221. As Chomsky does point out in Cartesian Linguistics (New York: Harper and Row, 1966), 73, Kant’s views are not given the attention a fuller study would need to devote to them.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

128   The Metaphysical Deduction ferences that exist between that seventeenth-century tradition and Kant’s revolutionary Critical approach as significant makes clear that considerably more work remains to be done in exploring the philosophical and conceptual foundations of Chomsky’s approach to linguistics and its historical background. While Brown and Chomsky both recognize and argue for attributing some universal set of concepts or rules to which all agents must be committed, Richard Mason has argued that many modern treatments of logic and its attendant analysis fail to take into consideration the historical choices that come “before logic.” Mason wishes to reject as misleading the crude alternatives of taking logic either to be embodied in a mutable and contingent language and culture or to be a set of rules that essentially characterize the world, “which have nothing to do with human choice.”62 For Mason, any given treatment of logic always indicates a choice, implicitly informed by a social, historical, linguistic, and religious context; he thus sees Kant’s conception of logic—the “transcendental hypothetical: ‘if anyone is to think or judge, then they must think like this .....’—as itself informed by Kant’s own philosophical concerns.” Following Heidegger, Mason regards such a transcendental approach to logic as transforming a radically subjectivist epistemology, grounded in my thoughts and judgments, into “nobody’s impersonal thoughts, judgments, or propositions.”63 Earlier I appealed to Manley Thompson’s use of a “neo-Tractarian or neo-Kantian way of speaking about logic,” and here I think Mason has missed the point of both Kant’s account of general logic and the 62. Richard Mason, Before Logic (Albany: SUNY Press, 2000), 120. 63. Ibid., 121–22. Mason here neglects the strategy I’ve outlined in the preceding chapters, where a set of rules is grounded in a radically subjective “I think”; on the assumption that any agent has that capacity, that set of rules can then also be seen as objective. Hence “nobody’s” judgments would become, in this context, everybody’s judgments, or at least would have to satisfy some set of universal and necessary conditions to be judgments. Mason remarks that “Kant seemed sure enough that the self was engaged both in moral choice and in logical reflection or calculation” (123), and I think he is certainly correct about that. I would simply challenge the notion of calling an entity who failed to do either, or especially both, a “self.” Mason also sees Kant’s project as motivated, in part, by “an alleged possibility that ‘I’ might be wholly or importantly mistaken in my thought” (121), but fails to make clear at all either what it would mean for someone to be wholly mistaken about “I” in its thought or whether we would attribute agency to such an entity.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Metaphysical Deduction   129 results of the metaphysical deduction when he writes “Kant distinguished logic in a narrow, calculating sense from ‘reason’ in his own wide sense.”64 While Kant no doubt does distinguish logic from reason (Vernunft), we have seen repeatedly that by “logic”—specifically general logic—Kant has in mind a set of rules for the possibility of thought simpliciter. Thus when Mason speaks of “choices” that precede logic, this neo-Kantian conception of logic would insist that the very notion of a “choice” presupposes alternatives, and thus the rejection (or negation) of one of them. In this sense, then, general logic provides a reflectively articulated set of constraints on rationality, in that one who rejects—or, better, fails to follow—that set of constraints could not be said to rational, and Mason obscures this point by contrasting his own account of what we must consider “before logic” with what we must consider “before reason” or “before rationality.”65 In short, a thoroughgoing process of transcendental reflection will result in the recognition that some set of constraints must hold for thought to be possible, and Kant suggests that the principle of contradiction is the most plausible candidate for membership in this set. I have tried to argue that any attempt to reject that principle, or any attempt to relegate its status to a solely historical or pragmatic convention or inductively based generalization, will itself be forced to make use of it and cannot do otherwise; in this way, we may wish to say with Wittgenstein that our spade is turned, or we may simply recognize that a minimal constraint on possible thought holds universally and necessarily without an alternative, and in that crucial sense is justified. Much earlier I suggested that, in his own tortuous way, Kant sought to provide a systematic and defensible account of what he regarded as, 64. Ibid., 113. 65. Ibid., 113. While one might, again, object to the following as “logic chopping,” it should be noted that Mason’s claim that an account of rationality would be “unequivocally historical, sociological, or anthropological” (113) assumes far more than Kant is willing to concede, and which he would in any case regard as empirical and contingent issues, beginning with identifying the proper subject-matter of these disciplines. It is also worth noting that Mason himself rejects various analyses as “question-begging” and “circular,” criticisms of positions that clearly presuppose the kinds of rules Kant has in mind in articulating his conception of logic. Mason’s appeal to history presupposes a rich semantic conception of time and thus, again, assumes precisely what Kant seeks to justify.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

130   The Metaphysical Deduction

Copyright © 2008. Catholic University of America Press. All rights reserved.

at bottom, “common sense,” and I think Kant’s approach here may best be seen in light of his reading of Rousseau.66 As is well known, Rousseau’s work has come to be recognized as a fundamental influence on Kant; I want to argue here, very briefly, that we underestimate that influence if we limit it to the realm of the practical or moral philosophy, for I think it is clear that Rousseau can also be seen as providing a crucial strategic insight into the nature of philosophical argument in general.67 Kant regarded Rousseau as the “Newton of the moral universe,” and in an oft-cited passage, praises Rousseau for having set him straight, transforming him from one who despised humanity to one who honored it. Kant reveals a similar attitude when he observes that “[a]n academy of sciences in Paris concocts 1000 times more errors than a village full of farmers.” As we have seen, Kant frequently utiliz66. Jean Ferrari, in Les sources françaises de la Philosophie de Kant (Paris: Librarie Klincksieck, 1979), 173, argues that Kant would have been familiar with Rousseau’s Discourse on the Sciences and Arts, the Discourse on the Origins and Foundations of Inequality, The Social Contract, Émile, The New Heloise, (probably) The Confessions, and possibly even the writings on music. One may discount various stories, such as Kant failing to take his regular walk due to his preoccupation with Émile (see M. Rischmiller, ed., Bemerkungen in den “Beobachtungen über das Gefühl der Schönen und Erhabenen” [Hamburg: Kant-Forschungen (vol. 3), 1991], 150), and the fact that a portrait of Rousseau was the sole picture hanging in Kant’s house, and still recognize the importance of Rousseau for Kant, particularly in the crucial decade of the 1760s. In his recent, and excellent, Kant: A Biography (Cambridge: Cambridge University Press, 2001), Manfred Kuehn takes issue with those who emphasize the salient influence of Rousseau on Kant. 67. An enormous amount has now been written on the relationship between Rousseau and Kant, although virtually all of it focuses exclusively on the moral philosophy. The key recent figure who has insisted on the importance of Rousseau for Kant’s philosophy has been, undoubtedly, John Rawls, particularly in his lectures (e.g., Lectures on the History of Moral Philosophy, ed. Barbara Herman [Cambridge, Mass.: Harvard University Press, 2000]). See also Richard Velkley, Freedom and the End of Reason (Chicago: University of Chicago Press, 1989), and V. Delbos, “Rousseau et Kant,” in J. J. Rousseau—Leçons faites à l’Ecole des Hautes Études Sociales (Paris: Alcan, 1912). Ernst Cassirer’s “Das Problem J.-J. Rousseau” in Archiv für Geschichte der Philosophie 41 (1932): 177–213, 479–513, was the first significant text in emphasizing the importance of Rousseau’s work for Kant. Klaus Reich’s Rousseau und Kant (Tübingen: J. C. B. Mohr [Paul Siebeck], 1936), which was generally overlooked on its publication, is the only discussion I’m familiar with that at least suggests that Rousseau’s influence on Kant extended beyond the practical philosophy; I have translated this piece in the Graduate Faculty Philosophy Journal 23, no. 2. See also Dieter Sturma, Jean-Jacques Rousseau (Munich: C. H. Beck, 2001).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Metaphysical Deduction   131 es a specific conception of grammar to clarify his account of the rules of logic and their status; thus it is not surprising to appeal to Molière’s “Le Bourgeois Gentilhomme,” who discovers to his astonishment that he has spoken prose—and as Kant notes, has spoken grammatically—his entire life.68 Kant’s lectures on logic, as well as his own unpublished notes, repeatedly refer to the general or common healthy understanding; the point seems to be that any agent is at least potentially able to reflect on its capacities to think, to judge, and to act. There is no guarantee of being able to delineate—or justify—the rules that range over the relevant domains, but anyone who is sufficiently insightful (Scharfsinnig) should be able to recognize that there are conditions that must be satisfied in order for us to think, judge, and act in the relevant ways. To be sure, on Kant’s view most people do not have in mind a conscious, articulated set of logical rules, let alone a formal system of logic; however, they do all think (or ought to think) in accordance with some modest set of rules, including the principle of contradiction, for otherwise we would not be able to recognize them as thinking. As we will see in the following chapter, such a procedure has some relevance for Donald Davidson’s account of intersubjective communication and what he calls “triangulation.” Yet here the question arises again: even if we are willing to accept the notion that some rules must be followed for judgments relative to a given domain to be possible, is this sufficient to establish the strong sense of universality and necessity required by Kant’s account? And even if we are willing to affirm a commitment to such universality as establishing a priori conditions on, say, thought, how can those conditions be identified? The key strategic move here is precisely captured by Rousseau in The Social Contract: that the “obedience to the law one prescribes to oneself is freedom.”69 As Rousseau argues in a number of texts, to be forced to obey a law externally imposed by another is slavery; selflegislation obviously does not abandon the notion of law, and the com68. The phrase about “honoring humanity” is from Kant’s “Remarks on the Observations on the Feeling of the Beautiful and the Sublime,” Ak. XX, 44; the cited remark is from the Blomberg Logik, Ak. XXIV.1, 22–23; the reference to Molière is Reflection 1620 (Ak. XVI, 39). 69. “L’obéissance à la loi qu’on s’est prescritte est liberté”; J.-J. Rousseau, Oeuvres complètes, ed. B. Gagnebin and M. Raymond (Paris: Gallimard [Pléiade], 1964), 3:365.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

132   The Metaphysical Deduction

Copyright © 2008. Catholic University of America Press. All rights reserved.

pulsion it brings with it. Rather, self-legislation provides a way of recognizing precisely that compulsion while, at the same time, maintaining a commitment to freedom. As Klaus Reich has noted, while the philosophical pedigree of self-legislation is quite a long one, and can be found in Aristotle, St. Paul, Polybius, Cicero, and others, it is Rousseau who seems to crystallize for Kant a crucial element of the Critical philosophy. Indeed, Rousseau’s emphasis on self-discovery as a means of determining for oneself precisely those laws that one imposes on oneself is a central theme, if not the Leitmotif, of Émile. Thus Rousseau remarks that “an hour of work will teach him [the student] more things than he will retain from a day of explanations.”70 As Cassirer characterizes this approach, “there is only one living source for this knowledge—the source of self-knowledge and genuine self-examination. And it is to this alone that Rousseau appeals; from it he seeks to derive all proofs of his principles and hypotheses.”71 In Kant’s hands, this becomes the strategy whereby conditions discovered through individual (or subjective) reflection reveal universal and necessary (or objective) conditions. Yet it should be clear that this is precisely the strategy Kant follows, not just in the practical philosophy, but also in his epistemology, metaphysics, and the philosophy of art.72 For Kant, “all necessity, without exception, lies in a transcendental condition” (A106), and the term “transcendental” refers to “the possibility of cognition or its a priori employment” (A56=B80). Thus through the process of philo70. Oeuvres complètes, 4:456. 71. Cassirer, The Problem of Jean-Jacques Rousseau, trans. Peter Gay (Bloomington: Indiana University Press, 1967), 50. 72. As Cassirer has observed relative to Goethe and Schiller, “for neither did ‘freedom’ and ‘necessity’ mean mutually exclusive opposites. They found between the two ideas a relation of correlation, not of opposition. This correlation Kant had revealed in the moral realm, explaining that ethical freedom is identical with ‘autonomy,’ with selfimposed law. The classicism of Goethe and Schiller carried this view into art: it was rooted in the principle that only law can give us freedom. Here for both the circle of the ‘subjective’ and the ‘objective’ was closed.” “Goethe and the Kantian Philosophy” in Rousseau, Kant, Goethe, trans. James Gutmann, Paul Oskar Kristeller, and John Herman Randall Jr. (New York: Harper and Row, 1963), 91. As Lewis White Beck points out, “Kant saw in mathematics a clue to the objectivity of all a priori knowledge, both analytic and what he considered to be synthetic .,... even empirical objects are constructions, and their necessary conditions are geometrical.” “Can Kant’s Synthetic Judgments Be Made Analytic?” in R. Wolff (ed.), Kant (Garden City: Doubleday, 1967), 22.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Metaphysical Deduction   133 sophical reflection, the agent comes to discover in herself (or at least has the capacity [Vermögen] to discover in herself) the necessary rules she must employ in making judgments, whether those judgments be empirical, mathematical, moral, aesthetic, or teleological. This agent, then, by reflecting on a specified domain, can come to identify a rule or set of rules that must hold in order to be able to make judgments relative to this domain, and that set of rules is thus able to be identified as a logic, presupposed or in place a priori as a set of minimal necessary conditions for the possibility of judgment, and in that sense such a logic can be said to be transcendental. Here we see the strength of the analogy Kant draws between general and transcendental logic. If we make the relatively modest assumption that we think, we can come to recognize, reflectively, that that thought must occur in accordance with certain rules; here, of course, Kant specifies the table of judgments and the amalgam of Aristotelian and Stoic logic and their further developments from the PortRoyal school on through Wolff, Meier, and Baumgarten. The point is that for thought to be possible, it must satisfy certain (minimal) conditions, and as I’ve tried to indicate, there are really no alternatives to the assumption at stake.73 This lends considerable plausibility to Kant’s strategy in the Transcendental Analytic, to find a set of rules that plays an analogous role, relative now to a passively received domain of sensibility. If we are to make coherent judgments relative to that domain— or if we assume that we have possible experience—we can come to recognize, reflectively, that such judgments must occur in accordance with certain rules. Thus in the Second Analogy, Kant can draw on this analogy to argue that for experience to be possible, it must incorporate some causal conception of an event, in order to distinguish it from a state of 73. Some might suggest that a particularly radical skepticism might remain untouched by this argument; as Laurence BonJour has pointed out, “There are certain versions of skepticism which are so deep and thoroughgoing that it is impossible to refute them,” The Structure of Empirical Knowledge (Cambridge, Mass.: Harvard University Press, 1986), 194. But it is not clear (a) that such a position can be formulated, and (b) what its interest would be, if it were. Once it is put forth as some kind of meaningful utterance, it would seem to be committed to at least the kinds of conditions under discussion; and if it is not formulated in such a way, it doesn’t seem that it could engage us in any philosophically significant way.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

134   The Metaphysical Deduction affairs. For his argument to be successful, he must then show that for a subject that judges using concepts, the very notion of an event that cannot be placed within a causal nexus is impossible.74 The Second Analogy, then, would provide a rule for empirical inquiry, qua logic, but no more determines the truth-value of a specific causal claim than general logic determines the truth-value of a judgment. To be sure, two claims that are explicitly contradictory may be rejected, as can invalid arguments, on the basis of general logic; but beyond such negative criteria, general logic is not in a position to determine the truth (or falsity) of specific claims. Similarly, a judgment about an event that fails to abide by the minimal constraints imposed by the Second Analogy will be rejected, but beyond that negative criterion, the determination of a judgment, such as one indicating a causal sequence, is not provided by transcendental logic. Among other things, Kant’s critics have argued that the Second Analogy seeks to secure the uniformity of nature, or give warrant to principles of inductive reasoning; they can then turn around easily enough and demonstrate how the Second Analogy fails to do so. But if we keep in mind that transcendental logic is a logic, we see that it is in no position to establish such results. Indeed, Kant takes the causal principle only to have proved that “all alterations take place in accordance with [nach] the law of the connection of cause and effect” (B232). Thus, given an event E, we can place it within a causal nexus, and know a priori that E was caused by an antecedent event. Undoubtedly, it is a principle developed from a quite immoderate claim: the concept of cause “makes strict demand that something A, be such that something else B follows from it necessarily and in accordance with an absolutely universal 74. Post-Newtonian developments in physics—both quantum and relativity theory—have been seen by many to call into question the causal principle argued for in the Second Analogy. An attempt to defend Kant against this objection raises a number of technical points that go well beyond the scope of this study. But one might attempt to begin such a defense by seeing a quantum account of an event as introducing probability functions to replace the more discrete Newtonian conception of an event, while still maintaining a temporally ordered causal sequence between event “smears.” While relativity theory perhaps raises even more difficulties for the Second Analogy, causal claims can still be regarded as maintaining their coherence, although now relative to an inertial frame of reference, and one can compare and evaluate causal claims by making the appropriate Lorentz transformations between those reference frames.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

The Metaphysical Deduction   135 rule (A91=B124; Kant’s emphasis). While strong enough (if successful) to stave off Hume’s attack on the causal principle, it is at the same time a modest claim, particularly given the historical context within which Kant writes and the subsequent claims made about what the Second Analogy establishes. As Kant freely admits, “How anything in general can be altered, and how it is possible that upon a state in one point of time an opposite one may follow in the next—of this we have a priori not the least conception. For this we require an acquaintance of actual forces, which can only be given empirically” (A206–7=B252; my emphasis). If we see a baseball hurled toward a window, which is then shattered in a resounding crash and shower of glass fragments, we may feel quite confident in asserting that the baseball broke the window. But if we are met with a compelling assertion that denies our claim, the truth of either judgment must be determined empirically; its truth is not delivered by a pronouncement forthcoming from Transcendental Analytic; indeed, the function of the Critique as a whole is in large part the securing of a method to go about inquiring after the truth of claims, and the securing of the limits to that inquiry. In sum, we have under consideration a specific kind of subject, or agent, who satisfies the following conditions: it can employ the firstperson pronoun, it can judge using concepts, and it can regard itself as free. As we have seen, Kant learned from Rousseau to “honor” humanity, and after what Karl Ameriks calls Kant’s “Rousseau experience,” it is clear that Kant attributes these abilities to the healthy general understanding of human beings. As Ameriks notes, “Kant’s philosophy has an unusual openness to a wide variety of considerations and a special respect for universal common sense.”75 In similar fashion, Dieter Hen75. Ameriks, Kant and the Fate of Autonomy, 67, 38. As noted earlier, there is no inprinciple limitation of such agency to human beings, although many of Kant’s remarks indicate that that is the model he employs. Many have criticized Kant for having restricted agency to men, and thus constructing a conception of human reason that is sexist and oppressive. As I have argued in “Kant and Feminism,” many of Kant’s most noxious remarks, including those about women, appear in texts such as the Lectures on Anthropology, not in the systematic works of the Critical philosophy, and that while Kant is deserving of strong censure for many of those remarks, he may well provide in those systematic works the grounds for a strong, progressive, and liberatory feminism, albeit along the lines of an Enlightenment feminism (as opposed to a competing “Standpoint” feminism).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

136   The Metaphysical Deduction rich has argued, “One could show that a clear connection exists between Kant’s claim that philosophy is based on natural reflection and his affiliation to Rousseau, in whose work the ordinary man in a sense knows everything from the very beginning.”76 While he may not be sanguine about all agents being able to articulate and defend a systematic treatment of logic qua logica docens, all such agents—to function as agents—will generally think, judge, and act in accordance with a set of rules, and have the normative obligation to eliminate those thoughts, judgments, and acts that conflict with those rules. To maintain a contradictory set of beliefs or to insist, for instance, on a non-causally ordered set of judgments is to fail one’s intellectual obligations, although we may have more success, on Kant’s view, in eliminating errors of general logic, while we are fated to continue to embrace transcendental illusions, and must return to the discipline that exposes such dialectical errors in our reasoning. Indeed, it is likely that anyone to whom we are willing to attribute agency thinks, judges, and acts in accordance with a wide variety of concepts, rules, conventions, and heuristics that go well beyond Kant’s own modest list. I think we can now see why Kant would take it as obvious that, given that possible thought cannot fail to follow certain rules—some version of general logic—and still be plausibly characterizable as thought, possible experience analogously must follow certain rules—a logic of possible experience—without which it would not, likewise, be characterizable as experience. To be sure, one can argue about the details and specifics of the concepts, principles, and rules Kant identifies as absolutely universal and necessary; indeed, those arguments continue today. But it seems at this level of analysis to be disingenuous to deny that some minimal constraints on thought, judgment—and for Kant, ultimately and most importantly action—must hold. As Ameriks quotes Franz von Herbert, “experience is ..... presupposed, postulated, or however one calls it. Now if a skeptic were dumb and shameless enough to say ‘But is there experience?’ there is really no answer for such a type other than a beating.’”77 76. D. Henrich, “Kant’s Notion of a Deduction and the Methodological Background of the First Critique,” in Kant’s Transcendental Deductions, ed. E. Förster (Stanford: Stanford University Press, 1989), 252n5. 77. Ameriks, Kant and the Fate of Autonomy, 65.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Ch a pter Five

Kant and Contemporary Philosophy

Copyright © 2008. Catholic University of America Press. All rights reserved.

e

If we construe Kant’s general logic in the way I have urged in previous chapters, we see emerging a conception of logic that functions as a minimal-constraint model of rationality. In short, for an agent to be regarded as an agent—or for ourselves to regard ourselves as agents—we must conform to some set of rules for thought to qualify as thought. In Kantian terms, for a judgment in general to be possible, it must conform to a set of rules, and I have focused here on the principle of noncontradiction as exemplary of such a rule. When we move to experience, we again see that for judgments—now ranging over the domain of a passively received sensible manifold—to be possible, they must conform analogously to such a set of rules. While it may seem to be straining at gnats to spend much time arguing for such a rule as the principle of non-contradiction as “justified” in some sense, it is precisely by showing the indispensability of such a principle that the strength of Kant’s approach in transcendental logic is revealed. Informally, to reject the epistemic force given to transcendental logic in virtue of its analogy with general logic is to put forth a conception of possible experience

137

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

138   Kant and Contemporary Philosophy that is sufficiently problematic that it would call into question the very idea of “experience.” Such judgments would not involve relatively enduring substances (the First Analogy), would lack any account of causal interaction, thus not allowing us to distinguish events from states of affairs (the Second Analogy), and would not recognize reciprocal causal interaction among such events and states of affairs (the Third Analogy). Kant’s point, of course, is that if it makes no sense to call a sentence that is explicitly self-contradictory a “thought,” then analogously it makes no sense to call a sentence that explicitly violates the rules that make such judgments possible a “judgment of possible experience.” Furthermore, while a sentence that satisfies these minimal syntactic conditions may be said to be well-formed, general logic tells us nothing—indeed, is in no position to offer anything—beyond the indication that it does satisfy such criteria. Similarly, while a judgment of possible experience that satisfies the minimal syntactic conditions relevant to such judgments—now ranging over a sensible manifold—can be seen as, in its own way, well-formed, transcendental logic tells us nothing about its truth-value. In this way, transcendental logic seeks to establish a framework within which judgments of possible experience may be investigated, or within which judgments may be regarded as truth-evaluable. Neil Tennant has noted, “One might break some logical rules some of the time, and get away with it. One might even break every logical rule some of the time, and get away with it. But what one cannot do is break some of the rules all the time, or all of the rules for any significant amount of time.”1 On Kant’s conception of logic as a set of rules that, on reflection, yield a set of universal and necessary rules that range over a given domain—albeit normatively—then Tennant’s characterization would serve to indicate Kant’s strategy. Without some set of necessary constraints on thought, thought is impossible; without some set of necessary constraints on judgments relative to sensibility, experience is impossible. In sum, an entity that indicated no thoughts and (conceivably) no cognitive interaction with its surroundings would not qualify as an agent, and certainly not as rational.2 1. Neil Tennant, “Logic and Its Place in Nature,” in Kant and Contemporary Epistemology, ed. Paolo Parrini (Dordrecht: Kluwer, 1994), 103. 2. Kant’s general motivation should be kept in mind here, as well; while the theo-

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant and Contemporary Philosophy   139 Having argued that the central thread running throughout the Critique of Pure Reason is the notion that Kant’s strategy there is explicitly logical, in the sense described above, I want to draw on these results to indicate their relevance for some issues in contemporary philosophy: Laurence BonJour’s defense of the a priori, Donald Davidson’s critique of the scheme-content distinction, and the general position adopted by those writers who have been—perhaps unfairly—labeled as “postmodernists.” In so doing, I hope to show that results of Kant’s strategy as argued for above not only are still largely defensible, but retain a good bit of relevance for approaches to current philosophical controversies.

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Laurence BonJour’s “Moderate Rationalism” In his 1985 The Structure of Empirical Knowledge, Laurence BonJour presented a compelling and articulate defense of a coherence theory of knowledge.3 Following what he called a “dialectical” strategy, he began by indicating the central issue at stake: the justification of empirical knowledge claims. He then argued that no available foundationalist or coherentist account could provide that justification, and that all such attempts either end in sheer dogmatism or succumb to skepticism. After a lengthy critical discussion, he turned to developing an argument for his own view, combining a correspondence theory of truth with a coherence theory of justification. He further suggested that such a view, to be adequate, must include a commitment to the a priori—indeed, the synthetic a priori—while recognizing that this could not be described as a pure coherence theory. However, this suggestion was relegated to an appendix, and put forward rather gingerly by claiming only that skepticism about the synthetic a priori was “by no means as intellectually mandatory as it is often thought to be” (SEK, 211). retical philosophy functions as the ratio cognescendi, the practical philosophy functions as the ratio essendi for philosophy in general. Clearly we would not be able to attribute moral notions, such as responsibility or duty, to an entity that didn’t satisfy such minimal constraints on rationality. 3. Laurence BonJour, The Structure of Empirical Knowledge (Cambridge, Mass.: Harvard University Press, 1985). All further references to this text are hereafter indicated by SEK, and given parenthetically.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

140   Kant and Contemporary Philosophy In a subsequent series of papers, which form the basis of his recent In Defense of Pure Reason, BonJour has pursued this aspect of his argument much further.4 Again following a “dialectical” approach, he criticizes at length both moderate empiricism, represented by Hume and Kant among others, and the “radical” empiricism of Quine, and then defends a view he calls “moderate rationalism,” which includes as an essential component a commitment to the a priori, thus offering a defense for the rational, albeit fallible, insight into necessity that the other approaches cannot provide. The epistemic stakes are quite high here, and BonJour concludes that any epistemological view that fails to include the synthetic a priori cannot succeed, generating skepticism of greater or lesser virulence, and thus risking “intellectual suicide” (5; 99), the loss of “cognitive sanity” (128), “irrationality and intellectual chaos” (138), and “giving up rational thought altogether” (152). Any discussion of the synthetic a priori obviously must include an examination of the views of Kant, who not only was the first to claim that synthetic judgments could be a priori, but took the questions of whether such judgments could occur, and if so how, to be central to his revolutionary approach to metaphysics in the Critique of Pure Reason. BonJour recognizes this need to take Kant’s account into consideration, and devotes a section of the first chapter of In Defense of Pure Reason to Kant’s views; furthermore, no philosopher is mentioned in the text more than Kant, with the (possible) exception of Quine. Unfortunately, BonJour so badly misunderstands Kant’s philosophical view that his critique of it cannot be accepted. What makes this something more than a relatively idle exercise in the history of philosophy is that his rejection of Kant’s theoretical philosophy forces BonJour to abandon his most powerful ally in defending the very views he puts forth. Elsewhere, I have argued that (1) BonJour and Kant largely agree on what an adequate epistemology must look like in its outlines, namely, the combination of a correspondence theory of truth, a co4. Laurence BonJour, In Defense of Pure Reason (Cambridge: Cambridge University Press, 1998); all parenthetical citations, unless otherwise noted, are to this text. The earlier papers are “Toward a Moderate Rationalism,” Philosophical Topics 23 (1995): 47–78; “Against Naturalized Epistemology,” Midwest Studies in Philosophy 19 (1994): 283–300; and “A Rationalist Manifesto,” Canadian Journal of Philosophy supp. 18 (1992): 53–88.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   141 herence theory of justification, and a commitment to the synthetic a priori; (2) BonJour’s dialectical strategy fails to persuade because of a lack of positive arguments for his own position; and (3) for sound epistemological reasons, BonJour should find the kind of argumentative strategies Kant deploys considerably more attractive than he does.5 Here I want to focus on the specifics of BonJour’s reading of Kant’s views, particularly those found in the Critique of Pure Reason, to provide further evidence for the claim that, on the basis of a misguided interpretation of Kant’s text, BonJour fails to embrace precisely the strategic option that would best support his own argument where it is at its weakest. As I hope to show, BonJour’s failure to identify correctly Kant’s strategy as grounded in a conception of logic leads to this misinterpretation. BonJour’s interpretation of Kant’s theoretical philosophy can be summarized fairly briefly, although some of its details will be brought out below. In general, he sees Kant’s view as representative of what is described as “moderate empiricism,” a position that incorporates “empiricist skepticism” about the a priori, and claims that “a priori justification concerns only ‘relations of our ideas’ as opposed to ‘matters of fact’” (17).6 While most associate this doctrine with Hume, BonJour argues that Kant is “in fact much closer to Hume” than he is to rationalism, and that Kant’s views are closer to Hume’s than to either Locke’s or Berkeley’s; indeed, this skepticism about the a priori, which has become so dominant in Anglo-American philosophy, largely derives from Hume and Kant (17). Furthermore, for Kant, “the mind so shapes or structures experience as to make the synthetic a priori propositions in question invariably come out true within the experiential realm” (23; BonJour’s emphasis). From this, and because Kant’s synthetic a priori claims do not pertain to an sich reality, “it appears to be for him [Kant] self-evident that we can have no a priori knowledge of independent reality except that which is analytic and ultimately trivial” (25). The synthetic concepts and principles Kant seeks to justify as a priori reduce to analytic trivialities, that “necessarily falsify reali5. “BonJour, Kant, and the A Priori,” Disputatio no. 7 (1999). 6. BonJour, as do many, uses “a priori” adjectivally, although as we have seen, Kant uses the term adverbially.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

142   Kant and Contemporary Philosophy ty” (152n15; BonJour’s emphasis), and the Kantian view “concedes that the skeptic is right, not only about induction but about knowledge of the real world generally, and then proceeds to offer us a pale substitute for the knowledge thus abandoned” (200n18). In sum, “if Kant had ever faced clearly the problem of the epistemological status of his own philosophical claims, he might have retreated into a more traditional rationalism” (25); since he did not do so, his project collapses into a uneasy mixture of dogmatism and phenomenalism, embracing rather than defeating skepticism, and thus utterly failing to offer any defense for the rational insight into necessity that the a priori must provide. Two general themes emerge from this critique: an objection to Kant’s metaphysics and a more explicitly epistemological concern with the a priori and its role in the justification of knowledge claims. BonJour is on much safer ground in putting forth the former, for the ontological claims that result from the perspectival shift in Kant’s “Copernican Revolution” have generated the greatest controversy of any of Kant’s doctrines, including reactions from Hegel, Schopenhauer, Nietzsche, Peirce, and Rorty, among many others. The basic complaint is made succinctly by Hegel: “Thoughts, according to Kant, although universal and necessary determinations, are only our thoughts—separated by an impassable gulf from what the thing in-itself is.”7 On this view, Kant’s is committed to an ontology of two kinds of objects: objects as they appear to human beings, and objects as they are “in themselves,” or what BonJour refers to as “an sich reality.” The obvious problem that results is determining the relationship between the two, a problem that becomes insoluble on Kant’s insistence that human beings, in principle, can never know the latter. Thus human beings are forever denied access to reality, quickly leading to skepticism due to the impossibility of bridging the impassable gulf to which Hegel refers. While this view (now generally referred to as the “two world” interpretation) of Kant’s ontology has a long history, an alternate interpretation (the “two aspect” view) has been put forth that yields several distinct advantages: having better textual support, ascribing to Kant a considerably more defensible view, explaining more plausibly chang7. G. W. F. Hegel, Encyclopedia (“Lesser”) Logic, § 41 z.2.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   143 es Kant makes in his revisions for the Critique of Pure Reason’s second edition, and, most important, not attributing a doctrine to Kant that he spends much of his effort in the First Critique rejecting. In brief, on the two-aspect reading, a given epistemic subject has two ways of regarding objects, or two different descriptions under which those objects can be taken; only one such description has any cognitive relevance for a being who makes judgments relative to a sensible manifold that is itself passively received within the subjectively contributed forms of time and/or space (e.g., a human being). On this reading, any references to the “thing in itself,” or the constituents of an sich reality, are cognitively vacuous, and in any case parasitic upon the objects of possible experience, the objects that satisfy those conditions necessary for the epistemic subject to have experience and make truth-evaluable judgments about them. As Arthur Melnick has written, “[T]he notion of a thing in itself is that concept of an object according to which being an object or a thing would make sense in abstraction from any idea of a (type of subject) and his intellectual organization of his experience.”8 If, in contrast, we initially take “objects” or “the real” to refer in some sense to objects as they “really” are, and then ask how our objects correspond to them, we embrace transcendental illusion, simply making it seem as if we were asking a sensible cognitive question. This is precisely what Kant is getting at when he tells us that the “true correlate of sensibility, the thing in itself, is not known, and cannot be known ..... and in experience no question is ever asked of it” (A30=B45; my emphasis; cf. A494=B522).9 By adopting without discussion the two-world interpretation, BonJour is able to show without much trouble that what one might call Kant’s epistemology is inadequate, and thus is forced to yield to skepticism about knowledge of the real world in general. Unfortunately, he does so by imposing a burden on Kant’s account that it need not 8. Melnick, Kant’s Analogies of Experience, 152; also see Bird, Kant’s Theory of Knowledge; Prauss, Erscheinung bei Kant; R. Meerbote, “The Unknowability of Things in Themselves,” in Kant’s Theory of Knowledge, ed. L. W. Beck (Dordrecht: Reidel, 1974); Allison, Kant’s Transcendental Idealism; Thompson, “Things in Themselves”; Buchdahl, “Realism and Realization in a Kantian Light.” 9. This is not to deny that the thing-in-itself plays a crucial role in Kant’s practical philosophy.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

144   Kant and Contemporary Philosophy confront, and thus need not discharge; for, on the two-aspect interpretation, there simply is no question of explaining the cognitive relationship between objects of possible experience and an sich reality. Indeed, explaining that relationship is precisely the obligation Kant argues that neither the empiricism of Locke (and Hume) nor the rationalism of Descartes and Leibniz can meet; in short, BonJour attributes to Kant, and criticizes him for being unable to defend, a view—transcendental realism—that Kant explicitly rejects. As Kant writes, it is the transcendental realist “who afterwards plays the role of empirical idealist. After wrongly supposing that objects of the senses, if they are to be external, must have an existence by themselves, and independently of the senses, he finds that, judged from this point of view, all our sensuous representations are inadequate to establish their actuality [Wirklichkeit]” (A369). BonJour argues, in spite of what Kant quite specifically denies, that he must provide an account of how the epistemic subject has cognitive access to that which has been stripped of the very conditions that would make it an object of possible experience, or an element of a possible cognitive judgment. Kant’s own view, in contrast, is that the epistemic subject contributes a set of universal and necessary conditions for the possibility of experience, and that any account of empirical judgments must recognize and justify that contribution. This in turn leads to the denial that we can have experience of the external world without that subject employing some theoretical or conceptual apparatus: a view BonJour himself exploits in criticizing at length foundationalist arguments that rely, surreptitiously or not, on “the given” (see SEK, chapter 4 passim, esp. 77–78).10 While I think it is clear that accepting the two-world interpretation of Kant’s ontology is, at best, problematic, it is only on that basis that BonJour can draw the remarkable conclusions he does relative to Kant’s account of logic, truth, justification, the a priori, and cognition. BonJour seems to find Kant offering a set of claims about our knowledge of objects that are restricted to appearances, and thus unable to 10. The point, of course, is most familiar from Wilfred Sellars’s classic “Empiricism and the Philosophy of Mind,” the influence of which BonJour’s acknowledges (SEK, 249n1) and which is itself very much in debt to Sellars’s reading of Kant.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   145 state anything about the real world of objects. On this basis, the mind constructs its own phenomenal experience; consequently, all objects of that experience will conform to the rules the mind imposes for that construction. From this, BonJour concludes that all synthetic a priori propositions are true, relative to this phenomenal experience (23), and thus are ultimately analytic, including specific principles such as the causal maxim (25n25). While he does not specifically address Kant’s own attempt to satisfy the “observation requirement” (SEK, 141–44), BonJour’s reading prevents him from seeing that Kant’s empirical realism embraces a notion quite similar to BonJour’s own, including a commitment to “extratheoretic input” couched in terms of a spatiotemporal intuition [Anschauung].11 One is tempted to see BonJour tending toward the interpretation Richard Rorty has offered, suggesting that a specifically Kantian “realist” transcendental argument must be such that “our ‘legitimating’ transcendental knowledge of the necessary truth that content will correspond to scheme, is made possible by the fact that our subjectivity (the scheme) creates the content.”12 Unfortunately, the interpretation given by Rorty (and BonJour) cannot be supported either directly or indirectly by Kant’s text. Kant’s metaphysics may well move in the direction of an anti-realism BonJour finds unattractive, as can be seen in his brief critical remarks on Dummett (150–51). There is, however, a tough-minded empirical streak in Kant that BonJour ignores, but that should not be minimized: “High towers and metaphysically great men resembling them, round both of which there is commonly much wind, are not for me. My place is the fruitful bathos of experience, and the word “transcendental” ..... does 11. An anonymous reader has correctly and helpfully pointed out that BonJour’s critique does not attribute to Kant the view here expressed by Rorty, and that BonJour’s account does not rest upon such a view. At the same time, BonJour has recognized that an adequate epistemology must take into consideration the relationship between the world and the empirical judgments made about that world, and it remains unclear on his reading how Kant addresses—or could address—the point. It should also be noted that this very question has been the source of a long controversy among Kant scholars, including Jaako Hintikka, Charles Parsons, Manley Thompson, and many others. 12. Richard Rorty, “Transcendental Arguments, Self-Reference, and Pragmatism” in Transcendental Arguments and Science, ed. P. Bieri, R. P. Horstmann, and L. Krüger (Dordrecht: Reidel, 1979), 79. As Wolfgang Carl points out in his response to Rorty (ibid., 106), this seems “to flatly contradict Kant’s own and well-known position.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

146   Kant and Contemporary Philosophy not signify something passing beyond all experience but something that indeed precedes it a priori, but that is intended to make cognition possible” (Ak. IV, 373n). Kant makes as explicit as possible, in the introduction to the Critique, the Transcendental Aesthetic, and elsewhere, that in human experience there is a fundamental and ineliminable passive component, which he calls sensibility. To be sure, the epistemic subject, on Kant’s view, contributes the forms of intuition— space and time—within which sensibility is received; yet this hardly entails that the subject constructs its own experience, or creates its own content. This recognition that there is a dimension of our experience that we confront, rather than create, is about as close as we can get to “the given” in Kant’s epistemology. One might compare here BonJour’s own insistence on an “observation requirement” in responding to the traditional “isolation” objection to pure coherentist theories of justification: namely, that certain beliefs, as “cognitively spontaneous,” indicate input from an independent source, or what he calls “extratheoretic input” (SEK, 141ff.). As we have seen, lacking what Kant calls “intellectual intuition,” we do not create content; all we can do, relative to this passive receptivity, is characterize reflectively the forms of intuition that condition it. But given this component of sensible intuition, we are responsible for contributing the intellectual apparatus that makes cognition of objects—experience—possible, a conclusion, as we have seen, BonJour himself seems to accept. Without that contribution providing justified universal and necessary conditions for the possibility of experience, the epistemic subject is simply left with a set of diverse and chaotic perceptions, not objective experience.13 It is not an ontological claim per se, nor a claim about the existence of objects;14 it is a claim about the possibility of delineating the structure 13. This is the essence of Kant’s critique of Hume. I haven’t seen any need to address directly BonJour’s claim that Hume and Kant argue largely for the same “moderate empiricist” conclusions; that they do not should already be sufficiently clear from what has been said, for Kant’s claims about the necessity involved in synthetic judgments made a priori can reduce to analytic phenomenalist claims only on BonJour’s adoption of the dubious “two world” interpretation of Kant’s ontology. BonJour’s view also makes it much more difficult to see why Kant spent so much time outlining the specific differences between his view and Hume’s, precisely in relation to this issue of necessity and the a priori. 14. In pure mathematics, it should be noted, questions about the existence of

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   147 of human understanding (and reason). On Kant’s perspective, it requires reflecting on experience to determine the necessary conditions that make that experience possible. What we thereby discover is a set of concepts and principles that we are forced to presuppose, a transcendental analytic that serves as a logic of experience. Such a logic provides only a modest, minimal structure for experience, but by satisfying the requirements of universality and necessity, it avoids leaving us with simply a rhapsody of perceptions without any organization or coherence. As I have attempted to show above, this is precisely the epistemic force provided to transcendental analytic by the argument of the metaphysical deduction and the analogy between general and transcendental logic. Just as we might try to justify a straightforward minimal principle of logic, such as the principle of non-contradiction, by showing its indispensability as a presupposition for any use of reason, we can see that the burden here, of justifying a minimal set of a priori rules for possible experience, while still difficult and fraught with controversy, is both less onerous and more plausible than that which BonJour imposes on Kant’s Critical philosophy. Indeed, it is a tactic that BonJour himself employs; as he points out, certain positions that reject a priori justification “can be cogent only if they are themselves justified, directly or indirectly, in the very way that they are supposed to call into question” (123). While adopting this approach on occasion (16; 131; 155; 171), BonJour never pursues nor develops it in any systematic fashion. In his In Defense of Pure Reason, BonJour argues, largely on the basis of examples, that certain claims carry with them a degree of epistemic compulsion that cannot be accounted for by either a moderate or a radical empiricist. Nor, he suggests, can the traditional rationalist view be accepted, insofar as its claims of rational insight are both infallible and impervious to empirical testing or revision (16–17; 144). The inability of empiricism to provide an adequate explanation mathematical objects are for Kant questions about the constructibility of mathematical objects. BonJour, both in his discussion of Kant and more generally throughout In Defense of Pure Reason, tends not to distinguish logic and mathematics in any significant sense, nor does he indicate that such a distinction is of fundamental importance for Kant (and others, e.g., Frege).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

148   Kant and Contemporary Philosophy of the a priori, BonJour argues, forces it to embrace skepticism; the obvious failures of traditional rationalism—as seen in the various cases where infallible rational insight was asserted relative to a claim later refuted—require significant revision of its commitment to the a priori. These negative results, supplemented with BonJour’s arguments concerning our intuitions about necessity, form the basis of his defense of the moderate rationalist position. What BonJour does not seem to recognize is the remarkable similarity between his results and Kant’s. As discussed earlier, in the First Critique and elsewhere, Kant describes two models of the human intellect, only to reject them in favor of his own model. As we have also seen, in contrast to both the empirical and the rationalist models, Kant specifies a set of synthetic concepts and principles, and argues that they serve a priori as conditions for the possibility of experience, relative to a passively received manifold of intuition. In this context, it is crucial to make clear the role justification plays, relative to cognitive claims, and the entirely distinct issue of truth-determination. For Kant, one determines the truth or falsity of a given empirical judgment in accordance with “the nominal definition of truth ..... the agreement of knowledge with its object” (A58=B82) While Kant repeats this formulation throughout the Critique of Pure Reason (A191=B236; A820=B848), it should be noted just how little Kant says about truth there, in contrast to how much he says about justifying the conditions for assigning truth-values. Kant is not so much interested in the truth of a given proposition “p” as he is in establishing the universal and necessary conditions that allow “p” to be true or false. The content of a given empirical judgment is contingent upon what is provided in the relevant sensible manifold; consequently, what we know a priori relative to such judgments is distinct from whether that judgment is in fact true or false; “a sufficient and at the same time general criterion [Kennzeichen] of truth cannot possibly be given” (A59=B83). In contrast to this account, BonJour criticizes Kant for not being able to provide a sufficient justification for any general claim, P, that appears to be synthetic a priori; worse, “the original proposition P turns out not to be knowledge of any kind and very possibly not even true” (25). Here BonJour utterly fails to take into consideration the

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   149 distinction Kant makes between truth and justification, wherein transcendental analytic cannot yield the truth, but serves only as a “logic of truth” (A62=B87); on Kant’s analysis, one must sharply distinguish the truth of a given judgment and what must be presupposed even to investigate a judgment as true or false. In terms of the logical analogy that structures the Critique, one would be hesitant to characterize the principle of non-contradiction as “true” (and certainly not as “false”), nor is it clear that such a principle would qualify as “knowledge” in any but the most attenuated sense—this is, again, Putnam’s point: such truths of logic “are so basic that the notion of explanation collapses when we try to ‘explain’ why they are true”; as Thomas Nagel succinctly puts it, “There are types of thought we cannot do without.”15 Similarly, Kant argues that the causal principle defended in the Second Analogy, qua a principle of the “logic of experience,” is justified as a synthetic rule employed a priori as a condition of possible experience; in turn, any specific empirical judgment appealing to that principle in characterizing an event will remain contingent. To fail to mark this distinction can, on the one hand, lead to what Kant sees as Hume’s fundamental error: “inferring from the contingency of our determination in accordance with the law the contingency of the law itself ” (A766=B794). On the other hand, failing to mark the distinction between what is presupposed in any investigation into the truth and the conditions that must be satisfied for a specific empirical judgment to be true is to present “the ludicrous spectacle of one man milking a he-goat and the other holding a sieve underneath” (A58=B83). Only on the basis of his (at best) problematic “two world” reading of Kant’s ontology, and his failure to distinguish sharply the role of the a priori in Kant’s account of cognition in contrast to the role of truth relative to what is justified a priori, can BonJour conclude that “a Kantian view ..... does not constitute a significant further alternative with respect to the issue of a priori justification” (25). While I think it clear from the above account that BonJour’s conclusion here is itself not justified, there are several, relatively minor, points of misinterpretation that should at least be noted, in addition to BonJour’s tendency to 15. Putnam, “Analyticity and APriority,” 138; Nagel, The Last Word, 32.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

150   Kant and Contemporary Philosophy conflate Kant’s views of logic and mathematics. First: BonJour seems to suggest that Kant views his claims about the cognitive capacities of the epistemic subject under consideration as necessary. It is more plausible to read Kant as saying in the Critique that if a subject satisfies certain minimal criteria—the ability to use the first-person pronoun, and to use concepts to judge passively received spatio-temporal content—then it must (necessarily) have certain characteristics. Second: while BonJour argues that certain, presumably synthetic, principles are ultimately analytic (e.g., the principle of the transcendental unity of apperception and the causal principle [25n25]), this is difficult to reconcile with Kant’s repeated insistence that all analysis presupposes synthesis (B130, A77=B102–3), a point BonJour entirely ignores.16 Third: BonJour claims, somewhat unclearly, that it is a “quasi-Kantian idea that concepts necessarily falsify the reality that they attempt to depict, i.e., that any thinking being (or perhaps any being at all like us) will inevitably misrepresent in certain pervasive ways the features of the world that he attempts to represent” (152n15). He adds that he “can find no intelligible rationale of any sort for such a view,” although he seems not to realize that there is even less plausibility in attributing this view to Kant, to any neo-Kantians, or even to any viewpoint one might very generally describe as Kantian; indeed, not only is the distinction BonJour draws on to make this attribution rejected by Kant, the results BonJour objects to on the basis of this distinction are, as we have seen, precisely those Kant spends the most time arguing against: transcendental realism. Fourth: BonJour asserts that “if Kant had ever faced clearly the problem of the epistemological status of his own philosophical claims, he might have retreated into a more traditional rationalism” (25). While one might quibble with the qualification of “clearly” here, it is quite obvious from Kant’s work, both published and unpublished, that he was aware of this is16. To be sure, Kant does claim that the principle of the transcendental unity of apperception is analytic (although in the first edition of the Critique he claims it is synthetic [A117n]); the issue becomes even murkier from the tortuous footnote at B422, where Kant’s account of apperceptive unity becomes all but impossible to follow. In any case these confusions, which naturally have spawned a great deal of debate in the literature, do not permit BonJour’s much more sweeping conclusions, nor do they justify ignoring Kant’s considered view that synthesis always precedes analysis.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   151 sue, and struggled with it throughout his career. His first critic, J. G. Hamann, registered the complaint in his well-known, if curious, piece “Metakritik über den Purismum der Vernunft”;17 while it was not published until 1800, it is a virtual certainty that Kant was aware of the thrust of Hamann’s critique. In any case, Kant recognized the problem well before the writing of the Critique. As he points out in one of his Reflexionen, “I seek in an understanding, that requires rules, the knowledge [Kenntnis] of these rules themselves. This is paradox.”18 Yet he came to believe he had solved this “paradox”—seeking rules of an understanding that must provide rules to the understanding—as he explains in the Phillipi Logik, a set of lectures delivered in 1772: “But this paradox disappears when we realize we are not conscious of them.”19 The point is not simply that we are not conscious of the rules we employ, but that we can come to recognize, on reflection, that we must employ them in any investigation. To be sure, transcendental reflection takes place in accordance with rules, and the presupposition of such rules may, at some level, qualify as dogmatic. But, as Kant points out in a phrase reminiscent of Wittgenstein, “We can proceed in no other way in logic. The exercise of rules isn’t something taught, since I can do nothing other else than by means of these rules.”20 If Kant’s provocative, albeit deeply controversial, analogy between general logic and transcendental logic can bear any weight whatsoever, then there may well be a path open to provide a similar kind of “justification” for the rules of a logic of experience. But, clearly enough, it is disingenuous to suggest that Kant did not recognize the issue of metacritique. The point here is not that BonJour fails to be a good Kant scholar, or that he doesn’t have a historian of philosophy’s grasp of the his17. J. G. Hamann, Sämtliche Werke (Vienna: Hamann Verlag, 1949–57), 3:282–89. 18. Reflexion 1592; Ak. XVI, 28. The editor of these notes, Erich Adickes, is unsure of the dating of this note, but places it between 1755 at the earliest, and 1764 at the latest. 19. Phillipi Logik, Ak. XXIV.1, 316. 20. Phillipi Logik, Ak. XXIV.1, 339. Cf. L. Wittgenstein, Philosophical Investigations (New York: MacMillan, 1958), §1: “Die Erklärungen haben irgendwo ein Ende.” Additionally, Kant might reply that any objection to his approach, to be communicable, must satisfy the minimal logical conditions he has articulated (cf. A820=B849), a tactic to which BonJour appeals without, again, systematically developing it.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

152   Kant and Contemporary Philosophy tory of philosophy,21 but that he misreads Kant; on the basis of this misreading, he rejects the most powerful weapon he has at his disposal for defending his own view, a view that is ultimately quite similar to Kant’s, and without the kind of argumentative apparatus Kant—or some kind of systematic approach—provides. BonJour’s case for moderate rationalism comes to depend largely on examples, and thus lacks the support required to defend the claims motivating the need for the a priori to be included in any adequate epistemology that BonJour has so compellingly revealed. In short, he seems to dismiss precisely those arguments that might provide him support where it is most needed, and this dismissal is grounded in a fundamental misunderstanding of Kant’s project. BonJour remarks, “[A] combination of a coherence theory of empirical justification with a correspondence theory of truth is most unusual from an historical standpoint, but it is arguably the only hope for avoiding both foundationalism and skepticism while preserving a dialectically independent basis for defending an account of empirical justification (and also a more or less commonsensical conception of reality)” (SEK, 158). It is precisely this combination that BonJour seems unwilling to find in Kant’s Critical philosophy; yet, as we have already seen, Kant is explicit in his commitment to a Tarskian (or Aristotelian) characterization of truth. As Frederick Schmitt has noted, “[T]here are idealists who have been correspondence theorists [about truth]. Immanuel Kant, despite the common attribution to him of a coherence theory of truth, may be a prominent example.”22 Furthermore, while his account of epistemic justification is notoriously complex, one can, without too much trouble, find Kant specifically appealing to precisely the coherentist criterion BonJour defends in The Structure of Empirical 21. Thus BonJour fails to mention that C. D. Broad’s remark about the failure of justifying induction as “a scandal to philosophy” (188; 196) is a clear reference—and particularly ironic in the present context—to Kant’s observation that it is a “scandal to philosophy” that philosophers (Kant specifies idealists) have had to accept the existence of the external world on faith, and have been unable to prove it (B xxxix n.); nor does he mention that the epistemological concerns about the reliability of memory (125ff.) are precisely those Descartes raises in his discussion of atheism in his Response to the Sixth Set of Objections. 22. Frederick F. Schmitt, Truth: A Primer (Boulder: Westview Press, 1995), 146; Schmitt cites as evidence A58 [=B82].

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant and Contemporary Philosophy   153 Knowledge and builds upon in In Defense of Pure Reason: “[T]he difference between truth and dreaming is not ascertained by the nature of the representations which are referred to objects (for they are the same in both cases), but by their connection according to those rules which determine the coherence [Zusammenhang] of the representations in the concept of an object, and by ascertaining whether they can subsist together [beisammen stehen] in experience or not.”23 Finally, as we have seen, Kant’s results relative to synthetic judgments put forth a priori can be viewed as analytic, and even vacuous, only on a reading of his texts that is difficult, if not in the end impossible, to sustain. In sum, we find in Kant’s theoretical philosophy a powerful set of arguments, strategically going well beyond a dialectical strategy that must ultimately rest on the intuitive appeal to examples that BonJour uses to supply his positive comments in support of the a priori, that includes a commitment to the a priori, a correspondence theory of truth, and a coherence theory of justification: precisely the epistemological desideratum BonJour pursues.

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant, Davidson, and Conceptual Schemes The philosophical work of Donald Davidson ranges over an enormous number of topics, including investigations into subjective probability and decision theory, intentionality, general questions about mind-body interaction, the meaning of metaphor, and a large number of other issues. Here I want to focus on the specific aspect of his work for which he is perhaps best known, the rejection of what Davidson calls the “third dogma of empiricism,” the distinction between scheme and content. On Davidson’s view, the contrast between an organizing “scheme” and a given content to be organized is itself ill-posed; this 23. Prolegomena, Ak. IV. 290. I have not gone into the question here of whether Kant’s appeal to empirical realism offers “a more or less commonsensical conception of reality,” in contrast to BonJour’s apparent endorsement of an Aristotelian or Thomistic ontology (see 183ff.). I will simply note that it is not entirely clear how one can reconcile Thomistic Realism with what BonJour has argued in SEK in criticizing contemporary foundationalism and its appeal to “the given.” BonJour admits his views are “highly tentative and exploratory” (185), and the issue merits more attention than I can give it here.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

154   Kant and Contemporary Philosophy leads to the result that because the very notion of alternate conceptual schemes is incoherent, the “very idea” of a conceptual scheme is untenable.24 As we have seen, Laurence BonJour has urged that any adequate epistemological account must incorporate an explicit commitment to the a priori; certain claims carry with them a degree of epistemic compulsion that cannot be accounted for by either a moderate or a radical empiricist. Yet, by misconstruing Kant’s strategy, BonJour is forced to rely on arguing by example, and thus fails to establish a sufficiently strong case for the a priori element in his own “moderate rationalism.” In contrast, Davidson seems to reject any talk of the a priori, and in general seems resistant to incorporating any strong sense of necessity into his views of language, cognition, and the complex strategic treatment of intersubjectivity he calls “triangulation.” It is here, however, that the rich confrontation between Davidson and Kant reveals the significance of Kant’s approach for Davidson’s work, and that, ultimately, Davidson’s own arguments suggest precisely the kind of commitment to necessity that grounds Kant’s strategy. In sum, by regarding Kant’s approach as fundamentally logical, I hope to show that Davidson’s results are in many ways those of Kant, while further showing that both positions require a notion of necessity that Kant explicitly recognizes but that Davidson eschews. Davidson’s argument is sufficiently well known that I do not need to go into any great detail in characterizing it. Various writers have proposed that different cultures (anthropological relativism—e.g., Benedict),25 different linguistic groups (linguistic relativism—e.g., Sapir and Whorf), different genders (on a standard reading of “difference” or “standpoint” feminism), different chronological eras in science (scientific/ontological relativism—e.g., Kuhn, and, on some readings, Quine), and still others perceive and organize the world in fundamentally different ways. At some conceptual distance between groups with different conceptual schemes, that distance becomes intractable, those 24. The locus classicus of the view is Davidson, “On the Very Idea of a Conceptual Scheme,” 183–98; unless otherwise noted, all citations (or paraphrases) come from this article. See also Davidson, “The Second Person,” 107–21. 25. R. Benedict, Patterns of Culture (Boston: Houghton Mifflin, 1934).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   155 schemes become incommensurable, and there is a complete inability for a group using one scheme to communicate its ideas to a group using a different, incommensurable scheme. In a nutshell, Davidson says this picture is incoherent. His argument is couched in terms of translation: “We may identify conceptual schemes with languages, then, or better, allowing for the possibility that more than one language may express the same scheme, sets of intertranslatable languages. Languages we will not think of as separable from souls; speaking a language is not a trait a man can lose while retaining the power of thought.” For Davidson, the question of incommensurability becomes a question of intertranslatability: “Can we then say that two people have different conceptual schemes if they speak languages that fail of intertranslatability?” His answer, in short, is no; language users, on his account, are creatures with beliefs; one to whom we attribute language, then, is one to whom we attribute the holding of beliefs (and vice versa). Our ability to understand another, it is further argued, can succeed only if we adopt a principle of charity: roughly, assuming another’s beliefs are largely true.26 Hence, an attribution of language or a conceptual scheme to another is to attribute largely true beliefs to it; and to do so, relative to another, is to undercut the possibility of that other having a radically distinct conceptual scheme. As William Taschek succinctly puts the point: “If I am going to suppose that an untranslatable language is possible, then I am supposing two things: I am supposing that something is both a language and untranslatable. Davidson’s point will be that the very conditions for my coherently supposing that something is a language at all are inconsistent with the possibility that it might be in principle untranslatable by me.”27 26. See “Thought and Talk” in Inquiries into Truth and Interpretation, passim; “The basic claim is that much community of belief is needed to provide a basis for communication or understanding; the extended claim should then be that objective error can occur only in a setting of largely true belief ”; “The Method of Truth in Metaphysics,” in ibid., 200. It should be noted that Davidson takes Tarski’s criterion of truth (“Convention T”) as embodying “our best intuition as to how the concept of truth is used”; ibid., 195. See also D. Davidson, “The Structure and Content of Truth” (1989 Dewey Lectures), Journal of Philosophy 87, no. 6 ( June 1990): 279–328. 27. W. Taschek, “Making Sense of Others: Donald Davidson on Interpretation,” Harvard Review of Philosophy 10 (2002): 38.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

156   Kant and Contemporary Philosophy To summarize, then, Davidson rejects the scheme-content distinction on the basis of the following argument: 1. Propositional thought requires language; one is a language user if and only if one has beliefs. 2. Conceptual schemes are languages. 3. A language can be recognized as a language only if it is translatable into one’s own language. 4. Translation failure can be either complete or partial.28 5. There can be no complete failure to translate a language while construing it as a language. 6. There can be no partial failure to translate a language while accepting that one is engaged in translation. Therefore 7. The notion of incommensurable—a complete or partial failure of translation—conceptual schemes, or languages, is incoherent and

Copyright © 2008. Catholic University of America Press. All rights reserved.

8. The very notion of a conceptual scheme—a scheme-content distinction—is incoherent. It is important to see precisely what Davidson is claiming—and what he is not claiming—here. As we will see in a slightly different way below, in discussing his triangulation strategy, he is not making the rather implausible suggestion that either in fact or even ideally the argument in question results in perfectly smooth communication, or some sort of idealized semantic transparency between two language users, although he has been interpreted in that way. Thus Lorraine Code has written, relative to Davidson’s argument, that “[t]ranslation and communication succeed often enough in practice to suggest that extreme relativism is implausible. Yet the rarity of perfect translation 28. It is important to see that Davidson takes such a partial failure of translation to be a situation where some range of sentences “could be translated and some could not.” As we will see, Davidson’s view here has been interpreted as making a much stronger claim than he in fact asserts.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   157 and the translinguistic elusiveness of many concepts and ideas recommend respect for a significant amount of relativism. Perhaps it is logically possible for all linguistic utterances to be translated into any other language without remainder. But in practical terms, conceptual barriers are often difficult to assail.”29 Certainly Davidson is not calling for “perfect translation”; his frequent references to Quine and his indeterminacy of translation thesis indicate sufficiently that for Davidson, a “perfect translation” would not be a rare occurrence, but would be a non-occurrence. This is not to say that all translations are imperfect; rather, it is that Davidson is more concerned with those conditions under which any translation—qua communicative act—could be successful.30 Indeed, Davidson is quite explicit in making this point: “We have found no intelligible basis on which can be said that schemes are different. It would be equally wrong to announce the glorious news that all mankind—all speakers of language, at least—share a common scheme and ontology. For if we cannot intelligibly say that schemes are different, neither can we intelligibly say that they are one.” Consequently, he is not claiming that one language can be translated into another “without remainder.” To be sure, conceptual barriers are often difficult—or worse—to assail. But this is not to recommend respect for a “significant amount of relativism,” at least significant enough to make the notion of radically distinct conceptual schemes coherent. Rather, it is to recommend respect for certain areas of agreement—initially, at least, those principles that I have argued cannot be relativized, specifically those principles Kant views as belonging to logic. As Davidson notes, his method “is not designed to eliminate disagreement, nor can it; its purpose is to make meaningful disagreement possible, and this depends entirely on a foundation—some foun29. L. Code, What Can She Know? (Ithaca: Cornell University Press, 1991), 59. I think Andrew Cutrofello makes a similar mistake by inferring from Davidson’s argument the “grandiloquent extrapolation that all rational beings must share the same beliefs,” in “On the Transcendental Pretensions of the Principle of Charity” in The Philosophy of Donald Davidson, ed. L. E. Hahn (Chicago: Open Court Press, 1999), 334. 30. Furthermore, those who actually translate literature or, for that matter, any other kinds of texts, would insist that the term “perfect translation” is an oxymoron. For some discussion of this point, see G. Steiner, After Babel (New York: Oxford University Press, 1975).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

158   Kant and Contemporary Philosophy dation—in agreement.” It is, of course, precisely the modal status of what would constitute such a foundation where Kant and Davidson disagree. In his later work, Davidson begins to discuss intersubjective claims in terms of a triangulation strategy, which he characterizes as a situation “that involves two or more creatures simultaneously in interaction with the world and with the other agent.” He continues by glossing triangulation as “the result of a threefold interaction, an interaction which is twofold from the point of view of each of the two agents: each is interacting simultaneously with the world and with the other agent.”31 In this context Davidson can appeal to the earlier results, relative to conceptual schemes; we attribute claims made by another, about a world we share, and we must share a sufficiently similar world in order to recognize each other as making claims, for “there would be no saying what a speaker was talking or thinking about, no basis for claiming he could locate objects in an objective space and time, without interaction with a second person.”32 Thus, Davidson appeals to the idea that in interacting with another, both agents “triangulate,” relative to each other as well as relative to a shared objective world. Such interaction is fundamentally linguistic, and by virtue of attributing interpretable linguistic behavior to another, we concomitantly identify an objective world we share with that other agent. Finally, to interpret that other’s claims, we assume, in accordance with the principle of charity, that her claims are “largely true,” an attribution we of course also make relative to our own beliefs. This is, again, hardly to say that we all make the same judgments, speak the same language, or share identical beliefs: it is, rather, that we share a sufficiently broad set of assumptions and perceptual claims to make meaningful disagreement possible. As noted, it is the modal status of such assumptions where Kant 31. Davidson, “The Emergence of Thought,” reprinted in Subjective, Intersubjective, Objective (Oxford: Clarendon Press, 2001), 128. 32. “The Second Person,” 121. I cannot go into some of the technical issues involved here, such as the distinction between distal and proximal stimuli in perception, and how Davidson sees triangulation as different from Kripke’s way of responding to Wittgenstein’s private-language argument, both for reasons of space and to maintain the focus on the interaction between Davidson and Kant.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant and Contemporary Philosophy   159

Copyright © 2008. Catholic University of America Press. All rights reserved.

and Davidson seem to part company. One might be tempted to take a short argumentative path here, and characterize Davidson’s position as suggesting that we must make the kinds of attributions and assumptions he specifies in order for intersubjective communication to be possible. Here we would see the modal combination that is the hallmark of Kant’s transcendental strategy, by identifying a concept or rule that is reflectively identified as necessary for judgments, ranging over a given domain, to be possible.33 Yet Davidson resists the strong sense of the a priori so fundamental to Kant’s approach (in addition to embracing Quine’s rejection of the analytic-synthetic distinction); the question I want to explore here is whether or not, at the conceptual level of Kant’s logical claims, this is at best a distinction without a difference. With his forms of intuition, tables of judgments and categories, as well as the various other features of his oft-abused “Architectonic,” it is easy to see Kant not just as a preeminently systematic philosopher; it is also easy to label him as committed to precisely the kind of schemecontent distinction of which Davidson is so sharply critical. One who has certainly read Kant in this way is Richard Rorty. In his well-known paper “The World Well Lost,” Rorty writes: The possibility of different conceptual schemes highlights the fact that a Kantian synthesized intuition can exert no influence on how it is to be synthesized—or, at best, can exert only an influence we shall have to describe in a way as relative to a chosen conceptual scheme as our description of everything else. Insofar as a Kantian intuition is effable, it is just a perceptual judgment, and thus not merely “intuitive.” Insofar as it is ineffable, it is incapable of having an explanatory function.34

Rorty reads Kant as identifying two fundamental aspects of cognition: a set of “unsynthesized intuitions,” to which we apply a set of concepts, the apparatus we use in order to make sense of those intu33. From a much different direction, and on the basis of a much different reading of Kant, Richard Rorty has claimed that Davidson “seems to have found a transcendental argument to end all transcendental arguments.” “Transcendental Arguments, SelfReference, and Pragmatism,” 78. 34. Rorty, “The World Well Lost,” Journal of Philosophy 69 (1972): 650.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

160   Kant and Contemporary Philosophy itions, or even create them.35 In this way, Rorty—who is here representative of a common, although thoroughly inaccurate, interpretation of Kant’s ontological claims—sees Kant as succumbing to the “myth of the given”:

Copyright © 2008. Catholic University of America Press. All rights reserved.

The picture which holds traditional philosophy captive is that of the mind as a great mirror, containing various representations—some accurate, some not—and capable of being studied by pure, nonempirical methods. Without the notion of the mind as mirror, the notion of knowledge as accuracy of representation would not have suggested itself. Without this latter notion, the strategy common to Descartes and Kant—getting more accurate representations by inspecting, repairing, and polishing the mirror, so to speak— would not have made sense.36

On this reading, we have on one hand a “world” (or “the world”) and, on the other, a subject who brings with it an ahistorical, neutral framework that is somehow meant to map onto, or “mirror” that world, and such a view is clearly committed to the kind of scheme-content distinction we have seen Davidson reject. I think Davidson is correct to do so, but I also think that such an interpretation of Kant’s epistemology is unsupportable (and I should add that I have not found anything in Davidson that forces us to adopt the kind of interpretation Rorty puts forth). For Kant, there is no notion of “fitting” the categories to the “given”; rather, the logic of experience articulated in the Transcendental Analytic—which of course must be supplemented by both the Tran35. Thus Rorty groups Kant with Hegel, in that both were guilty of “having confused the idea that nothing has such [an intrinsic] nature with the idea that space and time are unreal, that human beings cause the spatio-temporal world to exist”; Contingency, Irony, and Solidarity (Cambridge: Cambridge University Press, 1989), 4. As noted earlier, only in judgments of pure mathematics can we be said to “create” objects, and there the claims are about constructability, not existence. 36. Rorty, Philosophy and the Mirror of Nature (Princeton: Princeton University Press, 1979), 12. Wilfrid Sellars, who introduced the term “myth of the given,” claims that Kant was not altogether free of it. See the expanded edition of his original article in W. Sellars, Empiricism and the Philosophy of Mind, ed. R. Brandom (Cambridge, Mass.: Harvard University Press, 1997), 14. However, I think that little of what Sellars says in this article, or in his other work on Kant, supports attributing this to Kant, and that Kant himself can plausibly be interpreted as suggesting Sellars’s criticisms of a traditional epistemic reliance on the “given.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   161 scendental Aesthetic and what Kant says about the regulative employment of reason—isolates (reflectively) the intellectual component that is necessary to make experience possible. This structure then serves as a set of conditions for making judgments and investigating the truth of those judgments: not in the sense of breaking through a “veil of perceptions” to the world as it really is, but in the sense of recognizing that any cognitive investigation must conform to the conditions of possible experience. Instead of seeking a match between our appearances and what “really” appears, all such questions come within appearances: “The true correlate of sensibility, the thing in itself, is not cognized and cannot be cognized, through representations [of sensibility], and in experience no question is ever asked about it” (A30=B45; my emphasis). Rorty refers to “unsynthesized” intuitions, and attributes to Kant a conception of synthesis that Rorty describes as a “metaphor” for truth, on the basis of which he claims further that “Kant wanted to consign science to the realm of second-rate truth—truth about a phenomenal world.”37 On the contrary, within Kant’s epistemology, truth is the “agreement of cognition with its object,” made possible through judgment (A58=B82).38 Transcendental Analytic provides a set of conditions that a judgment of possible experience must satisfy to be able to be true. If we take Kant’s conception of objects of experience to be veiled representations of objects as they really are, then we can ask how phenomena correspond to, or are caused by, the thing-initself. As discussed earlier in calling into question BonJour’s adoption of a “two-world” characterization of Kant’s ontology, I think this is to operate with a degenerate cognitive sense of the thing-in-itself, and a very unKantian conception of truth as the agreement between objects as they appear and objects as they are in themselves. If, on the other hand, we make cognitive claims only about objects as they appear, or what one might consider objects in a non-degenerate sense, we arrive 37. Rorty, “Transcendental Arguments,” 97; Contingency, Irony and Solidarity, 4. 38. As well as A157=B197; A191=B236; A237=B296; A642=B670; A820=B848, inter alia. Cf. Frege: “Whenever anyone recognizes something to be true, he makes a judgement”; Posthumous Writings, trans. R. White and P. Long (Oxford: Blackwell, 1997), 251.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

162   Kant and Contemporary Philosophy at the Kantian conception of truth, which relies on a quasi-Tarskian (nominal) definition of truth, and is grounded in a transcendental analytic that provides only a logic of truth (A63=B87). At the same time, it is clear that in Kant’s epistemology such a logic qualifies, in a modified sense, as a “scheme,” relative to the content introduced, within the forms of intuition, as sensibility. Furthermore, there is no question that qua scheme, it is seen by Kant to impose absolutely universal and strictly necessary rules for epistemic claims to be possible. But, again, we can look at the analogy Kant draws between general and transcendental logic to see that the kind of scheme-content dualism Davidson rejects is distinct from that which Kant endorses. L. W. Beck has argued that “the philosopher begins where he is,” and the point here can be made by recognizing, with Kant, that everyone begins where he or she is.39 We begin with a congeries of judgments—empirical, mathematical, moral, teleological, and aesthetic, among them—although we can focus here on ordinary empirical judgments. One who has a philosophical bent may further explore what those judgments presuppose; if, for the sake of the argument, we grant Kant’s results in the Transcendental Analytic, such judgments must be made about relatively enduring substances, causally interacting. If Kant’s claim is at all plausible that the rules of Transcendental Analytic yield a logic, as a set of rules ranging over the domain of passively received sensibility, then we see that one who puts forth as a judgment of possible experience a claim about an uncaused event is making a logical mistake, at the level of transcendental logic. We saw earlier that a judgment about two straight lines enclosing a space (in, presumably, Euclidean geometry) is a claim that can be shown to impossible only on the basis of an attempt to construct it in pure intuition. In the same way, a reference to an “uncaused event” is not a contradiction at the level of general logic (as would be, say, a reference to an “uncaused causal event”), but only within the domain of judgments of possible experience. Without presupposing some set of rules, then, experience would not be possible, in the same way that one who embraces contradictions 39. See Beck’s “Toward a Meta-Critique of Pure Reason,” in Essays on Hume and Kant, esp. 31.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   163 ultimately results in thinking nothing. For Kant, of course, this is the result of a long, sometimes tortuous process, which is designed to isolate those necessary conditions that can be discovered only reflectively and is embedded in his radically subjective approach to philosophical reflection. But if we pursue this strategy, and find some such set of rules, we also find that we regard them as universal and necessary—a priori—and thus as providing constraints on anyone putting forth judgments. As Manley Thompson has observed, Kant’s first-person strategy imposes upon us such rules; “Unless one simply sees an impossibility in accounting for a bang of thunder heard today and happening tomorrow,” such an argument has no force in helping us characterize our cognitive practices.40 In short, for Kant we discover reflectively a set of rules that we must adopt in order to make sense, and that we also must attribute to anyone else if we are to regard him or her as making sense. This hardly means, of course, that Kant believes everyone thinks, experiences, and reasons in the same way, although he has, I think, been interpreted this way. Thus Robin May Schott writes of Kant’s views, “Not only are all objects known in exactly the same way by the individual thinker, but all thinkers know each object in the same way as every other thinker.”41 It is difficult to know precisely how to interpret this claim. It would seem perverse to claim on the basis of a general acceptance of certain minimal canons of rationality, for example, the principle of non-contradiction, modus ponens, or even some version of a causal principle, that we all think “alike.” For Kant is not claiming that all cognitive agents make the same cognitive judgments, nor can he be seen as claiming that all such agents should make the same cognitive judgments. Transcendental analytic provides an framework within which judgments of experience are possible, hence, a framework within which meaningful disagreement can take place. If, for example, two people find themselves arguing over the relationship between interest rates and the effect on the stock market, both presuppose a basic cause-and-effect framework within which they pursue the specifics of their respective 40. Thompson, “Things in Themselves,” 41. 41. Robin May Schott, Eros and Cognition (Boston: Beacon Press, 1988), 131.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

164   Kant and Contemporary Philosophy causal (and empirical) analyses. The structure that makes this debate possible is that which Kant argues for in the Transcendental Analytic, most specifically the Second (and Third) Analogy. In the same way, the argument over whether a rise in interest rates is “good” or “bad” presupposes a normative constraint that cannot be wholly accounted for in terms of an empirical account; that account takes place within a framework within which such evaluative terms are already present. While Kant’s strategy is grounded in his radical first-person approach, Davidson proceeds intersubjectively, advocating that cognitive claims be adjudicated by “triangulating” among two subjects and a shared external world: “the basic triangle of two people and a common world is one of which we must be aware if we are to have any thoughts at all.”42 Making sense of the other’s utterances—translating them, in a sense—may well be a long and even fruitless process. If I take another to be a language-user, I attribute beliefs to this agent; as Davidson puts it, “we should each provide the other with something understandable as a language”; yet for Davidson, this at best indicates a necessary intent to communicate and “does not involve following shared rules and conventions.”43 Davidson here seems to adopt what Wittgenstein characterizes as simply “the way we go on” in explaining how one can be seen as following a rule, and that “language is necessarily social.”44 Kant’s view is clearly much closer to that generally identified as the Wittgenstein of the Tractatus; simply put, if we choose to identify an entity as an agent, or as rational—which is itself a contingent attribution—we attribute, at the same time, certain constraints to its thought and judgments. We may discover these constraints by philosophical reflection 42. “The Irreducibility of the Concept of the Self,” reprinted in Subjective, Intersubjective, Objective, 87. Kant says very little about what is generally discussed under the problem of “other minds,” although I think it is clear from what has been said that when we (contingently) identify another as rational, or as an agent, we then necessarily regard it as subject a priori to the same constraints by which we recognize ourselves as constrained. In a passage he eliminated for the second edition of the Critique of Pure Reason, Kant does state, “It is obvious that, if one wishes to represent to oneself a thinking being, one must put oneself in that being’s place, thus substituting [unterschieben], as it were, one’s own subject for the object one seeks to consider (which does not occur in any other investigation)” (A353=B354). 43. “The Second Person,” 114. 44. “The Emergence of Thought,” 129.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   165 (the method of the First Critique), although there is no in-principle reason why they could not be discovered by the kind of triangulation strategy for which Davidson argues. (And one might point here to Kant’s argument in the “Refutation of Idealism” as leading precisely to the result Davidson relies on, that we share a common, external, objective world as part of this “triangle.”) The question becomes whether, at the end of our philosophical reflections, we are forced to assume that the other speaker must conform to the conditions Kant calls “logical” if we are to have any chance of taking that speaker to be a language user. Kant’s point, in this context, would seem to be that one who made sounds that, if even interpretable, were contradictory, were inconsistent, and failed to make any reference to a spatio-temporal “world” shared in the most minimal sense by both speakers, would simply be making noise. Davidson speaks in terms of what function as rules and conventions, or in terms of our best epistemic practices, where success is regarded in terms of smoothness of communication. In contrast, Kant would respond that, ultimately, at the level of logic—whether relative to possible thought or to possible experience—there is a difference between a situation where we follow our “best” available option and a situation where there is no option. In short, on the latter view, we have no choice but to presuppose some set of rules as necessary for a given judgment to be possible. It is important, however, to point out that a principle or set of principles can no more serve to constitute a “logic” in any rich sense than can logic constitute a full, or even particularly rich, account of rationality. The point is, rather, that an agent who rejects all principles, that is, who fails to adopt any principle that could be characterized as a principle of logic, could not qualify as rational. The well-known historian of logic Józef Bocheński has argued that if we accept as “criteria of rationality” “the rules of acceptance or rejection of statements,” we discover three rules—modus ponens, the dictum de omni [et nullo], and the principle of coherence (that contradictions should be avoided)—that “all the logics known, with no exception,” contain.45 45. J. Bochen´ski, “History of Logic and the Criteria of Rationality,” in Logic Counts, ed. E. Zarnecka-Bialy (Boston: Kluwer, 1991), 86.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

166   Kant and Contemporary Philosophy

Copyright © 2008. Catholic University of America Press. All rights reserved.

One might think in terms of the principle of identity, the principle of excluded middle, and the principle of non-contradiction (i.e., the traditional “laws of thought”); other candidates that might be added include consistency, bivalence, modus ponens, quantificational reference, double-negation elimination, and the hypothetical syllogism or “chain argument”; still others might include the successor function (as in the Peano postulates for arithmetic), Tarski’s “Convention T,” and perhaps even some constraint on inductive reasoning, such as a probability calculus. Still other candidates, of course, might be proposed as additions to such a list. At the same time, various criticisms have been brought forth against virtually all the members of this list.46 The principles discussed here, then, serve as candidates for being necessary—and not sufficient—conditions of rationality, and when we move to the domain of transcendental logic, we introduce sensible content, and we are no longer simply reflecting on the conditions of thought simpliciter. Furthermore, it is worth emphasizing again that this claim functions as a normative constraint on reason; frequently agents are, in fact, inconsistent; frequently we commit the simplest mistakes in reasoning and unknowingly embrace contradictory or inconsistent beliefs.47 But when made cognizant of such mistakes, we have an intellectual obligation to eliminate the inconsistency.

46. Thus, for instance, some quantum logics call into question bivalence, as do intuitionist logics, which leads to a non-classical treatment of double negation and of excluded middle; paraconsistent logics are willing, as we have seen, to countenance contradictions. Irving Copi, in his Introduction to Logic (6th ed.) (New York: MacMillan, 1982), 319ff., gives a brief survey of traditional criticisms of the various laws of thought; cf. L. S. Stebbing, A Modern Elementary Logic (London: Methuen, 1943), 146ff. For some discussion of so-called “deviant” and/or alternative logics, see S. Haack, Philosophy of Logics (Cambridge: Cambridge University Press, 1978) and Deviant Logics (Cambridge: Cambridge University Press, 1974). 47. For example, on most standard models of economic behavior, the principle of transitivity of preference-ordering is taken as axiomatic, although, in practice, such a principle is not infrequently violated. There is also extensive evidence that the kinds of results found in basic truth-functional semantics are violated with surprising frequency; for some of this evidence, see A. Tversky, P. Slovic, and D. Kahneman, Judgments under Uncertainty (Cambridge: Cambridge University Press, 1982); more generally, R. Nozick, The Nature of Rationality (Princeton: Princeton University Press, 1993).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant and Contemporary Philosophy   167

Copyright © 2008. Catholic University of America Press. All rights reserved.

An Excursus into the Postmodern

Introduction As we have seen, Kant is committed to a very strong sense of the a priori, which he glosses in terms of its two complementary criteria of absolute universality and strict necessity; where all such necessity lies grounded in a transcendental condition (A106). “Transcendental,” in turn, is characterized as “such cognition as concerns the a priori possibility of cognition, or its a priori employment” (A56=B80). As I have tried to show, Kant regards the concepts, and the consequent rules developed from those concepts, as transcendental conditions relative to a given domain. Thus the rules of general logic serve as universal and necessary (a priori) rules for the possibility of thought, or judgments simpliciter, while the rules of transcendental logic—specifically transcendental analytic—serve as universal and necessary (a priori) rules for the possibility of judgments about objects. Kant’s argument, then, is that just as we can show, reflectively, that thought must conform to certain conditions to qualify as thought—for example, by conforming to the principle of non-contradiction—we can also show that thought about objects must analogously conform to certain conditions to qualify as judgments about objects, although within this domain we introduce a mind-independent source, sensibility, passively received within the forms of time and space. The conditions involved here are, of course, those presented in the Axioms of Intuition, the Anticipations of Perception, the Analogies of Experience, and the second-order modalities of the Postulates of Empirical Thought. If this analogy is at all plausible, we can then see Kant as constructing a logic of experience, yielding a set of necessary conditions for the possibility of experience, or transcendental conditions for making judgments about objects of possible experience, in that the “conditions of the possibility of experience in general [überhaupt] are at the same time the conditions of the possibility of the objects of experience” (A158=B197). It is worth remarking again in this context that establishing that a given judgment satisfies these transcendental conditions does not determine that judgment’s truth-value; rather, it simply reveals that such a judgment qualifies as a “judgment of possible experience,” and as such is truth-evaluable.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

168   Kant and Contemporary Philosophy

Copyright © 2008. Catholic University of America Press. All rights reserved.

We have also seen Laurence BonJour’s argument for a “moderate” rationalism, that insists on a commitment to the a priori, albeit a commitment that is motivated by example and ultimately based on an inductive argument that is too weak to establish the necessity to do the philosophical work required. On the other hand, Donald Davidson resists any explicit reliance on the a priori, or, indeed, on any sense of necessity sufficiently strong to satisfy the demands of Kant’s project, although I have argued above that Davidson’s strategy may well have to adopt implicitly very much the same necessity found in Kant’s own arguments. I now want to turn to the strongest and most explicit rejection of any philosophical commitment to necessity, that which is found in a diverse group of philosophers who have—fairly or not— been labeled as “postmodernists.” Without any attempt to adjudicate various claims with any precision about what qualifies as modernism or postmodernism, and whether postmodernism is a development from or radical rejection of modernism, I want to look specifically at the critique of Kant that has been developed from a general perspective that I will refer to—given that proviso—as “postmodern.” I hope to show that this critique is grounded in an inaccurate understanding of what Kant believes he establishes, specifically in the realm of epistemology and metaphysics, and a systematic underestimating of Kant’s own critical conception of the limits of philosophy. The Tenets of Postmodernism While the historical roots of what has come to be called postmodernism may go back at least as far as J. G. Hamann, for the purposes of this brief discussion we can identify Nietzsche as largely responsible for the systematic and sustained critique of those values—of modernism, of humanism, and of the Enlightenment—that has given rise to the various versions of postmodern philosophy.48 There is a rela48. To discuss postmodernism is immediately to invite oversimplification, and I will not here even attempt to provide a taxonomy of the various strands of thought— deconstruction, poststructuralism, neo-pragmatism, etc.—that have come under the rubric of the postmodern. Of course, in addition to postmodern philosophy, broadly construed, one also finds its influence in fiction (Barthelme, Barthes, Borges, Coover, Cortazar, Delillo, Jackson, Pynchon), art (L. Freud, Holzer, Scher), architecture ( John-

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant and Contemporary Philosophy   169 tively short list of figures against whom Nietzsche reacts, throughout his writings, with some degree of ambivalence. Most prominent on any such list, of course, are Wagner and Socrates—but we can also add Plato, Christ, Kant, and Schopenhauer. Nietzsche makes no sharp break with Kant, as in the celebrated and well-documented conflict with Wagner; instead, we find a grudging intellectual respect together with an almost palpable revulsion for what Nietzsche takes Kant to represent: as early as the unpublished 1872 essay “Der letze Philosoph” and as late as Ecce Homo (1889), we see Nietzsche’s acknowledgement of Kant’s achievement always combined with his guarded antipathy for Kant as a “philosophical laborer.” Such philosophical laboring, the “philosophy” of the past, is Nietzsche’s target, when he writes in Beyond Good and Evil:

Copyright © 2008. Catholic University of America Press. All rights reserved.

Are these coming philosophers new friends of “truth”? That is probable enough, for all philosophers so far have loved their truths. But they will certainly not be dogmatists. It must offend their pride, also their taste, if their truth is supposed to be a truth for everyman—which has so far been the secret wish and hidden meaning of all dogmatic aspirations. “My judgment is my judgment”: no one else is easily entitled to it—that is what such a philosopher of the future may perhaps say of himself.49

Standardly, the conflict between Nietzsche and Kant has been seen as one focusing on ontology, between Nietzsche’s perspectivism and, in contrast, what he views as Kant’s ossified realism. Whether due to his son, Venturi), and music (Cage, Glass, Wilson, Eno, Anderson). A number of debates that continue to swirl around this taxonomy—e.g., Is Bakhtin a postmodernist? Is Wolff ’s To the Lighthouse a “modernist” novel in contrast to Rushdie’s postmodern The Satanic Verses?—will simply be ignored here. 49. Nietzsche, Beyond Good and Evil, trans. Walter Kaufmann (New York: Vintage Books, 1966), § 43. In Nietzsche on Truth and Philosophy (New York: Cambridge University Press, 1990), Maudemarie Clark has argued that in Nietzsche’s mature work, he is committed to some sort of commonsense realism, although he consistently rejects “metaphysical realism.” In Nietzsche and Metaphysics (Oxford: Clarendon Press, 1995), 22–25, Peter Poellner has criticized her account on various grounds. Yet, as Poellner himself notes in his concluding chapter, Nietzsche is faced with the possibly crippling problem of self-reference, a problem that may be avoided only by adopting the commitment to common-sense realism and a minimal, Tarski-style, criterion of truth Clark urges, a criterion Davidson also seems to regard as indispensable to giving even a minimally coherent conception of truth.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

170   Kant and Contemporary Philosophy

Copyright © 2008. Catholic University of America Press. All rights reserved.

rather unsystematic reading of Kant (and others) or to his over-reliance on Schopenhauer’s interpretation of Kant,50 Nietzsche clearly adopts the “two-world” interpretation of Kant’s phenomenon/noumenon distinction criticized above. As Rüdiger Grimm has succinctly summarized the point, “Nietzsche views Kant as having shared that traditional, Western intellectual paradigm which we have broadly characterized as the correspondence view. Nietzsche finds evidence of this in Kant’s refusal to abandon the noumenon-phenomenon distinction which, even if one regards the noumenon negatively as a ‘Grenzbegriff,’ nevertheless implies the existence of an objective realm which is the necessary condition for the possibility of experience.”51 I believe that the real conflict between Kant and Nietzsche arises not so much in the ontological dispute over the thing in itself and its viability, but rather over the more fundamental modal difference between them concerning necessity and the very possibility of naturalizing Kant’s program. Nietzsche’s perspectivism, then, would make otiose, on his reading of Kant’s critical strategy, the demand of Kant’s quaestio juris; but by adopting a facile reading of Kant’s epistemology, Nietzsche misses the opportunity to develop his own critique by defending it against one of its strongest competitors.52 In any case, Nietzsche’s philosophical perspective can be seen as responsible for generating what some have called a “hermeneutics of suspicion,” strongly influencing Heidegger, and in turn Foucault and Derrida, among many 50. The appendix (“Die Kritik der kantsichen Philosophie”) to volume 1 of Die Welt als Wille und Vorstellung. 51. R. Grimm, Nietzsche’s Theory of Knowledge (Berlin: de Gruyter, 1977), 60. It should be noted that Grimm is virtually alone among Nietzsche’s commentators in suggesting that Kant’s theoretical philosophy might simply omit the thing-in-itself (56). Others, however, have suggested this line of thought relative to Kant: thus Hilary Putnam, The Many Faces of Realism (LaSalle, Ill.: Open Court, 1988), 41, writes, “I think that almost all of the Critique of Pure Reason is compatible with a reading in which one is not at all committed to a Noumenal World, or even ..... to the intelligibility of thoughts about noumena.” I think this is also Arthur Collins’s point when he writes, “Kant is a truly anti-idealist thinker. His major failing is not inconsistency. It is his underestimation of the tenacity of Cartesian assumptions among his readers.” Possible Experience (Berkeley: University of California Press, 1999), 41. I have discussed briefly why Kant does not take this step in my comments on Hoke Robinson’s “Langton and Traditionalism on Things in Themselves,” Southwest Philosophy Review 18 (2002). 52. I have argued for this in more detail in “Nietzsche, Kant, and the Thing in Itself,” International Studies in Philosophy 25, no. 2 (1993): 66–77.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant and Contemporary Philosophy   171

Copyright © 2008. Catholic University of America Press. All rights reserved.

others, and allowing postmodern philosophy to issue a broad-ranging challenge to those philosophical values seen as characteristic of modernism and, specifically, the Enlightenment. While recognizing the risks of reductionism and oversimplification, I will here adopt Jane Flax’s useful summary of the basic tenets of the Enlightenment; she characterizes postmodernism as the attempt to throw these beliefs into radical doubt, and this will allow us to put into relief the contrasting claims of modernism and postmodernism.53 1. The existence of a stable, coherent self. Distinctive properties of this Enlightenment self include a form of reason capable of privileged insight into its own processes and into the “laws of nature.” 2. Reason and its “science”—philosophy—can provide an objective, reliable, and universal foundation for knowledge. 3. The knowledge acquired from the right use of reason will be “true.” 4. Reason itself has transcendental and universal qualities. 5. There are complex connections between reason, autonomy, and freedom. All claims to truth and rightful authority are to be submitted to the tribunal of reason. 6. If claims to authority are grounded in reason, the conflicts between truth, knowledge, and power can be overcome. Truth can serve power without distortion. 7. Science, as the exemplar of the right use of reason, is also the paradigm for all true knowledge. 8. Language is in some sense transparent. There is a correspondence between word and thing (as between a correct truth claim and the real). Objects are not linguistically (or socially) constructed; they are merely made present to consciousness by naming and the right use of language. 53. This list of beliefs, with some of the details omitted, is taken from Jane Flax, “Postmodernism and Gender Relations in Feminist Theory,” in Feminism/Postmodernism, ed. Linda J. Nicholson (New York: Routledge, 1990), 41–42. I would like to thank Professor Mary Klages for bringing this article to my attention. While this list is not intended to capture some “essence” of the postmodern—which would be oxymoronic, in any case—I think it is valuable in helping characterize the basic positions most postmodernists would not find particularly controversial.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

172   Kant and Contemporary Philosophy

Copyright © 2008. Catholic University of America Press. All rights reserved.

By calling these precepts into radical doubt, presumably we are left with something like the following views as characteristic of the postmodern (without, again, attempting any sort of “essential” characterization): 1. There is no stable, coherent self that can assert a privileged status relative to its mental processes or into the “laws of nature.” 2. Philosophy does not serve as a neutral adjudicating “science,” and thus cannot provide an objective, reliable, and universal foundation for knowledge. 3. If there is no necessarily “right” use of reason, the “knowledge” gained from it cannot be said to be “true” in any substantial sense. 4. There are no universal and transcendental qualities attributable to reason. 5. The tribunal of reason does not provide a disinterested neutral way of adjudicating competing claims relative to reason, autonomy, and freedom. 6. Conflicts among truth, knowledge, and power cannot be overcome by reason, and thus reason cannot serve power without distortion. 7. Science provides neither an exemplar of the right use of reason nor a paradigm for all true knowledge. 8. Language is not a transparent vehicle for naming objects or for bringing to consciousness linguistically transparent relations between words and things. From such a list, the standard claims of postmodernism become clear: a rejection of truth and objectivity; skepticism relative to any universal or “essentialist” claims about human beings, nature, and the relationship between them; doubt about the ability of science to provide assertions that are not embedded within social and political contexts that are subject to at least implicit power relations; and a deep suspicion about the ability of philosophy to deliver grand narratives that delineate the conceptual relationships that describe human beings and their place in the universe.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   173 Kant and the Postmodern As with Davidson, it is tempting here to take a rather short argumentative line, and suggest that the postmodern critique of Kant fails to distinguish necessary conditions from sufficient conditions, and further fails to recognize something about which Kant is explicit: namely, that all empirical claims are contingent. On the former point, for instance, we can find John McGowan observing that “postmodern theory has at times questioned the serene Kantian confidence in the powers of reflection by adopting the more troubling view that we are least likely to understand our own motives and conditions of existence.”54 Yet, in addition to the fact that Kant quite explicitly denies that we do understand (or know) our own motives—a crucial strategic aspect of his practical philosophy—it should be noted that there is a significant distinction between identifying necessary conditions for thought and intersubjective communication and taking those conditions to establish sufficient conditions. Only on the latter interpretation could one conclude that “reason will inevitably lead all who employ it properly to the same conclusions.”55 What reflection—taken here to be transcendental reflection, or the determination of a set of rules that must hold for judgments relative to a given domain to be possible—establishes are assertability conditions, as well as conditions for the possibility of disagreement. It should again be emphasized, as well, that Kant is a fallibilist, and nowhere argues that simply because judgments must conform to a specified set of constraints on judgment that those judgments will make the same assertion. Particularly in the case of empirical judgments—those of “ordinary” experience as well as those of natural science—as soon as it is recognized that an ineliminable component of those judgments is passively received sensible information, such judgments are contingent. While they are conditioned a priori by such constraints as space, time, and causality, the content of the individual judgments themselves is not determined a priori, nor is the judgment’s truth (or falsity). The inquiry into the truth (or falsity) of those 54. John McGowan, Postmodernism and Its Critics (Ithaca: Cornell University Press, 1991), 42. 55. Ibid., 34.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

174   Kant and Contemporary Philosophy judgments is an unending empirical investigation into what Kant calls the “thoroughgoing contingency of all natural things” (A562=B590). As we saw in discussing Davidson’s work, logic provides certain necessary constraints for that investigation to be possible, but does not and cannot provide—and certainly cannot do so a priori—agreement on specific empirical claims. To reject those constraints is not just to risk making communication difficult, but to make it impossible, generating a skepticism sufficiently radical to question its coherence.56 While I think that this argumentative line is in fact compelling against the postmodernist, it is also worth looking at some of the specific postmodern criticisms relative to Kant’s argument, which I think will reflect a misunderstanding of it. Thus Jane Flax has suggested that in contrast to the Enlightenment conception of a “stable, coherent” self, the postmodernist challenges this conception and the privileged insight into its own mental processes it brings with it. “What Kant’s self calls its ‘own’ reason and the methods by which reason’s contents become present or self-evident, are no freer from empirical contingency than is the so-called phenomenal self.”57 I have argued earlier that Kant attempts a reflective discovery of the conditions that must hold for judgments to be possible, and that this is ultimately grounded in the combinatory function of the transcendental unity of apperception. There is little doubt that the relationship between that apperceptive unity and the other aspects—for example, phenomenal, noumenal, moral—is horrifyingly complex, if not ultimately incoherent, and it is sufficiently clear from the tortuous footnote at B422 and Kant’s own Reflexionen that he never articulated this relationship to his own satisfaction. At the same time, it is also clear that Kant saw that aspect of the self, represented by the “I think,” or what Kant calls in the Par56. As BonJour notes in The Structure of Empirical Knowledge, 104, “there are certain versions of skepticism which are so deep and thoroughgoing that is impossible to refute them.” I think it is important to distinguish between the postmodern doubts about grand narratives and truth from a skepticism so radical that it is not clear that it could even be formulated or communicated, and even if it could be, what interest it could still retain. At the same time, the rejection of Kant’s conception of logic, which I have been characterizing as a set of minimal constraints on rationality, clearly risks advocating precisely that kind of radical skepticism. 57. Flax, “Postmodernism and Gender Relations in Feminist Theory,” 43.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant and Contemporary Philosophy   175

Copyright © 2008. Catholic University of America Press. All rights reserved.

alogisms “this I or He or It (the Thing) that thinks” (A346=B404), as empty of all content. The key to his argument is summarized at B134: “[O]nly insofar as I can grasp [begreifen] the manifold of representations in one consciousness do I call them one and all my representations; for otherwise I would have a self as many-colored and diverse as I have representations of which I am conscious to myself ” (B134). This is a straightforward argument in modus tollens. Without a logically simple subject grounding a set of diverse representations, it would be impossible for that subject to judge that its different representations were all its representations, and hence it could not combine and unify them into a coherent, rule-ordered experience. Thus, if I don’t have a unifying consciousness, my self is as diverse as its representations. Kant takes it as self-evident here that the subject is not as diverse as its representations;58 therefore it must have some kind of unified, or unifying, consciousness. As I have tried to make clear, the combining role of the transcendental unity of apperception functions as a necessary condition for any judgment to be possible, and it is simply not clear how that self, the qualitative unity of which is presupposed by any judgment, and which lacks all content, can itself be empirically contingent. Genevieve Lloyd I think makes a similar mistake by criticizing Kant’s attitude toward women in terms of “female consciousness.”59 If 58. Kant takes it as self-evident, as I say, here, or for the purposes of this specific argument. But the point is really an outcome of a general theme running throughout the Transcendental Doctrine of Elements, and particularly emphasized in the second edition of the Critique—that the unity of the subject is reflected in the unity of objects. This notion has been clearly outlined in Melnick, Kant’s Analogies of Experience, 144–49, and discussed in detail in Ewing, Kant’s Treatment of Causality, chapter 5, and Michael Washburn, “The Problem of Self-Knowledge and the Evolution of the Critical Philosophy 1781–1787” (dissertation, University of California San Diego, 1970). As Kant himself insists, self-cognition [Selbsterkenntnis] is quite different from self-consciousness [Selbstbewußtsein]; the former requires intuition (time), which is determinable only by its exhibition [Darstellung] in space (B156; cf. A33=B50). This “dialectical” aspect of Kant’s argument seems to be ignored by Frederick Jameson when he writes that Michel Foucault’s work on Kant has shown “the necessity of reversing Kant’s transcendental deduction: it is not the unity of the world that demands to be posited on the basis of the unity of the transcendental subject; rather, the unity or incoherence and fragmentation of the subject—that is, the accessibility of a workable subject position or the absence of one— is itself the correlative of the unity or lack of unity of the outside world.” Postmodernism, or, The Cultural Logic of Late Capitalism (Durham: Duke University Press, 1991), 137. 59. G. Lloyd, The Man of Reason (Minneapolis: University of Minnesota Press,

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

176   Kant and Contemporary Philosophy

Copyright © 2008. Catholic University of America Press. All rights reserved.

referring to the consciousness represented by the “I think,” she commits a category mistake, in that the apperceptive unity, as formal, is detached from any such characteristics. If, on the other hand, she is referring to empirical consciousness, Lloyd has simply changed the subject by neglecting the in-principle contingent aspect of all empirical claims, including those about self-cognition or self-knowledge. We may wish to gloss the apperceptive unity represented by the “I think” in terms of Kant’s conception of it as a “vehicle of all concepts,” or Wittgenstein’s Tractarian “metaphysical subject” that is necessary to see the visual world but is not part of the visual world;60 Hume observes that he never stumbles upon a perception of himself but fails to recognize, as has been noted, that it is this aspect of the self that is doing the stumbling. Such a conception of the self certainly qualifies as “stable and coherent,” but it should be kept in mind that it is without content, and, as such, plays only a logical role. And, in spite of what has sometimes been said, such a self does not provide any sort of “privileged access” to the cognitive functions of that self, which, as we have seen repeatedly, are isolable only by philosophical reflection upon possible thought and possible experience.61 Thus those postmodernist critics of Kant who wish to reject the Kantian conception of the self, qua apperceptive unity, must show that they themselves, in articulating their criticisms, both do not conform to the conditions Kant specifies and, 1981), 70. This very detachment of Kant’s transcendental unity of apperception has been the basis for much of the feminist critique of Kant. It is, of course, the very “emptiness” of the “I think” that allows Kant to present the powerful critique of Descartes’s cogito argument, the focus of the Paralogisms of Pure Reason, and not only are many of the criticisms of Kant better directed against Descartes (see Susan Bordo, The Flight to Objectivity [Albany: State University of New York Press, 1987]), but Kant provides a good bit of the strategy to develop those criticisms. This is not to suggest that Kant does not deserve strong censure for his comments about women; see my “Kant and Feminism.” 60. Wittgenstein, Tractatus Logico-Philosophicus, §5.633. 61. Thus Rorty has attributed to Kant the thesis “that the mind notoria est quam corpus,” and that Kant’s Transcendental Idealism is grounded in the “privileged access” discursive intellects have to their own “mysterious noumenal cognitive faculty”; “Transcendental Arguments, Self-Reference, and Pragmatism,” 79; similarly he attributes “Cartesian privileged access” to Kant in Philosophy and the Mirror of Nature, 138. Rorty is correct to identify in this context such a noumenal cognitive faculty as “mysterious,” and I think we have seen that it would be particularly mysterious to Kant.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant and Contemporary Philosophy   177 at the same time, do not succumb to a skepticism or relativism sufficiently radical to qualify as incoherent.62

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant, Foucault, and Foundationalism To conclude this too-brief discussion of the postmodernist reaction against Kant, I want to look at a small fragment of the work of Michel Foucault, the postmodern thinker whose engagement with Kant is perhaps the most intense, from his early translation and notes on Kant’s Lectures on Anthropology, through The Order of Things, on to the very late essay, the very title of which echoes Kant’s own “What Is Enlightenment?”63 I can hardly hope to do justice to Foucault’s rich confrontation with the works of Kant, and instead will focus on a representative claim from The Order of Things, in order to move to the question of foundationalism and a consideration of what kind of foundationalism can be attributed to Kant. In a consideration of Edmund Husserl, Foucault writes: 62. It is worth noting that Kant considers alternate conceptions—in addition to the better-known conceptions of Descartes and others on whom he focuses his critique in the Paralogisms—of the self, represented by the “I,” such as the comparison to elastic balls that “communicate” through a series of interactions (A363–64), as well as the account of “elanguescence” in discussing Mendelssohn (B414–15, esp. B415 n.b.). While some postmodernists might view the following reaction as mere “logic chopping,” it is fair to ask, from a Kantian perspective, what implicit commitments one makes in writing grammatical prose and logically structured arguments in favor of postmodernism; what is the strength of the commitment, in these cases, to grammar and logic? Some postmodernists—e.g., Jacques Derrida, Julia Kristeva, and Luce Irigaray—have recognized this issue as one worth some attention, and have thus challenged, from a wide variety of directions, precisely those commitments, with word-play, neologisms, puns, jokes, parentheticals, and unconventional grammar, and have introduced various ways of writing phrases, as in Derrida’s indicating the “trace” of a sign by taking it “sous écriture.” 63. E. Kant, Anthropologie du point de vue pragmatique, trans. M. Foucault (Paris: Vrin, 1979); Les mots et les choses (Paris: Gallimard, 1966), translated as The Order of Things, trans. A. Sheridan (New York: Vintage Books, 1973); “What Is Enlightenment?” in P. Rabinow (ed.), The Foucault Reader (New York: Pantheon Books, 1984), based on an unpublished French manuscript. I have followed the current practice of viewing Foucault as a postmodern thinker, although he was—unsurprisingly—suspicious about the very term “postmodernism”: “I do not understand what kind of problem is common to the people we call post-modern or post-structuralist,” in Politics, Philosophy, and Culture: Interviews and Other Writings, 1977–1984, ed. L. Kritzman (New York: Routledge, 1988), 34.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

178   Kant and Contemporary Philosophy

Copyright © 2008. Catholic University of America Press. All rights reserved.

[T]he phenomenological task that Husserl was later to set himself is linked, in its profoundest possibilities and impossibilities, to the destiny of Western philosophy as it was established in the nineteenth century. It is trying, in effect, to anchor the rights and limitations of a formal logic in a reflection of a transcendental type, and also to link transcendental subjectivity to the implicit horizon of empirical contents, which it alone contains the possibility of constituting, maintaining, and opening up by means of infinite explications. But perhaps it does not escape the danger that, even before phenomenology, threatens every dialectical undertaking and causes it to topple over, willy-nilly, into an anthropology. It is probably impossible to give empirical contents transcendental value, or to displace them in the direction of a constituent subjectivity, without giving rise, at least silently to an anthropology—that is to a mode of thought in which the rightful limitations of acquired knowledge (and consequently of all empirical knowledge) are at the same time the concrete form of existence, precisely as they are given in that same empirical knowledge.64

While Foucault here refers to the nineteenth century, I think it is unquestionable that his critique would include Kant, who, as we have seen, attempted to “anchor the rights and limitations of a formal logic in a reflection of a transcendental type.” The concern Foucault expresses here is characteristic of postmodernism’s suspicion about the Enlightenment’s attempt to claim universality and necessity for contingent human truths (and being unable to do, the Enlightenment falls “willy-nilly” into anthropology). It is clear enough that Kant has identified a set of rules—the forms of intuition, the rules of general logic, the rules of transcendental analytic—as absolutely universal and strictly necessary. The question here is whether this topples into anthropology, and surreptitiously takes anthropological conventions (which may indeed be widespread and of longstanding acceptance) to be necessary constraints on thought, experience, and behavior, which Foucault sees as an illegitimate inference (and which Kant might well describe as committing a fallacy of subreption).65 64. The Order of Things, 248. I would like to thank Brad Hume for insisting on the significance of this text for understanding the relationship between Kant and Foucault. 65. Indeed, given that Kant’s logical focus is on the quaestio juris, one might see

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   179 Before we turn to the question of whether Foucault’s interpretation forces us to see Kant as imposing the kind of foundationalism and “grand narrative” that troubles so many postmodernists, including Foucault, some specific remarks on this passage are in order. Foucault suggests that those working in the tradition he describes are attempting to give empirical contents “transcendental value.” But, as we have seen repeatedly, Kant consistently uses “transcendental,” in characterizing his philosophical strategy, to identify conditions that are necessary relative to possible judgments within a given domain. Thus it is inaccurate to suggest that Kant is doing what Foucault attributes to him, if we accept that Kant works in this tradition. Kant is quite explicit in emphasizing that empirical content is understandable only by being judged; what is “transcendental” here, then, is not empirical content, but the conditions for making empirical judgments (as well as mathematical judgments and, in general logic, judgments simpliciter). While Foucault’s attribution here may be ambiguous in terms of what he means precisely by “transcendental value,” a straightforward reading would suggest that Kant does not claim anything but that “making” or “giving” empirical contents transcendental value is impossible. Hence, this cannot be a criticism against Kant, in that he agrees with the basic point. Nor does Kant suggest that transcendental subjectivity “constitutes,” in any strong sense, the possibility of empirical contents; again, he is quite explicit in his empirical realism, insisting on the ineliminable sensible component of that which is received within space and time for making empirical judgments. Hence the danger Foucault sees confronting the tradition he describes—of “toppling” into anthropology—conflates two distinct issues. Insofar as human beings satisfy certain minimal constraints that must hold for judgments to be possible, those constraints are characteristic, obviously enough, of human beings and, in Foucault’s genealogical investigations as focusing on the distinct quaestio facti. Kant does not, in principle, reject the latter as without interest; his point, however, would be that historical or empirical investigations—such as Locke’s account of the mind, which he characterizes as “physiological”—simply cannot address questions of legitimacy or justification. Because the two questions are conceptually distinct, the responses address distinct issues, and frequently may well not even conflict. Thus Foucault and Kant may ultimately be talking past each other, relative to the debate over necessity and convention.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

180   Kant and Contemporary Philosophy

Copyright © 2008. Catholic University of America Press. All rights reserved.

that sense, would provide some material for a philosophical anthropology.66 At the same time, as shown above, Kant regards these constraints to hold for any agent who can employ the first-person pronoun, can use concepts to judge, and can regard itself as free; that is, an agent who makes judgments (and thus does not simply “react” to external stimuli), and judgments that can be true or false (and thus not the kind of divine understanding who possesses intellectual intuition and so does not in this sense judge at all). In sum, human beings must satisfy these constraints in order to be able to make judgments; clearly Kant has human beings in mind in outlining the concepts and rules that constitute Transcendental Aesthetic and Transcendental Analytic, and human beings are the only agents who we can currently be confident do satisfy them.67 At the same time, the set of possible agents is not exhausted by human beings, and those constraints, therefore, cannot be identified as human constraints. In what sense, then, do such conceptual or structural constraints function to provide “foundations”? In the postmodernist attack on the Enlightenment’s alleged commitment to a universal and ahistorical framework, with its attendant “grand narrative”—“there is no foundational principle, no originating or final cause”68—Kant is certainly regarded as not simply representative of that Enlightenment tradition, but quite possibly its standard-bearer. Thus he has been seen as provid66. It is perhaps worth remembering here that the Lectures on Anthropology carry the subtitle “from a pragmatic standpoint” (i.e., not from a critical standpoint). The Lectures on Anthropology may be found in vol. 7 of the Academy edition, and have been translated by Mary Gregor (The Hague: Martinus Nijhoff, 1974). These lectures were published in 1797, although Kant had been giving them “for some thirty years,” as Gregor points out in her translation (ix), or from a period well before Kant’s “Critical” turn to the mature Transcendental Idealism of, e.g., the Critique of Pure Reason. 67. To be sure, Kant makes some remarks about non-human animals that are very troubling and contentious (although perhaps a bit less noxious than those made by Descartes, among others). Kant is clear that he does not identify the constraints imposed by the Critical philosophy as relative only to human beings when he writes such things as “The human mind (as, I believe, must necessarily be the case with every rational being [jedem vernüftigen Wesen]) takes a natural interest in morality” (A830=B858n.). As mentioned earlier, it is important to distinguish the attribution of agency to another and what follows from that attribution. For Kant, the former is contingent; only the conditions that follow can be seen as holding a priori. 68. Thomas Flynn, “Foucault’s Mapping of History,” in The Cambridge Companion to Foucault, ed. Gary Gutting (Cambridge: Cambridge University Press, 1994), 39.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant and Contemporary Philosophy   181

Copyright © 2008. Catholic University of America Press. All rights reserved.

ing precisely the kind of framework that Rorty characterizes as “a set of rules which will tell us how rational agreement can be reached on what would settle the issue on every point where statements seem to conflict.”69 Yet Kant’s foundationalism—assuming such a term is appropriate—not only must be distinguished from the research program in contemporary (analytic) epistemology of the same name,70 but also must be distinguished from the kind of foundationalism postmodernists have attributed to their opponents, from Descartes on through the tradition of modern philosophy, culminating in Kant. Kant certainly does not claim any sort of indubitability about empirical claims, as some epistemologists—including some quite sympathetic to Kant, such as C. I. Lewis—have attempted to do in terms of qualia, raw feels, immediate apprehension of phenomenological content, sensa, selfjustifying beliefs, or immediate perceptual reports, with the exception, perhaps, of the extremely attenuated sense that the “I think” is an “indeterminate empirical intuition” (B422n).71 For, as Kant insists, experience only “teaches us that a thing is so and so, but not that it cannot be otherwise” (B3). At the same time, Kant’s strategy is radically distinct from one such as Descartes’s: not only does it not incorporate any sort of theological guarantee for what Descartes considers “clear 69. Rorty, Philosophy and the Mirror of Nature, 316. Here again, I do not want to take up the question of whether Rorty qualifies as a “postmodernist.” 70. As William Alston has noted, “Contemporary writers on foundationalism do not seem to notice that Descartes and Locke have a quite different view of knowledge and, hence, that if they hold that knowledge rests on foundations, this will mean something rather different.” “Two Types of Foundationalism” in Epistemic Justification (Ithaca: Cornell University Press, 1989), 19n1. Ernest Sosa describes the difference by distinguishing “formal” from “substantive” foundationalism in “The Raft and the Pyramid,” in Empirical Knowledge: Readings in Contemporary Epistemology, ed. P. Moser (Totowa: Rowman and Littlefield, 1986). David Tarbet has argued that Kant’s foundational imagery “has little to do with the Critique’s arguments” in “The Fabric of Metaphor in Kant’s Critique of Pure Reason,” Journal of the History of Philosophy 6 (1968): 270; as we will see, while Tarbet illuminates Kant’s military and jurisprudential metaphors, I think Kant’s foundations metaphor is also crucial to understand his general project in terms of what Kant believes philosophy both can and cannot establish. 71. For an example, see Paul Moser’s “A Defense of Epistemic Intuitionism,” Metaphilosophy 15 (1984): 205: “[W]e have good reasons to believe that the immediate apprehension of phenomenological content does not require conceptualization.” As I have tried to indicate above, I think Sellars (following Kant) has shown that any such attempt risks succumbing to the “myth of the given.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

182   Kant and Contemporary Philosophy and distinct ideas,” it explicitly rejects any sort of privileged access to mental content, and, at best, shows how to justify those conditions that we identify, reflectively, as necessary for the possibility of experience. In what sense, then, is Kant a “foundationalist”? Kant himself uses the notion of foundations in introducing the Critique: “Now it may seem natural that as soon as one has abandoned the ground of experience, one would not immediately erect an edifice with cognitions that one possesses without knowing whence, and on the credit of principles whose origin one does not know, without having first assured oneself of its foundations [Grundlegung] through careful investigations” (A=B7). That is, this procedure (reminiscent of that for which Descartes argues in the “First Meditation”) seems natural if “natural” means what we ought to do, rather than what we in fact do. For it is “the common fate of human reason” to rush headlong into the construction of elaborate speculative edifices, only later considering “whether the foundations are reliable [ob auch der Grund dazu gelegt sei]” (A5=B9). The “positive” dimension of Kant’s epistemology yields, by an investigation into experience, the specification and justification of foundations—presented by Transcendental Aesthetic and Transcendental Analytic—with the consequence that no meaningful claims, qua judgments, can be made beyond the limits of possible experience. Such a position qualifies as an immoderate, or even radical, foundationalism, in that the concepts, principles, and rules it delineates are characterizable as irrevisable, incorrigible, indubitable, infallible, and certain. Kant, of course, describes them as absolutely universal and strictly necessary. To return to the basic position I have been arguing for, Kant is a foundationalist in the sense that he provides necessary and universal conditions for the possibility of making judgments, relative to a given domain. Consequently, within the domain of experience, it hardly suggests, as Rorty indicates, that this foundationalist structure will “settle” every conflict; Kant’s general critical discussion of the employment of the understanding (and reason) does not eliminate further inquiry into knowledge claims. Rather, it grounds and secures the necessary conditions, and limits to, that inquiry. Even with a structure of synthetic rules for the understanding, employed a priori, empirical state-

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant and Contemporary Philosophy   183

Copyright © 2008. Catholic University of America Press. All rights reserved.

ments do conflict, and will continue to conflict. What cannot conflict are the principles that condition experience—the general presuppositions that make experience possible. Conflicting empirical claims are not resolved by a pronouncement forthcoming from Transcendental Analytic. Rather, we must turn to the “fruitful bathos of experience,” where we don’t find the glittering structures dreamed of in dogmatic rationalism, although we do find sufficient shelter from the skepticism Kant sees as the consequence of an uncritical empiricism. This is the payoff in what I have characterizing as Kant’s “logical” strategy. Just as no one would expect formal logic to determine whether a given claim “p” is true or false, or expect a grammar to determine whether a given sentence is true or false, Kant does not suggest that, given an empirical claim “p,” we can point to specific criteria that make “p” true (beyond the nominal definition of truth discussed at A58=B82); indeed, very little Kant says in the Transcendental Analytic has to do with such (conjectured) criteria. Rather, Transcendental Analytic is a “logic of truth”; Kant’s concern there is to specify the intellectual conditions that must hold if “p” is to be a truth-evaluable empirical claim. What he seeks to justify is a conceptual framework or logic of experience, which grounds our investigations into specific empirical claims.72 In contrast to this logical strategy, Foucault urges, along Nietzschean lines, a genealogical approach: [C]riticism is no longer going to be practiced in the search for formal structures with universal value, but rather a historical investigation into the events that have led us to constitute ourselves and to recognize ourselves as subjects of what we are doing, thinking, saying. In that sense, this criticism is not transcendental, and its goal is not that of making a metaphysics possible: it is genealogical in its design and archeological in its method.73

But to return to an earlier point, on the reading of Kant proposed here, it becomes increasingly unclear how one proceeds genealogically, with72. Judgments of pure mathematics, as constructions in pure intuition, are distinct in terms of both the ontological and the alethic issues involved. It should also be noted that the regulative employment of the ideas of reason plays a crucial role in empirical inquiry; see Susan Neiman, The Unity of Reason (Oxford: Oxford University Press, 1997). 73. Foucault, “What Is Enlightenment?,” 45–46.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

184   Kant and Contemporary Philosophy out presupposing precisely the kind of framework Kant articulates. To not put too fine a point on it, can one engage in a genealogical investigation without presupposing some sort of general spatio-temporal framework,74 some relatively rich (possibly richer than Kant’s own) conception of causal interaction, and some sufficiently informative notion of “event”? Indeed, if, as it has been argued, the nervus probandi of the Second Analogy is to distinguish a “state of affairs” from “event,” the fundamental idea that motivates historical cum genealogical inquiry would have to accept Kant’s result of that argument. To be sure, it is stated as a rule to be presupposed a priori: the concept of cause “makes strict demand that something A, should be such that something else, B, follows from it necessarily and in accordance with [nach] an absolutely universal rule” (A91=B124; Kant’s emphasis). But, as Kant freely admits about specific causal interactions, which would presumably include the kinds of historical events, our understanding of which Foucault has done so much to illuminate, “How anything in general can be altered, how it should be possible that upon a state in one point of time an opposite state may follow in the next moment—of these we have a priori not the least concept. For this acquaintance [Kenntnis] with actual forces is required, which can only be given empirically” (A206–7 =B252; my emphases). And if the reading of the Transcendental Analytic that has been proposed here is at all plausible, all Kant establishes as a priori in any rich sense is a relatively modest framework, analogous to general logic, that allows us to make judgments about causally interacting substances in space-time. While the results are couched in the strongest language possible, as absolutely universal and strictly necessary conditions of possible experience, we will see in the following chapter precisely how modest those results are. The burden, then, seems to be on the postmodernist to reject those results and maintain some ability to provide informative analyses that are not subject to a sufficiently radical skepticism, or relativism, that deprives them of the coherence required to make them informative. 74. Flynn argues that Foucault shifts from time to space “as the paradigm guiding his approach to historical topics”; “Foucault’s Mapping of History,” 41. In the context of Kant’s discussions of space and time (and, in contemporary terms, space-time), I think this shift must be seen as metaphorical.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Contemporary Philosophy   185 As has been noted by John McGowan, Christopher Norris, and others, late in Foucault’s career, as in his essay “What Is Enlightenment?” Foucault is much more sympathetic to Kant’s strategy and goals than in his earlier work (such as The Order of Things).75 The difficulty Foucault encountered was what Norris refers to as “poststructuralism and kindred forms of modish ultra-relativist doctrine.”76 Briefly put, the kinds of political and cultural critique that engaged Foucault throughout his career become problematic within a context of contextualism that is so thoroughgoing as to be indistinguishable from a radical relativism. To cite Norris again: “[T]he truth-values of enlightened thought—of reason in its jointly epistemological and ethico-political modes—cannot be abandoned without at the same time renouncing any claim to promote or articulate the interests of justice, autonomy, and human emancipation.”77 To be sure, the postmodern critique of the views conceived as constitutive of the Enlightenment has had, among its more salutary effects, the result that many of the standards put forth as “eternal” and “universal” could be seen as providing a convenient and apologetic mask for a privileging of power, generally the power of the white, the male, and the wealthy. Yet the consequences of a consistent critique of the values of the Enlightenment may well lead us down another slippery slope. For when all such Enlightenment notions are abandoned, we are in the position Camus has eloquently described as “confusion,” resulting in a simple failure to take responsibility: as he puts it, “Such ..... confusion is made up of cowardice rather than generosity and eventually justifies whatever is worst in this world.”78 To abandon our intellectual obligations here— to reject, in some sense, Camus’s notion of responsibility—is, as Louise Antony has argued, to reduce the debate to one about power. For 75. McGowan, for instance, refers to “Foucault’s reconciliation” with Kant in this late essay [“What Is Enlightenment?”], a reconciliation that contrasts strongly with the submerged debate with Kant that informs the much more hostile description of modernity found in The Order of Things; Postmodernism and Its Critics, 124n22. Christopher Norris argues for a similar view in “‘What Is Enlightenment?’” The Cambridge Companion to Foucault, 159–96. 76. Norris, “’What Is Enlightenment?’”185. 77. Ibid., 184. 78. A. Camus, “Reflections on the Guillotine” in Resistance, Rebellion, and Death, trans. J. O’Brien (New York: Knopf, 1961); 231.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

186   Kant and Contemporary Philosophy those who hope to speak for those who have traditionally been, and continue to be, oppressed, such a reduction cannot be to the advantage of those out of power: in short, as Antony eloquently puts it, such tactics, “given current distributions of power,” are “really stupid.”79 Furthermore, it begins to become quite unclear to what—if any—standards one appeals when one insists on the merits of a given view, and the mystery only grows with the strength of that insistence. It is not clear to me that Foucault was ever successful in reconciling his genealogical methodology with his concomitant concern about abandoning the Enlightenment or modernist insistence on necessity, truth, and the just.80 But, as Roy Boyne has observed in his discussion of the shifts in Foucault’s thought, by the time of the last two (posthumous) volumes of The History of Sexuality, we can see in Foucault’s thought the recognition that a “self ruled by the desires is unfree.” Furthermore, “the exercise of self-mastery is closely connected to the state of freedom.”81 Here, of course, we return to the earlier discussion of self-legislation, and Rousseau’s dictum that genuine freedom is manifest in obeying those laws we impose upon ourselves. When this approach is taken up by Kant, we see revealed precisely his transcendental strategy, where the thinking, judging, and acting subject discovers, reflectively, the conditions that must hold for thought, judgment, and action to be possible.82 Yet, as has been argued here, that radically subjective process reveals such conditions to be universal and necessary for any agent to whom we wish to attribute even an impoverished sense of rationality. 79. L. Antony, “Quine as Feminist: The Radical Import of Naturalized Epistemology,” in A Mind of One’s Own, ed. C. Witt and L. Antony (Boulder: Westview Press, 1993), 220n12. 80. One might well be excused for failing to discover in much postmodern writing on Kant the potential for a liberatory (albeit ultimately bourgeois) politics, although others have seen this, including the Marburg neo-Kantian (and socialist) Hermann Cohen, Karl Vorländer, in adumbrating a number of Kantian-inspired views in his Kant und Marx (Tübingen: J. C. B. Mohr [Paul Siebeck], 1926), and Lucien Goldmann in his Introduction à la philosophie de Kant (Paris: Gallimard, 1967), translated by Robert Black as Immanuel Kant (London: NLB, 1971). 81. Roy Boyne, Foucault and Derrida: The Other Side of Reason (New York: Routledge, 1996), 144. 82. Presumably, Kant would emphasize here acting correctly or morally. It should be noted that the vast majority of postmodern writing on Kant focuses on his moral philosophy, rather than the epistemology and metaphysics as I have done here.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Kant and Contemporary Philosophy   187 We thus discover a priori conditions imposed subjectively, yet holding objectively, grounding freedom for that subject and attributable to any subject. As Lucien Goldmann has remarked, “[I]f reason will not subject itself to the law it gives itself, it will have to bow under the yoke of laws which others impose on it, for without any law whatsoever nothing, not even the greatest nonsense, can play its hand very long.”83 The challenge, then, posed by Kant to postmodernism is whether a self can be constructed “all way down.” If it can be, must we then abandon any strong sense of necessity? And does this not ultimately lead to an incoherent relativism and skepticism, thus grounding precisely the kind of untenable—and, ultimately on such a view, uncriticizable—political, cultural, and ideological manifestations of power that postmodernism itself seeks to challenge, if not reject?

Copyright © 2008. Catholic University of America Press. All rights reserved.

83. Goldmann, Introduction à la philosophie de Kant, 120. From a different direction, but with much the same result, Stephen Pinker has argued that “radical empiricism is not necessarily a progressive, humanitarian doctrine. A blank slate is a dictator’s dream,” The Language Instinct, 427.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Ch a pter Six

The Modesty of the Critical Philosophy

Copyright © 2008. Catholic University of America Press. All rights reserved.

e

It is not uncommon in introductory philosophy courses to receive a picture of Kant that paints him as a paradigmatic old-fashioned, detached Prussian scholar, whose walks were so predictable that the residents of Königsberg could set their clocks by them, and whose life passed as the most regular of regular verbs. Such an image is reinforced by the overwhelming attention to what might be called the “positive” dimension of Kant’s philosophy, for example, the Transcendental Analytic. There we find obscure arguments couched in arcane, inscrutable terminology, as well as intricate, elaborate proofs of synthetic a priori concepts and principles. All this makes it easy to neglect the perspective from which Kant viewed things—“Our age is, in especial degree, the age of criticism, and to criticism everything must submit” (Axi n.)1—and to forget the once-famous epithet Mendelssohn applied to Kant: der alles zermalmende, “the destroyer of all.” By adhering to this traditional picture, it is also easy to overlook 1. Kant gives some of the historical background to this claim in the Jäsche Logik (Ak. IX, 27ff.); some of its political implications are developed in the essay “What Is

188

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Modesty of the Critical Philosophy   189 Kant’s own view of the Critique of Pure Reason. To conclude this study, I want to look briefly at some of the things Kant himself says about the purposes of the Critique. As we saw earlier, the “positive” dimension is presented in the Transcendental Aesthetic and Transcendental Analytic; the remainder of the Doctrine of Elements, where Kant presents argument after argument against philosophical positions staked out by both rationalists and empiricists, is negative. This critical strategy reaches its zenith in the Antinomies, where Kant employs a skeptical method to show that in some cases, both rationalists and empiricists are making claims that are demonstrably false; in other cases, by failing to make crucial critical distinctions, they fail to recognize that presumably opposed claims are both demonstrably true. Furthermore, a great deal of the Doctrine of Method is critical as well, constantly reminding the reader of the necessity to discipline the employment of pure reason. In sum, the structure of the Critique reflects its title as a critique of pure reason, a feature often neglected by the overwhelming emphasis on the Transcendental Analytic—specifically the Transcendental Deduction—found in the literature. Indeed, this emphasis may be a result of just how successful the arguments of the Transcendental Dialectic are regarded. But this should not obscure the fact that the main thrust of the Critique is negative. And this is precisely what Kant says: “On a cursory overview of the present work it may seem that its usefulness is merely negative, teaching us that we must never venture with speculative reason beyond the limits of experience. Such in fact is its primary [erster] use” (Bxxiv; Kant’s emphasis). Elsewhere, Kant tells us that the critical investigation of pure reason “should be called a critique, not a doctrine, of pure reason. Its utility, in speculation, ought properly to be only negative, not to amplify, but only to purify our reason, and to keep it free from Enlightenment?” Ak. VIII, 35–42, esp. 40. It is worth comparing Foucault’s remark in his essay of the same name: “I have been seeking to stress that the thread that may connect us with the Enlightenment is not faithfulness to doctrinal elements, but rather the permanent reactivation of an attitude—that is, of a philosophical ethos that could be described as a permanent critique of our historical era.” “What Is Enlightenment?” 42. As I have tried to argue, that attitude ultimately cannot be inculcated without adopting some aspect of these “doctrinal elements,” as characterized in terms of Kant’s transcendental strategy establishing conditions that hold a priori.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

190   Modesty of the Critical Philosophy errors—which is already to gain a great deal” (A11=B25).2 Much later, Kant returns to make the same point: “[W]here the limits of our possible cognition are very narrow, where the temptation to judge is great, where the illusion that confronts us is very deceptive and the harm that results from the error is very serious, there the negative in instruction, which serves solely to guard us from errors, has even more importance than many a piece of positive information by which our cognition is augmented” (A709=B737; Kant’s emphasis). In general, “The greatest and perhaps the sole utility of all philosophy of pure reason is therefore only negative; for it serves not as an organon for the expansion but as a discipline for the limitation of pure reason, and, instead of discovering truth, has only the modest merit of guarding us against errors” (A795=B823).3 I think it is a serious mistake to underestimate the negative force of Kant’s project. This is not, at the same time, to deny the positive results of Kant’s analysis of human knowledge; nor is it to ignore Kant’s fundamental notion that the limitation of speculative reason is necessary to ground the employment of practical reason. I am simply maintaining that the claims Kant makes about the universal and necessary conditions that the subject contributes in objective cognition must be looked at within this more general critical context, underscoring the innocuous nature of these modest claims, in spite of Kant’s reputation otherwise. It is for such reasons that I think Kant can be placed firmly within a tendency in philosophy that might be called “deflationary.” As Robert Gibbs has characterized this tendency: In our time, Philosophy’s humility is announced by claims that we don’t know. We do not know what humanity is. We do not know God. We do not know what our world should be. We do not even know what philosophy is. If we once had to learn to ask, “How can we know?,” when capital “P” Philosophy was still proud, then in our time the question we are learning is “How can we not know?” But this is to lapse back into the very er2. Kant adds “in speculation” [“in Ansehung der Spekulation”] for the second edition. 3. See also Kant’s comment “I shall not boast here of the service which philosophy has done to human reason through the laborious efforts of its criticism, granting even that in the end it should turn out to be merely negative” (A831=B859).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Modesty of the Critical Philosophy   191 ror of proud Philosophy. For today’s question truly should be “How can we know—that we do not know?” Our humility, our professions of ignorance, may after all be too confident—the confidence of skepticism. The ignorance of philosophy itself requires a critical turn.4

A major portion of the responsibility for the tendency described here has been given to Wittgenstein. It was, of course, Wittgenstein who said that “philosophical problems arise when language goes on holiday.” In his later work, Wittgenstein tried to show that

Copyright © 2008. Catholic University of America Press. All rights reserved.

1) The problems arising though a misinterpretation of our forms of language have the character of depth. They are deep disquietudes; their roots are as deep in us as the forms of our language and their significance is as great as the importance of our language. 2) The real discovery is the one that makes me capable of stopping doing philosophy when I want to.—The one that gives philosophy peace, so that it is no longer tormented by questions which bring itself into question.

In general, Wittgenstein concludes that philosophy “leaves everything as it is.”5 Without getting into exegetical and interpretative issues about the relation between Wittgenstein’s later and earlier work, there are important similarities between these points cited and the strategy Wittgenstein adopts in the Tractatus. In the latter, Wittgenstein is much more optimistic about the possibility of articulating the limits of thought, beyond which we must remain silent. But the attempt to say something about those limits can involve a need to say, or attempt to say, something from beyond them. In a sense, Wittgenstein recognizes this: “Thus the aim of the book is to draw a limit to thought, or rather—not to thought, but the expression of thoughts: for in order to be able to draw a limit to thought, we should have find both sides of the limit thinkable (i.e. we should have to be able to think what cannot be said).”6 But, in another sense, in working out this limit, Wittgenstein is preoccupied with the difficulties inherent in the task. In trying to solve 4. R. Gibbs, “The Limits of Thought: Rosenzweig, Schelling, and Cohen,” Zeitschrift für Philosophische Forschung 43 (1989): 618. 5. These citations are from the Philosophical Investigations, I.38, I.111, I.133, I.124, respectively. 6. Tractatus, “Preface.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

192   Modesty of the Critical Philosophy them, he continues to both recognize this limit and try to go beyond it. Thus (in an image used by Sextus Empiricus), Wittgenstein concludes: “My propositions serve as elucidations in the following way: anyone who understands me eventually recognizes them as nonsensical, when he has used them—as steps—to climb up beyond them. (He must, so to speak, throw away the ladder after he has climbed up it.)”7 Both the later and earlier works acknowledge the depth of philosophical problems and that their solution lies in no longer attempting to ask questions where none can be meaningful and, therefore, responded to. Undoubtedly, the differences between how these notions are worked out in the Philosophical Investigations and the Tractatus are profound—the former providing “therapy” for an illness (or pathology), the latter seeking to establish a picture of the world by reflecting on the necessities involved in its logical expression. But both are interested in resolving certain philosophical difficulties by forcing us to recognize that they are deeply grounded in the desire to make claims that, ultimately, can be nothing but nonsense. I think Kant shares this basic goal with the later Wittgenstein, “to pass from a piece of disguised nonsense to something that is patent nonsense.”8 The connections between Kant and Wittgenstein have hardly gone unnoticed, with much of the attention being focused on Schopenhauer’s influence on Wittgenstein and the similarity between the transcendental unity of apperception and the claims of the Tractatus concerning the metaphysical subject.9 I have no intention of arguing that 7. Tractatus, § 6.54. “And again, just as it is not impossible for the man who has ascended to a high place by a ladder to overturn the ladder with his foot after his ascent, so also it is not unlikely that the Skeptic after he has arrived at the demonstration of his thesis by means of the argument proving the non-existence of proof, as it were by a step-ladder, should then abolish this very argument.” Sextus Empiricus, Against the Logicians II.481; trans. R. G. Bury (Cambridge: Harvard University Press, 1935). 8. Philosophical Investigations, I.464. 9. See Tractatus, §§ 5.641, 5.632, 5.633. For some discussion on this latter point, see W. H. Walsh, “Philosophy and Psychology in Kant’s Critique,” Kant Studien 57 (1960): 189–95; Bennett, Kant’s Dialectic, 69ff.; Allison, Kant’s Transcendental Idealism, 290–91; Thompson, “On A Priori Truth,” 475–76. More general discussions of the relationship are Erik Stenius, Wittgenstein’s Tractatus (Oxford: Blackwell, 1960), chapter 11; P. M. S. Hacker, Insight and Illusion, rev. ed. (New York: Oxford, 1986), chapter 7, part 4; Leslie Stevenson, The Metaphysics of Experience (Oxford: Clarendon Press, 1982); Jonathan

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Modesty of the Critical Philosophy   193 Wittgenstein—in either the early or late incarnation—was doing what Kant was doing in the Critique. But taking Wittgenstein’s general project as dispelling illusions that an uncritical employment of thought and language generate, I think it is clear that Kant’s project is similarly therapeutic. Thus, when Gibbs writes that “since Hegel announced his system, philosophy has been in a sort of joyous retreat,” I would suggest that much of this retreat stems from the Critique, a retreat from dogmatic metaphysics (and/or an uncritical commitment to empiricism). At the same time, such retreat obviously does not, for Kant, entail an embrace of skepticism, relativism, or indifferentism. By a critical investigation of reason, we establish its scope and limits, securing what we can by making legitimate reason’s justifiable claims, “even if, as a result, much that is comprised in our original aims, adopted without deliberation, may have to be abandoned as futile” (Bvii). As I have tried to show, particularly in reference to Transcendental Analytic and the role of the understanding, we arrive with Kant at a reflective discovery and articulation of absolutely universal and strictly necessary concepts and principles, providing “immoderate” foundations for objective experience to be possible. But the nature of the claims made by these concepts and principles, and the requirement that we supplement the understanding with the regulative employment of reason, reveals the modesty of these foundations. Such an immoderate yet modest foundationalism brings with it two important results: (1) the attempt to make cognitive claims outside the scope of experience—claims that violate or transcend the canonical logic of experience—invariably ends in nonsense, and (2) the traditional topics of metaphysica specialis—God, Freedom, and Immortality—lie beyond the domain of possible experience, and must be dealt with otherwise than cognitively: namely, by reason in its practical employment.10 Lear, “Leaving the World Alone,” Journal of Philosophy 79 (1982): 382–92; S. Morris Engel, “Wittgenstein and Kant,” Philosophy and Phenomenological Research 20 (1970): 483–513; Meredith Williams, “Wittgenstein, Kant, and the ‘Metaphysics of Experience,’” Kant Studien 81 (1990): 69–88; and my own “Was Wittgenstein a Neo-Kantian? A Response to Professor Haller,” Grazer Philosophische Studien 45 (1993). 10. In the tradition Kant works, mapped out by such predecessors as Clauberg, Geulincx, and especially Wolff, metaphysics is systematically divided into distinct parts. Metaphysica generalis has as its province “being qua being,” and the study of “ens

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

194   Modesty of the Critical Philosophy

Copyright © 2008. Catholic University of America Press. All rights reserved.

It is with this first result that we find the close similarity between Wittgenstein’s and Kant’s therapeutic strategies. Wittgenstein, famously, wants to “show the fly the way out of the fly-bottle,” and by dispelling the bewitchment by which language entrances us, reveals the emptiness of that picture which holds us captive. This brings philosophy peace, because then what “we are destroying is nothing but houses of cards and we are clearing up the ground of language on which they stand.”11 Kant’s goal is peace as well, and “In the absence of this critique reason is, as it were, in the state of nature, and it can establish and make secure its assertions only through war” (A751=B779; Kant’s emphasis). The uncritical positions outlined in the Antinomies are engaged in such battles; the way to philosophical peace is to demonstrate that the arguments are themselves mere “houses of cards”: “[T]he very fact of their being able to refute one another is evidence that they are really disputing about nothing, and that a certain transcendental illusion has mocked them with a reality [Wirklichkeit] where none is to be found” (A501=B529–30). But in spite of Mendelssohn’s derisive epithet, Kant did not want to destroy everything; this brings us to the second result. By showing what theoretical reason can establish, and just as importantly what it cannot, Kant makes possible the move to practical reason—again, a positive benefit from a negative result: when all progress in the field of the supersensible has thus been denied to speculative reason, it is still open to us to enquire whether, in the practical cognition of reason, data may not be found sufficient to determine reason’s in genere,” the traditional Aristotelian conception of “First Philosophy.” Importantly, Kant argues that the method of such investigations must be transformed, and that “the proud name of an Ontology that presumptuously claims to supply ..... synthetic a priori cognition of things ..... must give place to the modest title of a mere Analytic of pure understanding” (A247=B303). Metaphysica specialis treats the specific questions concerning (rational) theology, cosmology, and psychology. A brief sketch of the history of this division, and of its influence on Kant, is de Vleeschauwer’s “Wie ich jetzt die Kritik der reinen Vernunft entwicklungsgeschichtlich lese,” Kant Studien 54 (1963): 351–68. As de Vleeschauwer observes (353), “metaphysics as a special discipline at the time of Wolff and Kant was still very young.” 11. Philosophical Investigations, I.309; cf. Culture and Value, trans. Peter Winch (Chicago: University of Chicago Press, 1980), 43: “Thoughts that are at peace. That’s what someone who philosophizes yearns for.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Modesty of the Critical Philosophy   195 transcendent rational concept of the unconditioned, and so to enable us, in accordance with the wishes of metaphysics, and by means of cognition that is possible a priori, though only from a practical standpoint, to pass beyond the limits of all possible experience (Bxxi; my emphasis).

In sum, we have secure foundations that necessarily and universally condition all possible experience, and that reveal the emptiness of the attempts to go beyond the field of possible experience in making claims to objective cognition. Such foundations, furthermore, make minimal claims about knowledge—the synthetic concepts and principles, employed a priori, that constitute such foundations are empty without sensibility, and must be organized and unified through the regulative employment of reason. Finally, by clearly marking out the extent and limits of experience, we show that the deepest interests of humanity are preserved, but must look to reason in its practical employment for satisfaction. All of these results, both negative and positive, can be regarded as following from the therapeutic cleansing provided by a thoroughgoing critique of pure reason.12

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Epistemological Modesty If, has been argued above, Kant puts forth a doctrine that can be seen as providing a set of rules as necessary for possible experience, this set qualifies as well as a foundational set of claims, or as providing a priori a conceptual framework for the possibility of experience. At the same time, I have argued that the strength of those philosophical results should be regarded within the context of the negative thrust of the Critique of Pure Reason and that these results are intended to make possible judgments of experience—just as general logic identifies the conditions that must hold for thought to be possible—but offer nothing insofar as the truth or falsity of specific empirical claims is under 12. Compare the Jäsche Logik, Ak. IX, 17: “We would then have two parts of logic: the analytic, containing the formal criteria of truth, and the dialectic, containing the characteristics and rules by which we can tell that something does not satisfy the formal criteria of truth, although it appears to. Dialectic in this sense would have its good use then as a cathartic [Katharticon] of the understanding” (my emphasis); here again, what Kant says about general logic is equally characteristic of transcendental logic.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

196   Modesty of the Critical Philosophy consideration. The latter reveals Kant’s conception of what his critical idealism establishes, relative to cognition, as modest. In contrast, Paul Guyer has urged that Kant in fact puts forth a systematically dogmatic approach, and I will here take up a part of Guyer’s interpretation in order to deflect that criticism and thus reinforce what I have been calling a “modest” interpretation of Kant’s critical project. In Guyer’s Kant and the Claims of Knowledge, he introduces his lengthy discussion of Kant’s transcendental idealism with a relatively brief consideration of the thing-in-itself. It is to Guyer’s credit that he takes on as his opponent not the naïve phenomenalist interpretation of Kant, but the much stronger competition of more recent Kant scholarship, singling out for specific attention the interpretation of Henry Allison. As Guyer characterizes Allison’s reading, “Kant does not advocate an ontological duplication of realms of objects but a conceptual or semantical division: not two sets of objects, but two ways of thinking of or describing one and the same sets of objects.”13 This “twofold” view of the object Guyer associates with an attribution to Kant of “epistemological modesty.” In contrast, Guyer contends that “[t]ranscendental idealism is not a skeptical reminder that we cannot be sure that things as they are in themselves are also as we represent them to be; it is a harshly dogmatic insistence that we can be quite sure that things as they are in themselves cannot be as we represent them to be.”14 The crucial issue for Guyer is the claim that the “modest” interpretation of the thing-in-itself cannot support Kant’s claim, not just that space and time are conditions of our sensibility, but the stronger claim that we can know that things-in-themselves cannot be spatial and temporal. As evidence of this position—that which he calls “harshly dogmatic”—Guyer cites A26=B42: “[S]pace represents no property whatever of any things in themselves, nor does it represent them in their relation to one another, that is, any determination which attaches to the objects themselves and which would remain, even if one abstracted from all subjective conditions of intuition” (Guyer’s translation). 13. Paul Guyer, Kant and the Claims of Knowledge (Cambridge: Cambridge University Press, 1987), 334. 14. Ibid., 333.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Modesty of the Critical Philosophy   197 While he goes on in succeeding chapters to develop his own view in detail, it is worth noting that this central criticism is reminiscent of Trendelenburg’s “third possibility,” as put forth in the 1860s. Trendelenburg claims that while space and time can be seen as conditions of our receptivity, they may also be features of the world itself: “[S]pace and time have an a priori origin in the representing but are at the same time valid of things.”15 The question then becomes this: how can we know nothing of the thing-in-itself, yet know that it is non-spatial and nontemporal? This persistent complaint, registered by Trendelenburg and Peirce, among others, leads in Guyer’s hands to the charge that Kant, far from being epistemologically modest, is in fact committed to the sheerest kind of dogmatism. In turn, an interpretation that seeks to preserve such modesty “obviously begs the question of transcendental idealism by assuming from the outset that any necessary condition of knowledge is subjective rather than objective, even if this subjective status will be dignified by the title ‘transcendentally ideal’ to signify that it is an indispensable rather than an arbitrary aspect of subjectivity.”16 From this, Guyer concludes that we need not worry about confronting the well-known dilemma posed by Jacobi, that one cannot enter the Critical philosophy without presupposing the thing-in-itself, yet one cannot remain in the Critical philosophy because of the thing-in-itself: “One can enter the critical philosophy, or at least the transcendental theory of experience, without the presupposition of the thing in itself, because none of Kant’s arguments for the nonspatiality and nontemporality of things in themselves, certainly none of his arguments from legitimate claims of the transcendental theory of experience, succeeds.”17 The problem this poses for Kant was developed in a more general fashion by his immediate successors. In various ways, they took the thing-in-itself, or the transcendental object, to be in some sort of causal relationship with the thing as it appears and to cause the appearances by which human beings come to know objects.18 From this view 15. Quoted in Hans Vaihinger, Commentar zu Kant’s Kritik der reinen Vernunft (Stuttgart: Spemann, 1887), 2:136; cf. 35ff. 16. Guyer, Kant and the Claims of Knowledge, 339. 17. Ibid., 335. 18. For the purposes of this discussion, I must ignore the technical distinctions among the thing-in-itself, the transcendental object, and noumena. For an excellent

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

198   Modesty of the Critical Philosophy there quickly follows the idea of the thing-in-itself as some kind of limiting concept of an object, or as representing the object the human being strives to know better and better. Such a reading, at the hands of his critics, commits Kant to a version of the “veil of perceptions” epistemology; as we have seen, Rorty has put Kant in the tradition of those philosophers trying to reconstruct accurately a mirror of the world as it really is. In the Critique, Kant does occasionally speak of the thingin-itself as in a causal relationship with things as they appear: “[T]he categories represent no special object [Objekt], given to the understanding alone, but only serve to determine the transcendental object (the concept of something in general [von etwas überhaupt]), through that which is given in sensibility, in order thereby to cognize appearances empirically under concepts of objects [Gegenstände]” (A251). And in the Prolegomena: “[W]e indeed, rightly considering objects of sense as mere appearances, confess thereby that they are based [zum Grunde liegen] upon a thing in itself, though we know not this thing as it is in itself but only know its appearances, viz. the way in which our senses are affected by this unknown something.”19 It is easy to see why such remarks were construed by Kant’s critics as they were. However, Kant often expressly avoids such language, as in the above-cited remark at A26=B52, specifically rejecting the spatiality and temporality of the thing-in-itself, and thus denying the possibility of its causal relationship with the thing as it appears. The thrust of such a view is given in the “B” preface: “[T]hough we cannot cognize these objects as things in themselves, we must yet be in a position at least to think them as things in themselves: otherwise there would follow the absurd conclusion that there can be appearance without anything that appears” (Bxxvi–xxvii; Kant’s emphases). While Kant may be responsible for much of the confusion on the part of his readers, I think it is possible to present a picture of the thing-in-itself consistent with the minimal claim he makes here in the “B” preface. To begin in the most general fashion, it is essential to an underhistorical overview of the reactions, from Jacobi to Nietzsche, as well as a good account of these distinctions, see Moltke Gram, “Things in Themselves: The Historical Lessons,” Journal of the History of Philosophy 18 (1980): 407–31. 19. Ak. IV., 314–15.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Modesty of the Critical Philosophy   199 standing of the thing-in-itself not to take it as another kind of object in some sort of direct correspondence with objects as they appear. That is, first-order language about the thing-in-itself must be used very cautiously, to avoid giving the impression that we are talking about objects in any but a “degenerate” sense. To imply a causal connection by talking about objects as they are in themselves and as they appear, or to imply some kind of correspondence between them as objects, is to succumb to transcendental illusion. Thus, when Kant tells us that we may “entitle the purely intelligible cause of appearances in general the transcendental object, but merely [bloß] in order to have something corresponding to sensibility as a receptivity” (A494=B522), he avoids saying that the transcendental object corresponds to the thing as it appears, and uses the notion of correspondence carefully. The transcendental object is again put forth here solely to avoid the “absurd conclusion” of an appearance without something appearing. First-order discussions, about the thing-in-itself, at least in Kant’s more precise formulations, use the notion only in this degenerate sense, and always in contrast to objects as they appear or as they qualify as objects of possible experience for a judging subject. More specifically, in the Transcendental Aesthetic and Transcendental Analytic, Kant articulates the necessary sensible and intellectual conditions for the possibility of experience, and the objects of possible experience: objects in a “non-degenerate” sense. As we have seen, the possibility of human cognition of objects requires unifying and judging sensible representations received within the forms of space and time. This sensible and intellectual capacity constitutes the contribution of the judging subject, and in this sense the cognition of objects is contingent upon this contribution. It is this notion of an object that this fundamental to Kant’s presentation in the First Critique, a notion Melnick calls an “epistemic” conception of the object. In contrast, “the notion of a thing in itself is that concept of an object according to which being an object or a thing would make sense in abstraction from any idea of a (type of) subject and his intellectual organization of his experience.”20 It is this that I have been calling an object in its degener20. Melnick, Kant’s Analogies of Experience, 152.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

200   Modesty of the Critical Philosophy ate sense. If Kant is correct that an object is essentially bound up with the contribution of the thinking subject, eliminating the subject eliminates the object as an object of possible experience. But, as we have seen, the thing-in-itself remains, as a way of avoiding the “absurd conclusion,” and it remains as an object of possible thought that is not selfcontradictory. As Thompson points out, this kind of language can easily lead us into transcendental illusion, if we take this degenerate sense of an object as a limitation. For

Copyright © 2008. Catholic University of America Press. All rights reserved.

[a] limitation implies objects free of the limitation, as a limitation on the range of sounds we can hear implies sounds free of this limitation. We may thus take our statement about space and time as implying objects free of this limitation they suffer as objects of our possible experience. These free objects are then objects (or things) in themselves as opposed to objects in space and time, and if free objects act causally, their causality is free of spatiotemporal limitation and therefore not physical.21

The point could also be developed in a different way, in terms of Kant’s discussion of magnetism and iron filings in the Postulates of Empirical Thought. Because the “grossness of our senses does not in any way decide the form of possible experience in general” (A226=B273), the thing-in-itself is not something we could come to know better and better were our eyesight sharper, our sense of smell more acute, etc. Thus, in this sense the thing-in-itself does not function as a “Grenzbegriff.” If we begin with a non-degenerate notion of an object, it becomes clearer why Kant can say that the thing-in-itself is non-spatial and nontemporal without, at this stage of his argument, being committed to dogmatism. Space and time are radically subjective forms contributed by sensibly intuiting subjects (including human beings). If we then abstract from this conception of objects of possible experience, we are left with the object only in a degenerate sense, relative to which it is meaningless, or a category mistake, to inquire about it “really” being spatial and temporal. Such forms of intuition make sense only in reference to the kind of subject, and the objects about which it can judge, here under consideration. Thus, any attempt to talk about the 21. Thompson, “Things in Themselves,” 42.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Modesty of the Critical Philosophy   201

Copyright © 2008. Catholic University of America Press. All rights reserved.

thing-in-itself, as does Trendelenburg, is to fall into transcendental illusion. His question, in other words, only gives the impression of being meaningful. Rather than taking the object as an object of possible human experience, and then going on to consider the implications of this conception for things as they are in themselves (or, perhaps, the thing-in-itself), he reverses the procedure and produces another version of transcendental realism. For such reasons, Kant can say, “The non-sensible cause of these representations is completely unknown to us, and cannot therefore be intuited by us as object [Objekt]. For such an object [Gegenstand] would have to be represented as neither in space nor in time (these being merely conditions of sensible representation), and apart from such conditions we cannot think any intuition” (A494=B522). Kant’s language isn’t as precise as we might like it to be—as we have seen, this notion of “cause” must be dealt with cautiously, and the final clause of this citation at least needs to be spelled out better. Moltke Gram has, however, taken the idea Kant expresses here and constructed what amounts to a transcendental argument for its central claim: If you interpret Kant’s argument epistemically, then the relation of affection holds between phenomenal arguments and our forms of sensibility. And there is no problem about circularity in the definition of “affection.” All we are asked to believe is that we cannot compare what an object is like for our forms of sensibility with what it might be apart from those forms. This is impossible just because we would have to assume that we can compare what an object of possible experience is for us with what it might be under other forms of sensibility. But this, in turn, requires us per impossibile to be able to compare what an object of possible experience is like when it is not an object of possible experience.22

As we have seen, the notion of an object of possible human experience essentially involves the intellectual contribution of the judging subject. Indeed, it could be said that in the Critique, Kant presents two degenerate senses of “object.” The first is that notion of an object “given” to us without the application of intellectual condi22. Gram, “Things in Themselves,” 429; compare the argument Kant gives at A104.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

202   Modesty of the Critical Philosophy tions: “since intuition stands in no need whatsoever of the functions of thought, appearances would none the less present objects to our intuition” (A90–91=B123). The second would be the object taken as it is in itself, stripped of the sensible conditions that are necessary for an object to be given to our intuition. Neither description of an object yields an object of possible experience. Only the kind of object, what I have been calling the object in its non-degenerate sense, qualifies as an object of knowledge, cognition, or experience for the kind of judging subject under consideration here. Guyer, however, argues that in contrast to this result, Kant’s transcendental idealism is not “an anodyne recommendation of epistemological modesty.”23 Guyer’s argument is, simply put, that if Kant is not modest in his assertions about the thingin-itself, then he is dogmatic. Guyer wants to affirm this antecedent— to reject the idea that Kant’s idealism is a “salubrious recommendation of epistemological modesty”24—while I want to deny the consequent and claim that therefore Kant is being modest in making the claims he does, specifically about the thing-in-itself (and, more specifically, as those claims are presented in the Second edition of the Critique). Given what has already been argued, at this point I can be brief. The basic point to be made against Guyer is that it is the object of possible human experience—the object in its non-degenerate sense— that is the fundamental conception of the object operative in Kant’s account of human cognition (and of any agent who judges similarly). This notion is distinct both from the sense of an “object” given to us, which Kant contrasts with knowledge “properly so called” (A78=B103), and from the thing-in-itself. If we go on to ask whether the thing as it appears, as it is in itself, is “really” spatial and temporal, we are being seduced by a picture of two sets of objects in some sort of causal interaction.25 This picture leads to an obvious contradiction that one would 23. Guyer, Kant and the Claims of Knowledge, 336. 24. ibid., 333. 25. Similarly, the language that suggests human beings have “access” to objects only through phenomena is to be resisted. Guyer uses such language in rejecting an argument of Allison’s; Allison argues that “whatever is required for the recognition or picking out of what is ‘objective’ in our experience, must reflect the cognitive structure of the mind ..... rather than the nature of the object itself ” (Kant’s Transcendental Idealism, 27). Otherwise, we are involved in the contradiction of having “access” to an object

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Modesty of the Critical Philosophy   203 suppose Kant would also have noticed. For if, as he says, space and time are not applicable to things in themselves, they would lack just that (minimal) condition—temporality—necessary to be in a causal relationship. Thus, again, such notions must be stated very carefully, for even in saying that things in themselves are not spatial and temporal, we give the impression of making meaningful claims about objects. This is precisely what the modest interpretation of the thing-in-itself seeks to avoid, insisting that meaningful cognitive judgments cannot be made about a thing-in-itself, but only about things as they appear. Only if we begin with objects of possible experience, and then go on to abstract from this conception to arrive at the object in the degenerate sense of a thing-in-itself, can we avoid the perils of an ontological distinction and a “two-worlds” view, with its accompanying transcendental illusion. Rather than being a “harshly dogmatic” claim about things in themselves, such a view serves as an “anodyne” and “salubrious” recommendation of epistemological modesty not to venture beyond the realm of possible experience with any hopes of satisfying cognitive claims—a recommendation, that is, to be content within the “fruitful bathos of experience.”26 But neither is Kant’s “modest” epistemology to be seen as “prudish.” Beyond the limits of experience we cannot have any hopes of gaining the cognition of objects, but there is no obstacle thereby given to the fullest development of knowledge within those limits. As Kant puts it in the Prolegomena, “In mathematics and natural science human reason admits of limits [Schranken] but not of bounds [Grenzen].”27 It is again the point made by Melnick, which stripped of those conditions that would make such access possible. For Allison, this notion of “access” is used in a very weak sense, and only in contrast to what goes on in the human experience of objects. For Guyer, this notion of “access” is used in a strong sense in asserting that Kant assumes we can deny spatiality of objects while “continuing to assume that we have access to such objects even when spatiality is a necessary condition of our knowledge” (Kant and the Claims of Knowledge, 339). On the modest epistemic interpretation of Kant’s thing-in-itself, this idea of “access” plays a minimal role; as Melnick puts it, “It is important to understand how little the notion of what we have access to, or what is knowable to us, is the basic one in this interpretation of a thing in itself ” (Kant’s Analogies of Experience, 153). 26. It is worth noting again that Kant explicitly points out that within experience, no question is ever even asked about the thing-in-itself (A30=B45). 27. Ak. IV, 352.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

204   Modesty of the Critical Philosophy admirably captures the essence of Kant’s foundationalism, viewed as both immoderate and modest: “[A]t least part of Kant’s empirical realism is that everything is ‘left open’ that could be left open. By ‘left open’ here I mean undecidable on a priori grounds or not in any sense contributed by the subject.”28 Kant does not deal explicitly with the method he himself uses in the Critique, nor with the perspective from which he himself begins in presenting his theory of experience. I have tried to indicate that such a lack of explicit discussion results from the internal, yet global, nature of Kant’s critical strategy. That is, if what Kant argues in the Critique is sound, the doctrines articulated in the Critique must hold as well for the articulation of that doctrine itself. Kant’s general approach— carried out by means of a first-person or radically subjective orientation—has come to be known by a variety of names, usually associated with Copernicus: thus we find that approach called the “Copernican hypothesis,” the “Copernican method,” and, most commonly, the “Copernican revolution.”29 What I think must be seen is that his notion of the thing-in-itself must be regarded as complementary to that Copernican revolution, by fundamentally changing our conception of what constitutes an object, specifically, an object of possible experience. Hence, while Guyer is correct in rejecting Jacobi’s dilemma, I think his own reasons for this rejection are incorrect. As I have tried to show, the “legitimate claims” of Kant’s theory of experience essentially include his account of what constitutes an object of possible experience, an account that incorporates the fundamental “assumption” that human receptivity is conditioned by space and time. This is the basic notion of an object operative in the Critical philosophy, upon which any inquiries into the thing-in-itself are parasitic. Granting that this fundamental assumption may be itself “dogmatic,” Kant’s arguments for the non-spatiality and non-temporality of the thing-in-itself become much more plausible if we see that spatiality and temporality are no28. Melnick, Kant’s Analogies of Experience, 56. 29. Kant’s own reference is to a strategy that is analogous to “the first thoughts of Copernicus” (Bxxii n.); strictly speaking, hypotheses are found only within empirical science. Kant addresses the topic in “The Discipline of Pure Reason in Regard to Hypotheses” at A769ff.=B797ff.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Modesty of the Critical Philosophy   205 tions applicable only within his transcendental theory of experience. And, as Kant himself summarizes the more general point, “a complete review of all the powers of reason—and the conviction thereby obtained of the certainty of its claims to a modest possession [eines kleinen Besitzes], even in the case of the vanity of higher pretensions—puts an end to all conflict, and induces it to rest satisfied with a limited but undisputed property” (A768=B796).

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant and Common Sense Kant famously credits Rousseau for having taught him to “honor humanity,” and, particularly in his various lectures on logic, Kant repeatedly identifies the healthy and common human understanding as a faculty available to all. In a number of texts—the Foundations of the Metaphysics of Morals, the Critique of Practical Reason, and especially the Critique of Judgment—Kant praises “common sense.” Thus, in the latter he writes, “[W]e can assume common sense as the necessary condition of the universal communicability of our cognition, which must be presupposed in any logic and any principle of cognition which is not skeptical.”30 At the same time, he seems critical of common sense, as in the natural disposition human beings have to speculate uncritically, generating what Kant derides as “pious brainstorms,” succumbing to a wide variety of philosophical positions Kant finds suspect, including enthusiasm, skepticism, relativism, indifferentism, atheism, fanaticism, and superstition.31 He notes further, in the Prolegomena, that “it is not necessary that everyone study metaphysics,” implying that many are simply not equipped to do so, and in any case, in metaphysics, common sense—even sound common sense—“has no right to judge at all.”32 Such claims, obviously enough, seem to be in conflict, suggesting a problem Kant must address, namely: what respect does “sound common sense” deserve, if any? 30. Critique of Judgment, Ak. V, 66 (§21). 31. For a sustained treatment of some of these views, see Kant’s relatively early (1766), empirically oriented “Dreams of a Spirit-Seer, Elucidated by Dreams of Metaphysics.” 32. Prolegomena, Ak. IV 263, 260.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

206   Modesty of the Critical Philosophy

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s solution to this problem is to recognize that, in different ways, common sense and metaphysics, on Kant’s construal of the latter, play complementary roles relative to each other. Common sense, as Kant indicates by noting the natural human disposition to ask questions that are, fundamentally, metaphysical, focuses specifically on issues of crucial importance, precisely those that constitute the ratio essendi of philosophy (and philosophical criticism). At the same time, human beings—at the level of common sense as well as for those attempting a more “sophisticated” philosophical analysis—fail to submit their own speculative results to critical scrutiny. In sum, the philosopher’s role is to provide that scrutiny, thus providing at least a reminder of the dangers of failing to observe the limits of speculative philosophy, particularly because the “natural and unavoidable dialectic” that accompanies pure reason and is inseparable from it will constantly require attention and correction (A298=B355). To be sure, common sense has a somewhat diminished role in the theoretical philosophy, where the results of the Critical philosophy affect the “monopoly of the schools” (Bxxxii), although even here the critical results of Kant’s account cannot contradict the practices of the healthy human understanding. For it is in the practical philosophy where common sense can be seen to be an advantage, relative to the “abstruse philosophers”: [T]he most remarkable thing about ordinary reason in its practical concern is that it may have as much hope as any philosopher of hitting the mark. In fact, it is almost more certain to do so than the philosopher, because he has no principle which the common understanding lacks, while his judgement is easily confused by a mass of irrelevant considerations, so that it easily turns aside from the correct way.33

Yet here as well, “ordinary” reason is in need of the critical discipline required to determine if in fact its claims pass muster. Again, the results indicated by that discipline will not conflict with the practices of the moral agent; but it has the advantage of being able to give warrant to those practices, and indicate those practices that may appear justified but, in fact, are not.34 For, as W. H. Walsh has made the point, the 33. Foundations of the Metaphysics of Morals, Ak. IV, 404. 34. One example of this would be the famous “alleged” right to lie from benevolent

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Modesty of the Critical Philosophy   207 “philosophy of the uninstructed mind is either an affair of incoherent speculation or an unconvincing defence of common sense.”35 Thus both common sense and critical complement each other, and there is, or should be, no ultimate conflict between them. As Manfred Kuehn summarizes the point, Kant

Copyright © 2008. Catholic University of America Press. All rights reserved.

rejects common sense as a criterion or tool of philosophical inquiry—be it in theoretical, practical, or aesthetic contexts. And all his polemic against common sense concerns such use of it. But he just as clearly accepts common sense as the field of philosophical inquiry. Because metaphysics and common sense are for him inextricably intertwined, the latter is relevant for him even in speculative philosophy, though it is perhaps more apparent in practical and aesthetic contexts.36

McGowan has pointed that “only in Hegel ..... does the notion appear that some special form of interpretation is necessary to understand Reason’s principles, and that notion almost invariably induces certain groups of interpreters who have privileged and superior insight.”37 Particularly in the practical philosophy, Kant remains consistent in showing respect for the sound and healthy common understanding possessed by all those who demonstrate the capacity for reason—those whom we are willing to call “rational”—and, while recognizing the particular disciplinary provenance of the philosopher, he does not claim the kind of “privilege” McGowan refers to here. In contrast, as Karl Ameriks has put the point, “Kant’s philosophy has an unusual openness to a wide variety of considerations and a special respect for universal common sense.”38 morals; see C. Korsgaard, “The Right to Lie: Kant on Dealing with Evil,” Philosophy and Public Affairs 15, no. 4 (1986): 325–49. 35. W. H. Walsh, Kant’s Criticism of Metaphysics (Chicago: University of Chicago Press, 1975), 254. 36. Kuehn, Scottish Common Sense in Germany, 1768–1790, 202. As Kuehn notes, Kant “had to show as convincingly as he could that his own critical philosophy was rather different from naturalism” (193), and this led Kant to emphasize his differences with the Scottish philosophers who, as Kuehn persuasively argues, had such an effect on Kant’s thought. 37. McGowan, Postmodernism and Its Critics, 33. 38. Ameriks, Kant and the Fate of Autonomy, 38.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

208   Modesty of the Critical Philosophy

Copyright © 2008. Catholic University of America Press. All rights reserved.

Conclusion In reflecting upon my thought and experience, I find myself in a world I didn’t create. But I also discover that the very capacity to reflect reveals a spontaneous ability to think and to judge about that world. Pursuing this, I come to realize that for these thoughts and judgments to be meaningful, they must conform to certain minimal, but universal and necessary, conditions, conditions I thus discover reflectively that I must presuppose. A thought, if it is to make sense, cannot violate a rule that would, for instance, permit it to be self-contradictory. A judgment of experience, to be such that its truth-value could be investigated and determined, must conform to those rules that make that experience possible. Such rules are subjective—they are imposed by a thinking subject, and are presupposed by that subject in articulating the very conditions of objectivity. But such rules are also objective, in being universal and necessary conditions that must hold for any such thinking, judging subject, and range over the domain of possible thought and possible experience. This yields a sobering reevaluation of the pretensions of both dogmatic metaphysics and uncritical empiricism. Kant characterizes these results in traditional foundationalist terms, thereby revealing the scope and limits of his own foundationalism: “[A]lthough we had contemplated building a tower which should reach to the heavens, the supply of materials suffices only for a dwelling-house, just sufficiently roomy for our business on the level of experience, and just sufficiently high to allow our looking beyond it” (A707=B735).39 No doubt, Kant is committed in many cases to such claims as “this is how it must be!” Such claims invariably invite “skeptical” responses, ranging from simply ask39. This picks up on an image from the “A” preface, where Kant cites Persius: “Tecum habita, et noris quam sit tibi supellex” (Axx), trans. G. G. Ramsey in Persius and Juvenal (London: Heinemann, 1918), as “live in your house, and recognize how poorly it is furnished.” Kant makes a similar point in the much earlier “Dreams of a Spirit-Seer,” at a period when his thought was perhaps closer to Hume’s than at any other, by having Socrates exclaim, “How many things there are, and see how I don’t need all of them!” (Ak. II, 369.) It is worth noting in this context Wittgenstein’s remark that “I have arrived at the rock bottom of my convictions. And one might almost say that these foundation-walls are carried by the whole house.” On Certainty, trans. Denis Paul and G. E. M. Anscombe (New York: Harper and Row, 1969), §248.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Modesty of the Critical Philosophy   209 ing “why?” to putting forth an equally strong argument for a directly opposed claim, to perhaps an even more radical form of skepticism. Even if Kant’s arguments are successful in showing that any meaningful claim must presuppose certain necessary conditions for that claim to be possible, or meaningful, and even if a “transcendental argument” can prove the existence of an external world and the necessity of the causal principle, there are still skeptical candidates that remain unrefuted. But it is not clear (a) that such a position can be formulated and (b) what its interest would be, if it could be formulated (and, presumably, somehow communicated). Once it is put forth as a meaningful utterance or set of utterances, it would seem to be committed already to at least those kinds of conditions discussed above, that are necessary for the possibility of meaning in its most general, non-technical (common) sense. And if it is not formulated in such a fashion, it doesn’t seem that it could engage us in any philosophically significant way. As Wittgenstein puts the point in the Tractatus, “Skepticism is not irrefutable, but obviously nonsensical when it tries to raise doubts where no questions can be asked. For doubt can exist only where a question exists, a question only where an answer exists, and an answer only where something can be said.”40 Later, Wittgenstein moved away from some of the views that show remarkable similarity to Kant’s, and was content to talk in terms of moves in language games, and “forms of life”; what Kant would call universal and necessary conditions become practices that a community has adopted, insofar as “they make good sense for members of a given community.”41 But if we take Kant’s results seriously, we might wish to conceive of those conditions, qua universal, as holding for anyone who could qualify as a member of any community. As mentioned earlier, the attribution of rationality and agency to another is, for Kant, a contingent claim, albeit a claim that brings with it a priori commitment to such an agent being able to use the pronoun “I,” judge conceptually, and regard itself as free. The question then becomes whether one who fails to satisfy these minimal constraints—and it would have 40. Tractatus, §6.51. 41. Norris, “What Is Enlightenment?” 162.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

210   Modesty of the Critical Philosophy

Copyright © 2008. Catholic University of America Press. All rights reserved.

to be the case that “common sense” would unthinkingly embrace these constraints as well—would count as a member of any community. On Kant’s view, then, we have the following results: (1) an argument against a variety of skeptical positions, by offering secure and justified rules that are universal and necessary for the possibility of experience; (2) a thoroughgoing dismantling of the dogmatic pretensions of an uncritical employment of speculative reason; (3) rules for experience that preserve the essentially contingent nature of individual judgments of experience; (4) “making room for faith,” or for the deepest moral concerns of practical philosophy, by limiting theoretical reason to its legitimate (and modest) domain, thereby revealing the necessity of practical reason for those issues that lie beyond this domain. Undoubtedly, some would object that Kant’s logic of experience, as presented here in outline, is still too weak; others will argue that any conception of synthetic claims, made a priori, as an indispensable dimension of human knowledge always involves a commitment that is too strong. I would simply respond that Kant wants to acknowledge that in metaphysics and epistemology, “proud Philosophy” can do and establish many things—but not everything. For, as he reminds us at A727=B755, “it is precisely in knowing its limits that philosophy consists.”

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Bibliography

Copyright © 2008. Catholic University of America Press. All rights reserved.

Kant’s Works The standard source for Kant’s works is the “Academy” edition: Kants Gesammelte Schriften, ed. Königliche Preußische (later Deutscher) Akademie der Wissenschaften. Berlin: Walter de Gruyter, 1902–. The Critique of Pure Reason appears in volumes 3 and 4 of this edition. In addition, I have used the following: Guyer, P., and A. Wood, trans. Critique of Pure Reason. Cambridge: Cambridge University Press, 1999. Kemp Smith, Norman, trans. Immanuel Kant’s Critique of Pure Reason. London: Macmillan, 1929. Pluhar, W., trans. Critique of Pure Reason. Indianapolis: Hackett, 1996. Schmidt, Raymund, ed. Kritik der reinen Vernunft. Hamburg: Felix Meiner Verlag, 1954. All references to the Critique of Pure Reason are given in the customary fashion, citing the first (“A”) and second (“B”) editions parenthetically. Kemp Smith’s translation is generally followed, with minor emendation; substantial changes are indicated in the footnotes. I have also consulted the Guyer-Wood and Pluhar translations. All other references to Kant’s works are cited by listing the pagination of the Akademie edition by volume, preceded by “Ak.”; the translation used is then cited parenthetically, by the translator’s name and page number. (These translations usually also indicate Akademie pagination.) These translations are at times extensively revised; where no translator is given, it can be assumed that the translation is my own. It being the fashion of Kant’s day often to give long titles to works, I have used abbreviations for some of those cited (e.g., “Über eine Entdeckung” for “Über eine Entdeckung nach der all neue Kritik der reinen Vernunft durch eine ältere entbehrlich gemacht werden soll”).

211

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

212   Bibliography

Copyright © 2008. Catholic University of America Press. All rights reserved.

Works by Others Aarsleff, H. From Locke to Saussure. Minneapolis: University of Minnesota Press, 1982. Allison, H. Kant’s Transcendental Idealism. New Haven: Yale University Press, 1983. ———. “Transcendental Affinity—Kant’s Answer to Hume?” In Kant’s Theory of Knowledge, edited by L. W. Beck. Dordrecht: Reidel, 1974. Alston, W. “Two Types of Foundationalism.” In Epistemic Justification. Ithaca: Cornell University Press, 1989. Ameriks, K. Kant and the Fate of Autonomy. Cambridge: Cambridge University Press, 2000. Amrhein, H. “Kants Lehre vom ‘Bewußtsein überhaupt.’” Berlin: Kant Studien Ergänzunsheft 10, 1909. Antony, L. “Quine as Feminist: The Radical Import of Naturalized Epistemology.” In A Mind of One’s Own, edited by C. Witt and L. Antony. Boulder: Westview Press, 1993. Ashworth, E. J. Language and Logic in the Post-Medieval Period. Dordrecht: D. Reidel, 1974. Atkinson, R. F. “Hume on Mathematics.” Philosophical Quarterly 10 (1960). Bacon, Roger. Oxford Greek Grammar. Edited by E. Nolan and S. A. Hirsch. Cambridge: Cambridge University Press, 1902. Beck, J. S. Einzig-mögliche Standpunckt aus welchem die critische Philosophie beurrtheilt werden muß (Riga: Johann Friedrich Hartnoch, 1796; reprint Aetas Kantiana, Brussells: Culture et Civilisation, 1968). Beck, L. W. “Can Kant’s Synthetic Judgments Be Made Analytic?” In Kant: A Collection of Critical Essays. Edited by R. Wolff. Garden City: Doubleday, 1967. ———. Essays on Kant and Hume. New Haven: Yale University Press, 1978. Beiser, F. The Fate of Reason. Cambridge, Mass.: Harvard University Press, 1978. Benedict, R. Patterns of Culture. Boston: Houghton Mifflin, 1934. Benfey, T. “Über die Aufgabe des platonischen Dialogs Kratylos.” Abhandlungen Göttingen, Philologische-Historische Klasse 12 (Göttingen, 1866). Bennett, J. Kant’s Analytic. Cambridge: Cambridge University Press, 1966. Beth, E. W. “Hundred Years of Symbolic Logic.” Dialectica 1 (1947). Bird, G. Kant’s Theory of Knowledge. New York: Humanities Press, 1962. ———. Review of McDowell’s Mind and World, Philosophy 71 (1996). Bocheński, J. “History of Logic and the Criteria of Rationality.” In Logic Counts, edited by E. Zarnecka-Bialy. Boston: Kluwer, 1991. BonJour, L. In Defense of Pure Reason. Cambridge: Cambridge University Press, 1998. ———. The Structure of Empirical Knowledge. Cambridge, Mass.: Harvard University Press, 1985.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Bibliography   213 Bordo, Susan. The Flight to Objectivity. Albany: State University of New York Press, 1987. Boswell, T. Quellenkritische Untersuchungen zum Kantischen Logikhandbuch. Frankfurt a.M.: Lang, 1991. Boyne, R. Foucault and Derrida: The Other Side of Reason. New York: Routledge, 1996. Brandt, R. Die Urteilstafel: Kritik der reinen Vernunft A76–76; B 92–101. Hamburg: Meiner Verlag, 1991. Brittan, G. Kant’s Theory of Science. Princeton: Princeton University Press, 1978. Brook, A. Kant and the Mind. Cambridge: Cambridge University Press, 1997. Brown, D. Human Universals. Philadelphia: Temple University Press, 1991. Buchdahl, G. Metaphysics and the Philosophy of Science. Oxford: Blackwell, 1969. ———. “Realism and Realization in a Kantian Light.” In Reading Kant, edited by E. Schaper and W. Vossenkuhl. Oxford: Blackwell, 1989. ———. “Science and God: The Topology of the Kantian World.” Southern Journal of Philosophy 30, Supplement (1991). Bursill-Hall, G. L. Speculative Grammars of the Middle Ages: The Doctrine of the Partes Orationis of the Modistae. The Hague: Mouton, 1971. Campbell, D. “Evolutionary Epistemology.” In The Philosophy of Karl Popper, edited by P. Schilpp. LaSalle, Ill.: Open Court Publishing, 1974. Camus, A. “Reflections on the Guillotine.” In Resistance, Rebellion, and Death. Translated by J. O’Brien. New York: Knopf, 1961. Carboncini, S., and R. Finster. “Das Begriffspaar Kanon-Organon.” Archiv für Begriffsgeschichte 26 (1982). Carl, W. “Response to Rorty.” In Transcendental Arguments and Science, edited by P. Bieri, R. P. Horstmann, and L. Krüger. Dordrecht: Reidel, 1979. Cassirer, E. “Das Problem J.-J. Rousseau.” Archiv für Geschichte der Philosophie 41 (1932). ———. “Goethe and the Kantian Philosophy.” In Rousseau, Kant, Goethe. Translated by J. Gutmann, P. O. Kristeller, and J. H. Randall Jr. New York: Harper and Row, 1963. ———. The Philosophy of Symbolic Forms. Vol. 1. Translated by R. Mannheim. New Haven: Yale University Press, 1953. ———. The Philosophy of the Enlightenment. Translated by F. Koelln and J. Pettegrove. Princeton: Princeton University Press, 1951. ———. The Problem of Jean-Jacques Rousseau. Translated by Peter Gay. Bloomington: Indiana University Press, 1967. Cesa, C., and N. Hinske, eds. Kant und sein Jahrhundert. Frankfurt a.M.: Lang, 1993. Cherniak, C. Minimal Rationality. Cambridge, Mass.: MIT Press, 1986. Chomsky, N. Cartesian Linguistics. New York: Harper and Row, 1966. ———. Reflections on Language. New York: Pantheon, 1975. Clark, M. Nietzsche on Truth and Philosophy. New York: Cambridge University Press, 1990.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

214   Bibliography Code, L. What Can She Know? Ithaca: Cornell University Press, 1991. Cohen, H. Kants Theorie der Erfahrung. Berlin: Cassirer, 1925. Collins, A. Possible Experience. Berkeley: University of California Press, 1999. Conrad, Elfriede. Kants Logikvorlesungen als neuer Schlüssel zur Architekton der Kritik der reinen Vernunft. Stuttgart: Frommann-Holzboog, 1994. Cutrofello, A. “On the Transcendental Pretensions of the Principle of Charity.” In The Philosophy of Donald Davidson, edited by L. E. Hahn. Chicago: Open Court Press, 1999. Davidson, D. “The Emergence of Thought” and “The Irreducibility of the Concept of the Self.” Reprinted in Subjective, Intersubjective, Objective. Oxford: Clarendon Press, 2001. ———. “Intellectual Autobiography.” In The Philosophy of Donald Davidson, edited by L. E. Hahn. Chicago: Open Court Press, 1999. ———. “On the Very Idea of a Conceptual Scheme,” “Thought and Talk,” and “The Method of Truth in Metaphysics.” Reprinted in Inquiries into Truth and Interpretation. Oxford: Clarendon Press, 1984. ———. “The Second Person.” In Midwest Studies in Philosophy, vol. 17, edited by P. French, T. Uehling, and H. Wettstein. Notre Dame: University of Notre Dame Press, 1992. ———. “The Structure and Content of Truth” (1989 Dewey Lectures). Journal of Philosophy 87, no. 6 (1990). Delbos, V. “Rousseau et Kant.” In J. J. Rousseau—Leçons faites à l’Ecole des Hautes Études Sociales. Paris: Alcan, 1912. De Vleeschauwer, H. J. La déduction transcendentale dans l’oeuvre de Kant. Paris: Leroux, 1934–1937. ———. “Wie ich jetzt die Kritik der reinen Vernunft entwicklungsgeschichtlich lese.” Kant-Studien 54 (1963). Diamond, C. The Realistic Spirit. Cambridge, Mass.: MIT Press, 1995. Dummett, M. Elements of Intuitionism. London: Duckworth, 1977. Ebbinghaus, J. Review of Reich’s Die Vollständigkeit der kantischen Urteilstafel. Deutsche Literaturzeitung 44 (1933). Engel, P. The Norm of Truth. Toronto: University of Toronto Press, 1991. Ewing, A. C. Kant’s Treatment of Causality. London: Kegan Paul, Trench, Trubner, 1924. Ferrari, J. Les sources françaises de la Philosophie de Kant. Paris: Librarie Klincksieck, 1979. Flax, J. “Postmodernism and Gender Relations in Feminist Theory.” In Feminism/Postmodernism, edited by L. J. Nicholson. New York: Routledge, 1990. Flynn, Thomas. “Foucault’s Mapping of History.” In The Cambridge Companion to Foucault, edited by G. Gutting. Cambridge: Cambridge University Press, 1994. Forster, M. Hegel and Skepticism. Cambridge, Mass.: Harvard University Press, 1989.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Bibliography   215 Foucault, M. The Order of Things. Translated by A. Sheridan. New York: Vintage Books, 1973. ———. Politics, Philosophy, and Culture: Interviews and Other Writings, 1977– 1984. Edited by L. Kritzman. New York: Routledge, 1988. ———. “What Is Enlightenment?” In The Foucault Reader. Edited by P. Rabinow. New York: Pantheon Books, 1984. Frede, Michael. “The Origins of Traditional Grammar.” In Essays on Ancient Philosophy. Minneapolis: University of Minnesota Press, 1987. Frege, G. The Basic Laws of Arithmetic. Edited and translated by M. Furth. Berkeley: University of California Press, 1967. ———. The Foundations of Arithmetic. Translated by J. Austin. Oxford: Blackwell, 1978. ———. Posthumous Writings. Translated by R. White and P. Long. Oxford: Blackwell, 1997. Friedman, M. Kant and the Exact Sciences. Cambridge, Mass.: Harvard University Press, 1992. ———. “Kant, Skepticism, and Idealism.” Inquiry 49 (2006). ———. “Kant’s Theory of Geometry.” Philosophical Review 94 (1985). Genova, A. “Kant’s Notion of Transcendental Presupposition in the First Critique.” In Essays on Kant’s Critique of Pure Reason, edited by J. Mohanty and R. Shahan. Norman: University of Oklahoma Press, 1982. Gibbs, R. “The Limits of Thought: Rosenzweig, Schelling, and Cohen.” Zeitschrift für Philosophische Forschung 43 (1989). Gilead, A. “The Relationship between Formal and TranscendentalMetaphysical Logic According to Kant.” Monist 65 (1982). Goldmann, L. Introduction à la philosophie de Kant. Paris: Gallimard, 1967. Translated by R. Black as Immanuel Kant. London: NLB, 1971. Good, T. A Brief English Tract of Logick. Oxford: L. Lichfield, 1677. Gram, M. “Things in Themselves: The Historical Lessons.” Journal of the History of Philosophy 18 (1980). Grayeff, F. “The Relationship of Transcendental and Formal Logic.” KantStudien 51 (1959–1960). Grimm, J., and W. Grimm. Deutsches Wörterbuch. Leipzig: S. Hirzel, 1885. Grimm, R. Nietzsche’s Theory of Knowledge. Berlin: de Gruyter, 1977. Guthrie, W. K. C. A History of Greek Philosophy. Vol. 5. Cambridge: Cambridge University Press, 1986. Guyer, P. Kant and the Claims of Knowledge. Cambridge: Cambridge University Press, 1987. Haack, S. Deviant Logics. Cambridge: Cambridge University Press, 1974. ———. Philosophy of Logics. Cambridge: Cambridge University Press, 1978. Hamann, J. G. Sämtliche Werke. Vol. 3. Vienna: Hamann Verlag, 1949–1957. Hanna, R. Kant and the Foundations of Analytic Philosophy. Oxford: Oxford University Press, 2001. ———. “Rationality and the Ethics of Logic.” Journal of Philosophy 100 (2006).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

216   Bibliography Harrison, R. On What There Must Be. Oxford: Clarendon Press, 1974. Hegel, G. W. F. Enzyklopädie der philosophischen Wissenschaften. Vol. 1: Die Wissenschaft der Logik. Frankfurt a.M.: Suhrkamp, 1983. ———. Glauben und Wissen. Frankfurt am Main: Suhrkamp, 1970. ———. Wissenschaft der Logik 1. Edited by G. Lasson. Hamburg: Meiner Verlag, 1932. Heidegger, M. Kant und das Problem der Metaphysik. 4th ed. Frankfurt a.M.: Klostermann, 1973. Henrich, D. “Kant’s Notion of a Deduction and the Methodological Background of the First Critique.” In Kant’s Transcendental Deductions, edited by E. Förster. Stanford: Stanford University Press, 1989. ———. “The Proof Structure of Kant’s Transcendental Deduction.” Review of Metaphysics 22 (1969). Hintikka, J. “Kant’s Theory of Mathematics Revisited.” In Essays on Kant’s Critique of Pure Reason, edited by J. Mohanty and R. Shahan. Norman: University of Oklahoma Press, 1982. Horstmann, R. P. “The Metaphysical Deduction in Kant’s Critique of Pure Reason.” Philosophical Forum 13, no. 1 (Fall 1981). Jameson, F. Postmodernism, or, The Cultural Logic of Late Capitalism. Durham: Duke University Press, 1991. Jansohn, H. Kants Lehre von der Subjektivität. Bonn: Bouvier, 1969. Jevons, W. S. Elementary Lessons in Logic. London: MacMillan, 1889. Jungius, J. Logica Hamburgensis. Edited and translated (into German) by R. W. Meyer. Hamburg: Augustin, 1957. Kerstein, S. Kant’s Search for the Supreme Principle of Morality. Cambridge: Cambridge University Press, 2002. Kirwan, C. Aristotle’s Metaphysics, Books Γ, ∆, Ε. Oxford: Clarendon Press, 1971. Kitcher, P. Kant’s Transcendental Psychology. Oxford: Oxford University Press, 1990. Kneale, W., and M. Kneale. The Development of Logic. Oxford: Clarendon Press, 1962. Körner, S. Kant. Middlesex: Penguin, 1955. Korsgaard, C. “The Right to Lie: Kant on Dealing with Evil.” Philosophy and Public Affairs 15, no. 4 (1986). Krüger, L. “Wollte Kant die Vollständigkeit seiner Urteilstafel beweisen?” Kant-Studien 59 (1968). Kuehn, M. Kant: A Biography. Cambridge: Cambridge University Press, 2001. ———. Scottish Common Sense in Germany, 1768–1800. Kingston [Ont.]: McGill-Queen’s University Press, 1987. Laurie, S. “Kant’s Criticism of Pure Reason.” Journal of Speculative Philosophy 6 (1872). Lear, J. “Leaving the World Alone.” Journal of Philosophy 79 (1982). Lehmann, G. “Zu Kants Logik-Vorlesung um 1780, anläßlich einer neu aufgefundenen Nachschrift.” Kant-Forschungen 1 (1987).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Bibliography   217 Lloyd, G. The Man of Reason. Minneapolis: University of Minnesota Press, 1981. Locke, J. An Essay Concerning Human Understanding. Edited with an introduction by P. Nidditch. Oxford: Clarendon Press, 1975. Longuenesse, B. Kant and the Capacity to Judge. Translated by C. Wolfe. Princeton: Princeton University Press, 1998. ———. “Kant et les jugements empiriques.” Kant-Studien 86 (1995). Lorenz, K. “Kant’s Doctrine of the A Priori in the Light of Contemporary Biology.” In General Systems Yearbook, vol. 7, edited by L. Bertalanffy and A. Rapoport. Ann Arbor, Mich.: The System, 1962. Lukasiewicz, J. Aristotle’s Syllogistic from the Standpoint of Modern Formal Logic. Oxford: Clarendon Press, 1951. MacFarlane, J. “Frege, Kant, and the Logic in Logicism.” Philosophical Review 111 (2002). Machina, K. “Kant, Quine, and Human Experience.” Philosophical Review 81 (1979). Maddy, P. “Naturalism and the A Priori.” In New Essays on the A Priori, edited by P. Boghossian and C. Peacocke. Oxford: Oxford University Press, 2000. Magee, B. Men of Ideas. London: BBC, 1978. Mason, R. Before Logic. Albany: SUNY Press, 2000. McDowell, J. Mind and World. Cambridge, Mass.: Harvard University Press, 1994. McGowan, J. Postmodernism and Its Critics. Ithaca: Cornell University Press, 1991. Meerbote, R. “The Unknowability of Things in Themselves.” In Kant’s Theory of Knowledge, edited by L. W. Beck. Dordrecht: Reidel, 1974. Melnick, A. Kant’s Analogies of Experience. Chicago: University of Chicago Press, 1973. Moser, P. “A Defense of Epistemic Intuitionism.” Metaphilosophy 15 (1984). Mosser, K. “Kant and Epistemology.” Ph.D. diss., University of Chicago, 1990. ———. “Kant and Feminism.” Kant-Studien 90, no. 3 (1999). ———. “Should the Skeptic Live His Skepticism? Nietzsche and Classical Skepticism.” Manuscrito 21, no. 1 (1998). ———. “Stoff and Nonsense in Kant’s First Critique.” History of Philosophy Quarterly ( January 1993). Nagel, T. The Last Word. New York: Oxford University Press, 1997. Neiman, S. The Unity of Reason. Oxford: Oxford University Press, 1997. Nietzsche, F. Beyond Good and Evil. Translated by W. Kaufmann. New York: Vintage Books, 1966. Norris, C. “What is Enlightenment?” In The Cambridge Companion to Foucault, edited by G. Gutting. Cambridge: Cambridge University Press, 1994. Nozick, R. The Nature of Rationality. Princeton: Princeton University Press, 1993. Nussbaum, C. “Critical and Pre-Critical Phases in Kant’s Philosophy of Logic.” Kant-Studien 83, no. 3 (1992).

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

218   Bibliography Parsons, C. “Kant’s Philosophy of Arithmetic.” In Kant on Pure Reason, edited by R. C. S. Walker. Oxford: Oxford University Press, 1982. Paton, H. J. “Formal and Transcendental Logic.” Kant-Studien 49 (1957). ———. Kant’s Metaphysic of Experience. New York: Macmillan, 1936. Peirce, C. S. Elements of Logic. In Collected Papers of Charles Sanders Peirce. Edited by C. Hartshorne and P. Weiss. Cambridge, Mass.: Harvard University Press, 1931. Pinborg, Jan. “Speculative Grammar.” In The Cambridge History of Later Medieval Philosophy, edited by N. Kretzmann, A. Kenny, and J. Pinborg. Cambridge: Cambridge University Press, 1982. Pinder, T. “Kants Begriff der Logik.” Archiv für Geschichte der Philosophie 61 (1979). Pinker, S. The Language Instinct. New York: Morrow, 1994. Pippin, R. Kant’s Theory of Form. New Haven: Yale University Press, 1982. Poellner, P. Nietzsche and Metaphysics. Oxford: Clarendon Press, 1995. Posy, C., ed. Kant’s Philosophy of Mathematics: Modern Essays. Dordrecht: Kluwer, 1992. Pozzo, R. “Catalogus Praelectionum Acadamiae Regiomontanae 1709–1804.” Studi Kantiani 4 (1991). ———. George Friedrich Meier’s Vernunftlehre: Eine historisch-systematische Untersuchung. Stuttgart: Frommann-Holzboog, 2000. ———. Kant und das Problem einer Einleitung in die Logik. Frankfurt a.M.: Lang, 1989. ———. “Kant within the Tradition of Modern Logic.” Review of Metaphysics 52 (December 1998). Pozzo, R., and M. Oberhausen, eds. Vorlesungsverzeichnisse der Universität Königsberg 1720–1804: Mit einer Einleitung und Registern. Stuttgart: Frommann-Holzboog, 1999. Prauss, G. Erscheinung bei Kant. Berlin: De Gruyter, 1971. ———. “Zum Wahrheitsproblem bei Kant.” Kant-Studien 60 (1969). Priest, G., R. Routley, and J. Norman, eds. Paraconsistent Logic: Essays on the Inconsistent. Munich: Philosophia Verlag, 1989. Putnam, H. “Analyticity and APriority.” In Realism and Reason: Philosophical Papers, vol. 3. Cambridge: Cambridge University Press, 1983. ———. The Many Faces of Realism. LaSalle, Ill.: Open Court, 1988. Quine, W. V. “Methodological Reflections on Current Linguistic Theory.” In Semantics of Natural Languages, edited by D. Davidson and G. Harman. Dordrecht: Reidel, 1972. ———. “Two Dogmas of Empiricism.” In From a Logical Point of View. Cambridge, Mass.: Harvard University Press, 1980. ———. The Ways of Paradox and Other Essays. New York: Random House, 1966. ———. Word and Object. Cambridge, Mass.: MIT Press, 1960.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Bibliography   219 Ramsey, F. P. The Foundations of Mathematics. New York: Humanities Press, 1950. Rawls, J. Lectures on the History of Moral Philosophy. Edited by B. Herman. Cambridge, Mass.: Harvard University Press, 2000. Reich, K. “Die Vollständigkeit der kantischen Urteilstafel.” Dissertation, Rostock, 1932. ———. “Rousseau and Kant.” Translated by K. Mosser. Graduate Faculty Philosophy Journal 23, no. 2 (2002): 35–54. ———. Rousseau und Kant. Tübingen: J. C. B. Mohr [Paul Siebeck], 1936. Rescher, N. “Kant and the ‘Special Constitution’ of Man’s Mind.” In Akten des IV Internationalen Kant-Kongresses (Berlin: de Gruyter, 1974), part II.1. Rischmiller, M., ed. Bemerkungen in den “Beobachtungen über das Gefühl der Schönen und Erhabenen.” Hamburg: Kant-Forschungen [vol. 3], 1991. Rorty, R. Contingency, Irony, and Solidarity. Cambridge: Cambridge University Press, 1989. ———. Philosophy and the Mirror of Nature. Princeton: Princeton University Press, 1979. ———. “Transcendental Arguments, Self-Reference, and Pragmatism.” In Transcendental Arguments and Science, edited by P. Bieri, R. P. Horstmann, and L. Krüger. Dordrecht: Reidel, 1979. ———. “The World Well Lost.” Journal of Philosophy 69 (1972). Rotenstreich, N. Experience and Its Systematization. The Hague: Martinus Nijhoff, 1972. Rousseau, J.-J. Oeuvres complètes. Edited by B. Gagnebin and M. Raymond. Paris: Gallimard [Pléiade], 1964. Russell, B. The Principles of Mathematics. London: Allen and Unwin, 1903. Schmitt, F. Truth: A Primer. Boulder: Westview Press, 1995. Schott, R. M. Eros and Cognition. Boston: Beacon Press, 1988. Sellars, W. Empiricism and the Philosophy of Mind. Edited by R. Brandom. Cambridge, Mass.: Harvard University Press, 1997. Sextus Empiricus. Against the Logicians. Translated by R. G. Bury. Cambridge, Mass.: Harvard University Press, 1935. ———. Outlines of Pyrrhonism. Translated by R. G. Bury. Cambridge, Mass.: Harvard University Press, 1933. Shamoon, A. “Kant’s Logic.” Dissertation, Columbia University, 1979. Slaughter, M. M. Universal Languages and Scientific Taxonomy in the Seventeenth Century. Cambridge: Cambridge University Press, 1982. Smart, H. “Two Views on Kant and Formal Logic.” Philosophy and Phenomenological Research 14 (1955). Smit, H. “Kant on Marks and the Immediacy of Intuition.” Philosophical Review 109.2 (April 2000): 235–66. Sosa, E. “The Raft and the Pyramid.” In Empirical Knowledge: Readings in Contemporary Epistemology, edited by P. Moser. Totowa, N.J.: Rowman and Littlefield, 1986.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

220   Bibliography Steiner, G. After Babel. New York: Oxford University Press, 1975. Steiner, M. “Kant’s Misrepresentations of Hume’s Philosophy of Mathematics in the Prolegomena.” Hume Studies 12 (1987). Strawson, P. F. The Bounds of Sense. London: Methuen, 1966. ———. An Introduction to Logical Theory. London: Methuen, 1963. ———. “Sensibility, Understanding and the Doctrine of Synthesis: Comments on Henrich and Guyer.” In Kant’s Transcendental Deductions, edited by E. Förster. Stanford: Stanford University Press, 1989. Surber, J. P. Language and German Idealism: Fichte’s Linguistic Philosophy. Atlantic Highlands, N.J.: Humanities Press, 1996. Sutherland, D. “Kant’s Philosophy of Mathematics and the Greek Mathematical Tradition.” Philosophical Review 113 (2004). Svendsen, L. F. H. “Kant’s Critical Hermeneutics: On Schematization and Interpretation.” Dissertation, Faculty of Arts, Oslo, 1999. Swing, T. K. Kant’s Transcendental Logic. New Haven: Yale University Press, 1969. Tarbet, D. “The Fabric of Metaphor in Kant’s Critique of Pure Reason.” Journal of the History of Philosophy 6 (1968). Tarski, A. Introduction to Logic. New York: Oxford University Press, 1965. Taschek, W. “Making Sense of Others: Donald Davidson on Interpretation.” Harvard Review of Philosophy 10 (2002). Tennant, N. “Logic and Its Place in Nature.” In Kant and Contemporary Epistemology, edited by P. Parrini. Dordrecht: Kluwer, 1994. ———. “Rule-Circularity and the Justification of Deduction.” Philosophical Quarterly 55 (2005). Thompson, M. “On A Priori Truth.” Journal of Philosophy 78 (1981). ———. “Singular Terms and Intuitions in Kant’s Epistemology.” Review of Metaphysics 26 (1973): 314–43. Reprinted with postscript in Kant’s Philosophy of Mathematics: Modern Essays, edited by C. Posy. Dordrecht: Kluwer, 1992. ———. “Things in Themselves.” Proceedings of the American Philosophical Association 57 (1983). ———. “Unity, Plurality, and Totality as Kantian Categories.” Monist 72 (1989). Tonelli, G. “Conditions in Königsberg and the Making of Kant’s Philosophy.” In bewußt sein: Gerhard Funke zu eigen, edited by A. Bucher, et al. Bouvier: Bonn, 1975. ———. “Das Wiederaufleben der deutsch-aristotelischen Terminologie bei Kant während der Enstehung der ‘Kritik der reinen Vernunft.’” Archiv für Begriffsgeschichte 9 (1964). ———. “Die Voraussetzungen zur kantischen Urteilstafel in der Logik des 18.Jahrhunderts.” In Kritik und Metaphysik. Berlin: De Gruyter, 1966. ———. Kant’s Critique of Pure Reason within the Tradition of Modern Logic. Edited by David Chandler. Hildesheim: Olms, 1994. Trentman, J. A. “Speculative Grammar and Transformational Grammar: A

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Bibliography   221 Comparison of Philosophical Perspectives.” In History of Linguistic Thought and Contemporary Linguistics, edited by H. Parret. Berlin: De Gruyter, 1975. Tuschling, B., ed. Probleme der “Kritik der reinen Vernunft.” Berlin: De Gruyter, 1984. Tversky, A., P. Slovic, and D. Kahneman. Judgments under Uncertainty. Cambridge: Cambridge University Press, 1982. Vaihinger, H. Commentar zu Kants Kritik der reinen Vernunft. Stuttgart: Spemann, 1881. Van Frassen, B. “Presupposition, Implication, and Self-Reference.” Journal of Philosophy 65 (1968). Vázquez Lobeiras, M. J. Die Logik und ihr Spiegelbild. Frankfurt a.M.: Lang, 1998. Velkley, R. Freedom and the End of Reason. Chicago: University of Chicago Press, 1989. ———. “Kant on the Primacy and the Limits of Logic.” Graduate Faculty Philosophy Journal 11 (1986). Vorländer, K. Kant und Marx. Tübingen: J. C.B . Mohr [Paul Siebeck], 1926. Walker, R. C. S. Kant. London: Routledge and Kegan Paul, 1978. Walsh, W. H. Kant’s Criticism of Metaphysics. Chicago: University of Chicago Press, 1976. ———. “Philosophy and Psychology in Kant’s Critique.” Kant-Studien 57 (1966). Warda, A. Immanuel Kants Bücher. Berlin: Martin Breslauer, 1922. Washburn, M. “The Problem of Self-Knowledge and the Evolution of the Critical Philosophy 1781–1787.” Dissertation, University of California San Diego, 1970. Wedin, M. “Aristotle on the Firmness of the Principle of Non-Contradiction.” Phronesis 49 (2004). Wilson, K. “The Mistaken Simplicity of Kant’s Enthymematic Treatment of the Second and Third Figures.” Kant-Studien 66 (1969). Wittgenstein, L. Culture and Value. Translated by P. Winch. Chicago: University of Chicago Press, 1980. ———. On Certainty. Translated by D. Paul and G. E. M. Anscombe. New York: Harper and Row, 1969. ———. Philosophical Investigations. Translated by G. E. M. Anscombe. New York: MacMillan, 1958. ———. Remarks on the Foundations of Mathematics. Translated by G. E. M. Anscombe. Cambridge, Mass.: MIT Press, 1983. ———. Tractatus Logico-Philosophicus. Translated by D. Pears and B. F. McGuiness. London: Routledge and Kegan Paul, 1961. Wolff, M. “Der Begriff des Widerspruches in der ‘Kritik der reinen Vernunft.’” In Probleme der “Kritik der reinen Vernunft,” edited by B. Tuschling. Berlin: De Gruyter, 1984. Wolff, R. P., ed. Kant: A Collection of Critical Essays. New York: Doubleday, 1967.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

222   Bibliography

Copyright © 2008. Catholic University of America Press. All rights reserved.

Wubnig, J. “The Epigenesis of Pure Reason.” Kant-Studien 60 (1969): 147–52. Young, J. M., ed. and trans. Lectures on Logic. Cambridge: Cambridge University Press, 1992. Zöller, G. “Kant on the Generation of Metaphysical Knowledge.” In Kant: Analyse—Probleme—Kritik, edited by H. Oberer and G. Seel. Würzburg: Könighausen and Neumann, 1988.

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Index

Aarselff, H., xviii Allison, H., 6n Alston, W., 181 Ameriks, K., xvin3, 135, 207 Amrhein, H., 25n35 Antony, L., 185–86 Apperception, 8n7, 13, 20, 25n35, 97n6, 104–5, 109, 116–17, 120, 121–22, 150n16, 174–76, 192 Aristotle, xvii, 1n, 12n16, 34, 36–37, 42–43, 54, 57, 67, 74–75, 77, 78, 88–92, 100, 112n34 Arnauld, A., 75 Ashworth, E. J., 73 Atkinson, R. F., 16n23 Autonomy, xix, 132n72, 135n75, 171–72, 185 Bacon, R., 66 Beck, L. W., 16n23, 18n25, 25n35, 28, 84n47, 115n41, 132n72, 162 Beck, J. S., 104 Beiser, F., 28n39 Bennett, J., xix, 100 Berkeley, G., 28, 122n52 Beth, E. W., 36 Bird, G., 103 Bochenski, J., 165 BonJour, L., xx, 133n73, 139–53, 154, 168, 174n56 Boyne, R., 186

Brandt, R., 39n7, 60n4, 77, 94 Brown, N., 125–26 Brucker, J., 76 Buchdahl, G., 23n33 Carl, W., 145n12 Cassirer, E., 112n35, 132n72 Categories, derivation of, 97–101. See also Table of categories Causality, 16, 115, 173, 200 Cherniak, C., 87n50 Chomsky, N., 61, 126–28 Chrysippus, 30n42 Clark, M., 169n49 Code, L., 156–57 Collins, A., 170n51 Common sense, xvin3, 118, 129–30, 135, 205–7 Conceptual schemes, xxi, 153–66 Copernican Revolution, 3, 18, 20, 26, 33, 67, 70, 72, 127, 142, 204 Cutrofello, A., 157n29 Davidson, D., xvii, xx–xxiii, 153–62, 164–66, 168 Deduction: clue to metaphysical deduction; 79, 99–100, 114–16; metaphysical; x. xix, 63–64, 93–136; transcendental; 8n7, 15, 19, 25, 32, 94, 116 Derrida, J., 170, 177n62 Descartes, R., 28, 127, 144, 152n21, 181

223

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

224   Index De Vleeschauwer, H. J., xvn1 Diamond, C., 46

Jevons, W., 42 Jungius, J., 76, 85

Engel, P., 118n45 Epicurus, 76, 80 Epistemology, x, 48, 128, 143, 145–46, 152, 160–62, 170, 182, 198, 203; naturalized, 48, 119 Ewing, A. C., 8n7

Kemp Smith, N., 36, 95 Kitcher, P., 5, 26n35 Körner, S., 88 Korsgaard, C., 24n34 Kripke, S., xiii, 158n32 Krüger, L., 102n16 Kuehn, M., xvin3, 130n66, 207

Copyright © 2008. Catholic University of America Press. All rights reserved.

Farrell, D., 83n44 Fichte, J., 104–5, 122n52 First-person, 11, 135, 163–64, 204 Flax, J., 171–72, 174 Foucault, M., xxiii, 170, 177–87, 189n1 Foundationalism, 152–53, 177, 179, 181–82, 193, 204, 208 Freedom, xi, 126, 131–32, 171–72, 186–87 Frege, G., 44, 50, 54, 121, 161n38 Friedman, M., 90n53, 103n18 General logic. See Logic, general Genova, A., xvn1 Gibbs, R., 190–91, 193 Gilead, A., 108 Goldmann, L., 187 Good, T., 78 Gram, M., 201 Grammar, xviii, 56–73; transformational, 69n17; universal, 59, 65–70, 72 Grimm, R., 170 Guyer, P., 196–205 Hamann, J. G., xvii–xviii, 151, 168 Harrison, R., 124n56 Hegel, G. W. F., xvii, 11, 36, 112, 122n52, 142, 160n35, 193 Heidegger, M., 99, 128, 170 Henrich, D., 96n4, 136 Hintikka, J., 17n24 Horstmann, R.-P., 94–95, 99n10 Hume, D., 14–18, 22n31, 116, 141, 144, 146n13, 176 Jacobi, F. H., 197, 204 Jameson, F., 175n58

Lear, J., 124 Leibniz, G. W., 2, 8–10, 22, 66, 144 Lloyd, G., 175–76 Locke, J., 2, 10–18, 22, 25n35, 54, 123, 144 Logic: Aristotelian, xvii, 34–37, 42–43, 53–54, 57, 74, 77–78, 88–92; normativity of, 121; Stoic, 34, 37, 39n8, 53, 71 Logic, general: completeness of, 91–92, 101–3; formality of, 103–6; in contrast to quantificational 103n18, 114–15; in contrast to transcendental, ix, xiv, xix, 1, 31, 34, 62–64, 79–81, 83n44, 86, 90n53, 94, 111, 116–17, 133–34, 138, 151, 167, 184, 195n12; justification of, 123–24; priority of general and transcendental logic, 108–9 Logica docens/Logica utens, 74, 79, 87, 118n45, 136 Longuenesse, B., 60n4, 75n27 Machina, K., 48–49 Mason, R., 128–29 Mathematics, 10–11, 15–18, 50, 81–83, 86, 113n36, 146n14, 160n35, 183n72, 203 McGowan, J., 173, 185, 207 Meier, G. F., 75, 100 Melnick, A., xxiii–xiv, 8n7, 21–22, 64, 114–15, 143, 199, 204 Mendelssohn, M., 188 Metacritique, 1n1, 28n39, 118 Modality, xiv, 29, 77, 110, 113, 122n53, 158–59, 170 Modesty, 77, 188–207 Modistae School, 67–73, 127 Moser, P., 181n71

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Index   225 Nagel, T., 149 Natural language, xvii, 64, 67–69, 72, 78, 87 Nicole, P., 75 Nietzsche, F., 168–70 Non-contradiction, principle of, 1, 6, 42–43, 46, 48–55, 60, 107, 137, 147, 149, 166 Norris, C., 185n75 Oberhausen, M., 73 Objective reality, 84 Objective validity, 28–33, 61 Objectivity, 47n23, 125, 208

Copyright © 2008. Catholic University of America Press. All rights reserved.

Parson, C., 17n24 Paton, H. J., 38, 80, 99, 106, 110n31 Peirce, C. S., xvii, 36, 79, 197 Persius, 208n39 Pinborg, J., 67, 69n17 Pinker, S., 187n83 Pippin, R., 99 Plato, 49, 66 Poellner, P., 169n49 Postmodernism, 47n23, 168–77, 187 Pozzo, R., 73, 75n30, 79n38, 79n39 Prauss, G., 31, 85–86 Priest, G., 51–54 Putnam, H., 45–46, 49, 149, 170n51 Quine, W. V., 47–49, 78n37, 92n55, 123, 140, 157, 159 Ramsey’s Maxim, 26 Rationality, xxi, 41–43, 50, 53–55, 60, 69, 78, 87, 129, 163, 165, 166n47, 174n56, 209 Realism: empirical, xxiii–xxiv, 105n22, 122n52, 145, 153n23, 179, 204; transcendental, 23, 144, 150, 201 Reich, K., xvn1, 10n10, 102n16, 130n67 Rescher, N., 27 Ricoeur, P., 107n27 Rorty, R., 145, 159–62, 176n61, 181–82, 198 Rotenstreich, N, 108n28

Rousseau, J.-J., xix, 130–33, 135–36, 186, 205 Routley, R., 51–54 Russell, B., 57 Schmitt, F., 152 Schopenhauer, A., xix, 170, 192 Schott, R., 163 Sellars, W., 144n10, 160n36 Sextus Empiricus, 192 Shamoon, A., 77 Skepticism, xxiii, 4, 11, 14, 21, 22n31, 47n23, 72, 118n44, 133n73, 139–43, 152, 174, 183, 187, 209 Steiner, M., 16n23 Strawson, P. F., 5n, 36, 91–92, 97, 103 Svendsen, L., 64–65, 72 Swing, T. K., 97 Table of categories, xix–xx, 63, 92, 94, 96–100, 103, 109, 112 Table of judgments, xix–xx, 38, 63–64, 94, 96–101, 103, 105–6, 110, 112n34, 113–14 Tarbet, D., 181n70 Tarski, A., 98n7, 152, 155n26, 166, 169n49 Taschek, W., 155 Tennant, N., 138 Thing-in-itself, 21–22, 122n52, 143n9, 161, 170n51, 196–205 Thompson, M., 10n10, 16n23, 17n24, 45–46, 49–50, 53–54, 82, 84n46, 89–90, 103n18, 122n52, 163, 200 Tonelli, G., 35n3, 36, 46, 53, 73–74, 76nn32, 34, 97 Transcendental argument, 45, 145, 159n33, 201 Transcendental idealism, 23, 26n36, 105n22, 117n43, 122n52, 176n61, 196–97, 202 Transcendental unity of apperception, 8n7, 104, 109, 116–17, 121–22, 150n16, 174 Trendelenburg, F., 197 Truth, ix, 30–32; 61–62, 83–86, 107–8, 134–35, 155n26, 161–62, 171–72,

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

226   Index Truth (cont.) 174n46, 183, 195; a priori, 45–46; Nietzsche and, 169n49; theories of, 138–40, 145, 148–49, 152–53

Young, J. M., 97

Copyright © 2008. Catholic University of America Press. All rights reserved.

Vaihinger, H., 11 Velkley, R., 46

Wagner, R., 169 Walker, R., 97 Walsh, W. H., 25n35, 206–7 Wittgenstein, L., xiii, 45–46, 55, 119, 123– 25, 151, 164, 176, 191–94, 208n39, 209

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of

Copyright © 2008. Catholic University of America Press. All rights reserved.

Necessity & Possibility: The Logical Strategy of Kant’s “Critique of Pure Reason” was designed and typeset in Arno Pro by Kachergis Book Design of Pittsboro, North Carolina. It was printed on 60-pound Natures Natural and bound by Thomson-Shore of Dexter, Michigan. e

Mosser, Kurt. Necessity and Possibility : The Logical Strategy of Kant's Critique of Pure Reason, Catholic University of