Está en la página 1de 800


Bogazici University library


1!lI~mllJ~MH~J~lllml ~

.:. The mOSt accurate and informative English translation of Kant's most important philosophical work in both the 1781 and 1787 editions. :. Faithful rendering of Kant's terminology, syntax, and serrrenc~ structure. :. A simple and ~~~t :;rle suitable for readers at alJ levels. :. Distinct versions of all those portions of the work 'substantially revised"by Kant for the 1787 edition. :. All Kant's handwritten emendahons and marginal notes from his own personal copy reproduced for the first time in any edition, German or English. .) A large~scale introduction providing a -\ummary of the suucmre and arguments of the Critiqlte as well as the most informative account available in English of its long and complex genesis.
) An extensive editorial apparatus including informative

annotation and glossaries.

ISBN 0-521-65729-6


This entirely new translation of the Critique afPure Reason is the most accurate and informative English translation ever of this will make epochal philosophical text. Though its simple and direct it suitable for all new readers of Kant, the translation displays an unprecedented philosophical and textual sophistication that will enlighten Kant scholars as well. Through the comparison of the best modern German editions to the original 1781 and 1787 versions of the text, and careful attention to the precise translation of Kant's terminology, as well as rendering of the structure and syntax of Kant's prose, this translation recreates as far as possible a text with the same interpretative nuances and richness as the originaL Moreover, by including the complete text of handwritten emendations and marginal notes made by Kant in his own even no personal copy of the first edition, this volume does German edition has ever done: furnish the reader with a text as dose as possible to the one present in Kant's own library. The Cambridge Edition places the reader in the most mclepen1detlt yet best informed interpretative position by presenting entirety rate (though meticulously cross-referenced) versions of all the n{)ltln,n~ of the work that Kant revised heavily for the second edition: prefaces, the introduction, Transcendental Aesthetic, 'Ihnscendental Deduction, the chapter on Phenomena and Noumena, and the Paralogisms of Pure Reason. The extensive editorial apparatus includes informative amtlot:atton, detailed glossaries, a thorough but perspicuous index, a large-scale general introduction in which two of the world's preeminent Kant scholars provide a succinct summary of the structure and argument of the Critique as well as a detailed account of its long and complex genesis.


and Education Natural Science Lectures on Logic Lectures on MI?tal~hysics Lectures on Ethics

Critique of Pure Reason



General editors: Paul Guyer and AlIen W Wood '

/J,1m ,,11''.1

board: Henry Allison Reinhard Brandt Ralf Meerbote Charles D. Parsons Hoke Robinson J. B. Schneewind





Yale University

Bogazici University Library

ll"~llll CAMBRID


The Pin Building, Trumpington Street, Cambridge, CB2 rRP, United Kingdom

The Edinburgh Cambridge CE2 2RU, United Kingdom 40 West 20th Street, New York, NY lOO II -42 II, USA 10 Stamford Road, Oakleigh, Melbourne Australia Ruiz de Alarcon 13, 28014 Spain
CamtJri(:lge University Press 1998



H' .,.

is in copyright. Subject to statutory exception th/:: p]rovCisi,~ms of relevant collective licensing agreements, n\l,IeprcJdL,ction of any part may take place without;:W,.itt:en permission of Cambridge University Press. First published 1998 Reprinted 1998 First paperback edition 1999 Reprinted 2000 (twice) Printed in the United States of America Typeset in Janson
Library Cau/()f!.i1"f!:-,in-.f'u/,/iClltion Data


Kant, Immanuel, 1724~I804. [Kritik der reinen Vernunft. English] The critique of pure reason / edited [and translated] by Paul Guyer, Alien W. Wood. p. cm. - (The Cambridge edition of the works of Immanuel Kant) Includes bibliographical references and index. ISBN 0-521-3542-1 (hardcover) 1. Knowledge, Theory of. 2. Causation. 3. Reason. I. Guyer, Paul, 1948- Il. Wood, AlIen W Ill. Title. IV: Series: Kant, Immanuel,1724-1804 Works. English. "992. B2778.E5G89 1998 r 2 I - dC2 I 97-2959
A catalog '-ecord for this book is the British Library

ISBN 521 35402 1 hardback ISBN 0521 657296 paperback


General editors' preface Acknowledgments Introduction, by Paul Guyer and AlIen WWood
Note on translation Bibliography


73 77

Immanuel Kant, Critique ofPure Reason Editorial Notes Glossary Index





General editors'


Within a few years of the publication of his Critique Reason in 1781, Immanuel Kant (1724-184) was recognized by contemporaries as one of the seminal philosophers of modern times - indeed as one of the great philosophers of all time. This renown soon spread beyond Germ,m-speaking lands, and translations of Kant's work into English were published even before 1800. Since then, interpretations of Kant's views have come and gone and loyalty to positions and waned, but his importance has not diminished. Generations of scholars have devoted their efforts to producing translations of Kant into English as well as into other languages. There are four main reasons for the present of Kant's Wfltu-, 0-<:'
I. Completeness. Although most of the works in Kant's ones more than time have been translated before, the most once, only fragments of Kant's many important works have ever been translated. These include the Opus postumum, Kant's unfinished magmnn OpUJ on the transition from philosophy to physics; transcriptions of his classroom lectures; his correspondence; and his a commarginalia and other notes. One aim of this edition is to first prehensive sampling of these materials available in English for time.
2. Availability. Many English translations of Kant's especially those that have not individually played a large role in the sulbsc;qllerlt development of philosophy, have long been inaccessible or out Many of them, however, are crucial for the understanding of Kant's philosophical development, and the absence of some from l~ngh:5h language bibliographies may be responsible for erroneous or ered traditional interpretations of his doctrines by English-speaking philosophers.

3. Organization. Another aim of the present edition is to make Kant's published work, both major and minor, available in comprehensive volumes organized both chronologically and topically, so as to faCilItate serious study of his philosophy by English-speaking readers.


General editors' preface

many of Kant's works have most distinguished scholars of their some of been translated these translations are now and there is considerable terminological disparity among them. Our aim has been to enlist some of the most accomplished Kant scholars and translators to produce new translations, freeing readers from the philosophical and literary preconceptions of previous generations and allowing them to approach texts, as far as possible, with the same directness as present-day readers of the German or Latin originals. In pursuit of these goals, our editors and translators to follow several principles: 1. As far as seems advisable, edition employs a single general glossary, especially for Kant's technical terms. Although we have not attempted to restrict the prerogative of editors and translators in choice of terminology, we have maximized consistency by putting a single editor or editorial team in charge of each of the main groupings of Kant's writings, such as his work in practical philosophy, philosophy of religion, or natural science, so that there will be a high degree of terminological consistency, at least in dealing with the same subject matter. 2. Our translators try to avoid sacrificing literalness to readability. We hope to produce translations that approximate the originals in the sense that they leave as much of the interpretive work as possible to the reader. 3. The paragraph, and even more the sentence, is often Kant's unit of argument, and one can easily transform what Kant intends as a continuous argument into a mere series of assertions by breaking up a sentence so as to make it more readable. Therefore, we try to preserve Kant's own divisions of sentences and paragraphs wherever possible. 4. Earlier editions often attempted to improve Kant's texts on the basis of controversial conceptions about their proper interpretation. In our translations, emendation or improvement of the original edition is kept to the minimum necessary to correct obvious typographical errors. 5. Our editors and translators try to minimize interpretation in other ways as well, for example, by rigorously segregating Kant's own footnotes, the editors' purely linguistic notes, and their more explanatory or informational notes; notes in this last category are treated as endnotes rather than footnotes. We have not attempted to standardize completely the format of individual volumes. Each, however, includes information about the context in which Kant wrote the translated works, a German-English glossary, an English-German glossary, an index, and other aids to comprehension. The general introduction to each volume includes an explanation of specific principles of translation and, where necessary, principles of selection of works included in that volume. The pagination of the stanvm

4. Consistency

General editors' preface

dard German edition of Kant's works, Kant's (ie.,anznz,ette :",c!;'rzf:ten. by the Royal Prussian (later German) Acadl~mv of Sciences Georg Reimer, later WaIter de Gruyter & Co., 1900- ), is indicated throughout by means of marginal numbers. Our aim is to produce a comprehensive of Kant's wr-ttmlp's. embodying and displaying the high standards Kant ~f'hrd .. ,._ ship in the English-speaking world during the second of the twentieth century, and serving as both an instrument and a for the further development of Kant studies by English-speaking readers in the century to come. Because of our emphasis on literalness of trans[aulon and on information rather than interpretation in editorial practices, we hope our edition will continue to be usable despite evolution and occasional revolutions in Kant scholarship.




This translation of Kant's Critique ofPure Reason is the work of both of us over many years, during which we have had the of many students, friends, and colleagues. Those generous of their time and effort are owed special helped us in one way or another in the preparation and revision of the translation are Giinter Zoller, Charles Parsons, Stephan Wagner, students in Paul Guyer's Kant classes at the University of Pennsvlvania in 1994-95, and the students in AlIen Wood's "German PhilosOI)hical TimmerTexts" classes at Cornell University in 1990 and 1992. man made available to us detailed corrections of the RayrrlUDld :,chlfi1c1t (Meiner) edition of the German text, and Georg Mohr nu,,,i,;,,,; corrections of the Ingeborg Heidemann (RecIam) edition. people, induding Lewis White Beck, Rolf George and Martin We:atller~ston, offered us corrections of earlier English translations. Cooper Rega \Vood helped us with the identification and attribution of classical quotations. The trustees of the Florence R. C. Trust endowed the research fund that paid for facsimiles of original edJttic)l1S of the Critique and other research materials. Finally, special to Allison Crapo and Cynthia Schossberger for their generous and meticulous help with the proofreading and to Michael Rohlf and Feszczak for their help in identifying further corrections.


Critique of Pure Reason


Immanuel Kant's Critique of Pure Rem-on is one of the seminal and monumental works in the history of Western in May 1781, when its author was already fifty-seven years stantially revised for its second edition six years later, the the culmination of three decades of its author's very and the starting-point for nearly two more decades of his evolving but now very public philosophical thought. In the more than two centuries since the book was first published, it has been the constant ject of scholarly interpretation and a continuous source inventive philosophers. To tell the whole story of the book's mtlU(~n(:e would be to write the history of philosophy since Kant, and that is beyond our intention here. After a summary of the Critique's structure and argument, this introduction will sketch its genesis and evolution from Kant's earliest metaphysical treatise in 1755 to the of the first edition of the Critique in 1781 and its revision for the second edition of 1787.

The strategy of the Critique. In the conclusion to his second Cntlqlle, the Critique of Practical Reason of 1788, Kant famously wrote, "Two and awe things fill the mind with ever new and increasing the more often and more enduringly reflection is occupied them: the starry heavens above me and the moral within me.'" motto could just as well have served for all of Kant's sophical works, and certainly for the Critique of PUre Reason. From the outset of his career, Kant had been concerned to resolve a number of the most fundamental scientific controversies of his epoch to establish once and for all the basic principles of scientific knowledge of the world, thereby explaining our knowledge of the "starry heavens."

Introduction career, Kant was intent on showin g Almost as early in also presupp osition of moralit y freedom, unders tood not only as compat is as the ultimat e value served and advanced by the moral law, was Reason Pure of ible with the truth of modern science. The Critique the the work in which Kant attemp ted to lay the foundations both for . freedom certain ty of modern science and for the possibility of human of xity The book is complex, however, not just because of the comple Kant's own position, but also because he argues on several fronts against several different alternative positions represe nted in early modern phiwithin the Germa n Enligh tenmen t in particular. losophy generally his own dualistic defense of both modern sciIn order to make room felt ence and human autonomy, Kant, like Descartes, Locke, and Hume, was which he had to rein in the pretens ions of traditional metaphysics, and represe nted for him by the school of Christi an Wolff (1679-1754) 762).2 his followers, especially Alexander Gottlie b Baumg arten (1714-1 the Their position, which Kant called "dogmatism," was compar ed in Preface to the Critique to the despotic ministr y of an absolute monarchy - Kant held dogma tism to be caprici ous, opinion ated, faction l rationa ridden and conseq uently unstable and open to the contem pt of observers. Yet Kant wanted to distinguish his own critical stance toward dogma themtism from several other ways of rejecting it, which he regarde d as is selves equally danger ous to the cause of reason. The first of these advoto skepticism, the positio n Kant took David Hume (17II-1 776) cate.' Anothe r position Kant rejected was empiricism, which unders tood the "way of ideas" described in John Locke's (1632- 174) Essay concern on ing Human Understanding (1790) as ground ing knowle dge solely ideas acquired in the course of individual experience. Yet anothe r philosophica l stance Kant encoun tered was what he called indifferentism, which did not reject metaphysical assertions themselves but did reject he any attemp t to argue for them systematically and rigorously. Here subin had in mind a numbe r of popula r philosophers who were often the stantive agreem ent with dogmatists on metaphysical issues such as unconexistence of God and the immort ality of the soul, but who were and vinced by the scholastic subtlet y of the dogmat ists' propos itions need we proofs, holding instead that the beliefs on these matters that h for the successful conduc t of human life are simply given throug 4 "health y underst anding " or commo n sense. Yet while he attemp ted to criticize and limit the scope of traditio nal metaphysics, Kant also sought to defend against empiricists its underly what ing claim of the possibility of universal and necessary knowledge of Kant called a priori knowledge, knowledge origina ting indepe ndently experiexperience, because no knowledge derived from any particular and ence, or a posteriori knowledge, could justify a claim to universal

Introduction necessary validity. He sought likewise to defend its scientific chalracter against skeptics who dismiss its rigorous arguments as msuttlCllent against proponents of "common sense" who regard them as pellauluc and superfluous. As Kant compared dogmatic metaphysicians to deany fenders of despotism, so he likened skeptics to nomads who form of permanent civil society and are to or overthrow the monarchy of metaphysics, and Lockeans to calunmiators who would foist a false and degrading genealogy on the monarch. Those who would pretend indifference to metaphysical he char~~ed with being closet dogmatists, like supporters of a corrupt regime scoff at its defects and feign ironic detachment from it but have no independent convictions of their own. Kant's position thus required him not only to undermine the arguments of traditional metaphysics but also to in their a scientific metaphysics of his own, which establishes can a priori but also limits it to that which is required for ordinary ence and its extension into natural science. Kant therefore had to way to limit the pretensions of the dogmatists while still deterldlng metaphysics as a science which is both possible (as was denied skeptics) and necessary (as was denied by the indifferentists). Thus Kant had to fight a war on several different fronts,S in he had to establish the unanswerability of many metaphysical questions against both dogmatists and empiricists but also defend parts of the positions he was attacking, such as the possibility of a priori cognition of the fundamental principles of natural science, against both empiricists skeptics. And while he wanted to prove to the indifferentists that a science of metaphysics is important, he also wanted to embrace of their position, since he thought that in regard to some insoluble metawe can defend physical questions, indeed the most important a kind of commonsense belief - in God, freedom and imml)rt:alil:y cause our moral outlook has an inescapable stake in them. The structure of the Critique. This complex program led to the enormous complexity of the structure and argument of the Critique of Pure Reason. To many readers, the elaborate structure or "architectonic" of the Critique has been a barrier to understanding it, a brief account of the origin of the main divisions of the book can illuminate its contents. Although these contents are profoundly original, Kant actually borrowed much of the book's structure from well-known models. After the preface (which was completely rewritten for the second edition) and the introduction, the Critique is divided into two main parts, the "Doctrine of Elements" and tlle "Doctrine of " This distinction is a variation on a distinction common in German logic textbooks between "general logic" and "special" or applied logic;6 in Kant's hands, it becomes a rubric to distinguish between his fundamental ex3

Introduction cogniti on and its limits, in the a theory positio n of method ologica l his own reflections on "Doctr ine of Elemen ts," "Doctr ine of the of rubric implica tions of that theory, under the atical and mathem n betwee ts Metho d," where he provide s contras ng, as reasoni l practica and ical philosophical proof and betwee n theoret emic, dogmat and method well as contrasts betwee n his own critical pirical, and skeptical method s of philosophy. The "Doctr ine of Elemen ts" in turn is divided into two main (altal though very disprop ortiona tely sized) parts, the "Trans cenden rs conside which of first the the "Transc endenta l Logic," Aesthetic" ty, sensibili our of forms fundam ental the a priori contrib utions namely space and time, to our knowledge, and the second of which considers the a priori contrib utions of the intellect, both genuin e and spurious, to our knowledge. This division is derived from Baumgarten's or introdu ction of "aesthetics" as the title for the science of "lower" conor higher of "sensitive cogniti on" in contras t to logic as the science ceptual cognition;7 at the time of writing the Critique, however, Kant rejected Baumgarten's supposition that there could be a science of taste for (what we now call "aesth; tics"), and instead approp riated the term dge knowle to lity the contrib ution of the forms of sensibi his theory in general. s After a brief explana tion of the distinction betwee n "genscierallog ic" and "transc endenta l logic" - the former being the basic being latter the and ence of the forms of though t regardless of its object the science of the basic forms for the though t of objects (A 50-57! B74-lh ) - Kant then splits the "Transc endenta l Logic" into two main tal divisions, the "Trans cenden tal Analyt ic" and the "Trans cenden thsixteen a from derives Dialectic." Kant uses this distinction, which century Aristotelian distinction betwee n the logic of truth and the logic of probability, represe nted in eightee nth-cen tury Germa ny by the Jena n professor Joachim Georg Darjes (I714-I 792),9 to distinguish betwee ation cooper in g the positive contrib utions of the underst anding , workin and with sensibility, to the conditions of the possibility of experience t of attemp s spuriou the knowledge (the "Transc endenta l Analytic") and inysical metaph provide reason workin g indepen dently of sensibility to tal cenden "Trans (the sight into things as they are in themse lves two into divided turn in Dialectic"). The "Transc endenta l Analytic" is the books, the "Analytic of Concep ts" and the "Analytic of Principles," pure the of validity ry first of which argues for the universal and necessa of concepts of the underst anding , or the categories, such as the concepts validthe for argues substance and causation, and the second of which ity of fundam ental principles of empirical judgment employ ing those and categories, such as the principles of the conservation of substance the universality of causation. The "Transc endenta l Dialectic" is also divided into two books, "On

Introduction Pure Reason" "On the Dialectical Injrenenc:es the Concepts Pure Reason," in which Kant explains how pure reason geller'atc:s of metaphysical entities such as the soul, the as a and then attempts to prove the reality of those ideas terns of inference which are valid within the limits of beyond those limits. But it should be noted twofold division of the "Transcendental Analytic" into the Concepts" and "Analytic of Principles" with the main Dialectic, the "Dialectical Inferences of Pure Reason," rerJUc:atc;s traditional division of logic textbooks into three sections on cor,!cej)ts, judgments, and inferences: IO Kant uses this structure to argue cepts of pure understanding, when applied to the forms sen.,ibility, rise to sound principles of judgment, which constitute the heart critical metaphysics, but that inferences of pure reason peJrtormed out respect to the limits of sensibility give rise to mc:taphyslcal into lusion. The treatment of inferences is in turn sections, "The Paralogisms of Pure Reason," "The ArlurlOIny of Pure Reason," and "The Ideal of Pure Reason," which expose mc:talph'ysil:alJly fallacious arguments about the nature of the soul, about the size and origin of the world as a whole, and about the existence of God, re~;pectively. These divisions are also derived from Kant's predecessors: Baumgarten divided metaphysics into "general metaphysics," or --"",-",ogy," and "special metaphysics," in turn divided into "rational pS'1chology," "rational cosmology," and "rational theology." Kant replaces their "ontology" with the constructive doctrine of his own Analytic" (see A 247/ B 303), and then presents his criticism metaphysics based on pure reason alone by demolishing the special metaphysics of rational psychology, cosmology, and theology. Finally, Kant divides the "Doctrine of Method/' .in which he retlects on the consequences of his demolition of metaphysics and reconstruction of some parts of it, into four chapters, the "Discipline," the "Canon," the "Architectonic," and the "History of Pure Reason."" The first two of these sections are much more detailed the two. In the "Discipline of Pure Reason," Kant provides an extended contrast between the nature of mathematical proof and philosophical argument, and offers important commentary on his own new critical or "transcendental" method. In the "Canon of Pure Reason," he pn~pares the way for his subsequent moral philosophy by contrasting the rn(~thod of theoretical philosophy to that of practical philosophy, and giving the first outline of the argument that runs through all three critiques, namely that practical reason can justify metaphysical beliefs about God and the freedom and immortality of the human soul although theoretical reason can never yield knowledge of such things. The last two parts of the "Doctrine of Method," the "Architectonic of Pure Reason" and the

Introduction contras ts betwee n Kant's Pure Reason ," recapit ulate "Histor y of method and those the dogmatists, empirimvn critical which he began, treating these contras ts in both cists, and skeptics systematic and historic al terms. Indeed , althoug h Kant himself never the history of philoso phy as a scholar ly discipline, in cared much the few pages of his "Histor y of Pure Reason " he outline d the history lof modern philoso phy as the transce ndence of empiric ism and rationa aluse, still we that pattern ism by his own critical philoso phy, the heirs though of course we also have to add room to this pattern for the and successors of Kant's own philosophy. in With this analysis of the organiz ation of the Critique ofPure Reason s. content hand, we now provide a brief resume of its gh "Introd uction ": the idea of transce ndenta l philoso phy. Althou crithe , project his of side e Kant himself often suggests that the negativ pretique of dogmat ic metaphysics, is the most import ant, the Critique human of ts elemen sents Kant's positiv e doctrin e of the a priori atiknowle dge first. In the introdu ction, Kmt argues that our mathem judgcertain s cal, physical, and quotidi an knowle dge of nature require ments that are "synthe tic" rather than "analytic," that is, going beyond ts inwhat can be known solely in virtue of the conten ts of the concep identity of les princip volved in them and the applica tion of the logical that and contrad iction to these concep ts, and yet also knowab le a priori, expelar particu no since is, indepe ndently of any particu lar experie nce ry rience could ever be sufficient to establish the universal and necessa ic synthet how of n questio validity of these judgme nts. He entitles the reason" pure of problem a priori judgme nts are possible the "genera l it (B 19), and propos es an entirel y new science in order to answer (AIO-I 6/B 24-30). This new science, which Kant calls "transc endenta l" (A I I IB 25), does the not deal directly with objects of empirical cogniti on, but investigates the ing examin by them conditi ons of the possibility of our experience of obof on cogniti any have mental capacities that are require d for us to dge, jects at all. Kant agrees with Locke that we have no innate knowle by us in ted implan itions that is, no knowle dge of any particu lar propos experiual individ our of t God or nature prior to the comme ncemen g ence. I2 But experience is the produc t both of externa l objects affectin e respons in s facultie ve our sensibility and of L~e operati on of our cogniti a or "pure" have can we that to this effect (AI, BI), and Kant's claim is operathe by made nce priori cogniti on of the contrib utions to experie l tion of these faculties themselves, rather than of the effect of externa into ies capacit ve cogniti our objects on us in experience. Kant divides our receptiv ity to the effects of external objects acting on us and giving al us sensations, throug h which these objects are given to us in empiric by n intuitio of data the intuitio n, and our active faculty for relating

Introduction thinking them under concepts, which is 33), and forming judgments about them. As already suggested, this vision is the basis for Kant's division of the "Transcendental Doctrine of Elements" into the "Transcendental Aesthetic," which deals with sensibility and its pure form, and the "Transcendental Logic," deals with the operations of the understanding and as the spurious and the legitimate activities of theoretical reason. "Transcendental Aesthetic": space, time, and tr:lm>Ct~w::l.eJtlt:l1 idealism. Despite its brevity - a mere thirty pages in the first edition and forty in the second - the "Transcendental Aesthetic" argues for a series of striking, paradoxical and even revolutionary theses mine the course of the whole remainder of the Critique and that been the subject of a very large proportion of the scholarly work devoted to the Critique in the last two centuries. '3 In this section, Kant attempts to distinguish the contribution to cognition made our receptive faculty of sensibility from that made solely by objects time are pure of affect us (A2 I-2/B 36), and argues that space all intuition contributed by our own faculty of sensibility, forms of which we can have a priori knowledge. 'fhis is the basis for Kant's resolution of the debate about space and time raged between the Newtonians, who held space and time to be seJJ-s;ub'sistirlg' entities existing independently of the objects tl1at occupy Leibnizians, who held space and time to be systems of relations, conceptual constructs based on non-relational properties in the things we think of as spatiotemporally related. 14 Kant's alternative to both of these positions is that space and time are subsistent beare ings nor inherent in things as they are in themselves, objects of expeforms of our sensibility, hence conditions under rience can be given at all and the fundamental of their representation and individuation. Only in this way, Kant argues, can we adequately account for the necessary manifestation of space time throughout all experience as single but infinite magnitudes - the feametature of experience that Newton attempted to account for with physically incoherent notion of absolute space and time as the sensorium dei - and also explain the a priori yet synthetic character of the matical propositions expressing our cognition of the physical properties of quantities and shapes given in space and time - the epistemological certainty undercut by Leibniz's account of space and time as mere relations abstracted from antecedently existing objects (A22-5 IB 37-4I, A 30 - 2 IB 46-9). Kant's thesis that space and time are pure forms of intuition to the paradoxical conclusion that although space and time are empirically real, they are transcendentally ideal, and so are objects given in them. Although the precise meaning of this claim remains to deB

Introduction that it is only from the human bate,I5 in general terms it is the and the spatiote mporal ity time, space, standpo int that we can speak of these things not as cognize we that thus of the objects of experience, the conditions of under appear they as only they are in themselves but is Kant's famous This ). B49-73 32-48I A our sensibility (A26-3 0IB42- 5, throug hout the ed employ is which doctrin e of transcendental in a variety of s) critique uent subseq two Critique ofPure Reason (and the and "Distic" Aesthe l endenta "Transc ways, both positively, as in the ic a prisynthet of lity possibi the for t accoun cipline of Pure Reason," to l endenta "Transc the in as ely, negativ and ori cogniti on in mathematics, given nces appeara the to on cogniti our Dialectic," to limit the scope of of to our sensibility, while denyin g that we can have any cogniti on cons realitie ndent transce as is, that things as they are in thelTlselves, ve stituted as they are indepe ndently of the constit ution of our cogniti capacities. "Trans cenden tal Analyt ic": the metaph ysical and transce ndenthe tal deduct ions. The longest and most varied part of the Critique is conthe s: division main two "Transc endenta l Logic," contain ing the structive "Transc endenta l Analytic," which considers the understanding concep ts that yield a priori cognitions in conas the source of a ly junctio n with the forms of intuitio n already analyzed; and the primari of faculty the ates investig which destructive "Transc endenta l Dialectic," reason, in the first instance as a source of illusory argume nts and metaphysical pseudo-sciences, althoug h in the end also as the source of valuand able regulat ive princip les for the conduc t of human inquiry in is saw, we as c," Analyti tal practic al reasoning. The "Trans cenden with dealing ts," Concep of tic turn divided into two books, the "Analy the concepts of the underst anding , and the "Analytic of Principles," concerning the principles of the unders tanding that arise from the applica n. intuitio tion of those concepts to the forms of In the "Analytic of Concep ts," Kant presents the unders tanding as the the source of certain concepts that are a priori and are conditi ons of ts, concep basic twelve These possibility of any experience whatever. in object an of concepts ental fundam which Kant calls the categories, are conin and , objects of ts concep lar general, or the forms for any particu ic junctio n with the a priori forms of intuitio n are the basis of all synthet c" Analyti tal cenden "Trans the of a priori cogniti on. In an initial section which he named in the second edition of the (A66-8 I/B9I- 1I Critique the "metaphysical deducti on" of the categories CB 159), Kant s derives the twelve categories from a table of the twelve logical function judgall of aspects ant signific ly or forms of judgme nts, the logical s of ments. Kant's idea is that just as there are certain essential feature we which in ways onqing corresp all judgme nts, so there must be certain . objects about be may nts judgme form the concep ts of objects so that


There are four main logical features of judgments: scope of the of their P~~_~~(:~t;~::!~:EIlJIS,_ are relation, or assert a between a or between two or more subject-predicate judgments; or whether assert a p~ssjble, a~1:l!?:l, or I!{;<:{;s~ary each of heading~there are supposed to be a Juclgrnellt may lJ~ ~mive;sal, particular or slrlgLlla:l1; @!tlJt;matrve, nitc;;categorical~ ~ypothetical or dis;jUin:Cl:ivl~; a:nd;,!pl'O b,lelna tIC, toric;oi apodichc.Corresponding to rh,~~p:h"pl"p"lr.crH'" poss;ibllities, Kant holds there To be twelve fundamental categories conceiving of the quantity, quality, relation, and modality of objects 7o!B95, A8o/B I06). The plausibility of Kant's claim that there are ex'l.ctllv twelve logical functions of judgment and twelve corresponding categories for conceiving of objects has remained controversial since Kant first made it. ,6 Even if Kant establishes by this argument that we certain concepts a priori, it is a more ambitious claim that all of these conoeplts apply universally and necessarily to the objects that are given in our experience. Kant takes on this more ambitious project in the "Transcendental Deduction of the Categories," the chapter which he says in first edition of the Critique cost him the most labor (A xvi), but which he then rewrote almost in its entirety for the second edition 30/ B I 16-69) after other attempts in the intervening works, the Prolegomena to Future Metaphysics (1783) and Metaphysical foundations ofNatural Science (1786). In both versions of the Critique, although not in the intervening works, Kant centers his argument on the premise our experience can be ascribed to a single identical subject, via what he the "transcendental unity of apperception," onlyjf the elements of exas to present perience given in intuition are synthetically combined categories us with objects that are thought through the are held to apply to objects, therefore, not because these objects make the categories possible, but rather because the categories thE;m:sel'\Tes constitute necessary conditions for the representation of all possible idea of the of jects of experience. Precisely what is entailed by apperception, however, and what the exact relation between apperception and the representation of objects is, are obscure and controversial, and continue to generate lively philosophical discussion even after two centuries of interpretation. '7 Principles of pure understanding. Even if the transcendental deduction does establish that the categories do apply to all possible data it does so for experience, or (in Kant's terms) all manifolds of only abstractly and collectively - that is, it does not specify how each category applies necessarily to the objects given in experience or show

Introdu ction

that an of the categories must be applied to those objects. This is Kant's of task in Book II of the "Trans cenden tal Analyt ic," the "Analy tic "The s, Princip les." This book is in turn divided into three chapter Schematism of the Pure Concep ts of the Unders tanding ," the "System Principles of Pure Unders tanding ," and "On the Ground of the of Distinc tion of All Objects in Genera l into Phenom ena and Noume na." the In the first of these chapters Kant shows how the logical conten t of transcategor ies derived from the metaph ysical deduct ion is to be secformed into a conten t applicable to the data of our senses; in the the of all ond, he demon strates principles of judgme nt showin g that nts argume categories must be applied to our experience by means of ies that are sometimes held to prove the objective validity of the categor third indepe ndently of the prior transce ndental deduction; and in the chapter Kant draws out the consequences of the precedi ng two, arguaping that because the categories have a determ inate use only when time plied to spatiot empora l data and yet the forms of space and themselves are transcendentally ideal, the categories also have a deter, minate cognitive use only when applied to appearances ("pheno mena") in and therefo re that by means of the categor ies things as they are themselves ("noum ena") might be thought but not known. t In the "Schematism," Kant argues that the categories, whose conten nts, has thus far been derived solely from the logical structu re of judgme must be made applicable to objects whose form has thus far been speccan ified solely by the pure forms of space and time. He argues that this a ," be done by associating each categor y with a "transc endenta l schema a of form or relation in intuitio n that is an approp riate represe ntation y logical form or relation. In particular, Kant argues that each categor of must be associated with a temporal schema, since time is the form intuevery sensible intuitio n whatever, while space is the form of outer itions only. For example, the schema of the logical concep tion of ground al and consequence is the concep t of causality as rule-go verned tempor is , succession: the concep t of a cause, as opposed to that of a mere ground ing the concep t of "the real upon which, whenever it is posited, someth it is else always follows," or "the succession of the manifold insofar as second subject to a rule" (A 144!B 183)' As Kant will make clearer in the edition, however, the subseq uent chapter on the "Principles" will show that althoug h the conten t of the transce ndental schemata for the categories may be explicated in purely tempor al terms, the use of these ies schemata in turn depends upon judgme nts about the spatial propert the Thus nt. and relations of at least some objects of empirical judgme ies argume nt of the "Analytic of Principles" as a whole is that the categor space in both must and can only be used to yield knowledge of objects and time. The principles expressing the universal and necessary applily cation of the categories to objects given in space and time are precise

Introduction the synthetic a priori judgments that are to be demonstrated critical replacement for traditional metaphysics. In the second chapter of the "Analytic of Principles," the "System of All Principles," Kant organizes the principles of pure understan(1mp; under four headings corresponding to the four groups of categories. For each of the first two groups of categories, those listed "Quantity" and "Quality," Kant supplies a single "mathematical" ciple meant to guarantee the application to empirical objects of certain parts of mathematics, which are in turn supposed to be associated with under the certain parts of the logic of judgment. The first a mathematics of title "Axioms of Intuition," guarantees tllat extensive magnitudes, where wholes are measured by discrete parts, applies to empirical objects because these are given in space and time which are themselves extensive magnitudes (A 162-61 B z02-7). The general implication of tllis argument is that empirical use logical quantifiers (one, some, all) depends on the division empirical manifold into distinct spatiotemporal regions. The second ciple, under the title of the "Anticipations of Perception," guarantees that the mathematics of intensive magnitudes applies to the "real in space," or that properties such as color or heat, or forces such as weight or impenetrability, must exist in a continuum of degrees because our sensations of them are continuously variable (A 166-761 B207-18). Here Kant's argument is that since the use of the logical presence or functions of affirmation and negation is dependent on absence of sensations that come in continuously varying degrees, the empirical use of the categories of "Quality" is connected with the mathematics of intensive magnitudes in a way tllat could not have been predicted from an analysis of the logical content of these categories themselves (another example of how a synthetic a merely analytic judgment arises). Switching from "matllematical" to "dynamical" principles, section of the "System," the "Analogies of Experience," concerns the necessary relations among what is given in space and time, and thus gives expression to the necessary conditions for the application of tlle categories of "Relation" to empirical objects. Many interpreters consider this the most important section of the Critique. In the first analogy, Kant argues that the unity of time implies that all change must consist in alteration of states in an underlying substance, whose existence and quantity must be unchangeable or conserved (A 18z-6/B 224-32). In the second analogy, Kant argues that we can make determinate judgments about the objective succession of events as contrasted to merely subjective successions of representations only if every objective alteration follows a necessary rule of succession, or a causal law (A I 86-z I I 1 B Z 3z-56). In the third analogy, Kant argues that determinate judgments

Introduction states substance) in different regions of space exist sithat objects causal if such objects stand in the multaneously are possible -62). SIE2S6 (A2II-1 tion interac cal recipro relation of commu nity or to answer Kant's supply to ed suppos ly general is The second analogy the is analogy third the while y, causalit about Hume's skeptical doubts bebasis for Kant's refutati on of Leibniz's rejectio n of real interac tion "mon's Leibniz of thesis l essentia an ces tween indepe ndent substan to adology." In particular, both what the second analogy is intende d of matters been have proceed to ed suppos is prove and how the as ly intense as almost d dispute been have they exegetical controversy; ful. success is Hume to reply Kant's r whethe n the philosophical questio "System of In the first edition of the Critique, the final section of conditions s provide ht," Thoug cal Empiri of Principles," the "Postulates existence, ity, possibil of ies categor modal the of for the empirical use ies of categor the of use inate determ our and necessity, and argues the of sphere the to d confine fact in both possibility and necessi~y is given in experience (A2 18-35 lE 265-74 , which is that is, nt, 279-87). In the second edition, however, Kant inserte d a new argume the that show to ts attemp which 274-9), the "Refuta tion ofIdeal ism" (R very possibility of our consciousness of ourselves presupposes the exisreprese nted as of objects that are not tence of an external dently of our indepen exist to ed conceiv spatially outside us but are also of this artions implica the gh Althou them. subjective represe ntation s of claim in Kant's confirm to seems it , debated gumen t have been intensely l idealendenta "transc that sics Metaphy Future the Prolegomena to , idealism nal traditio unlike idealism l" ism" is a "critical" or "forma t withou n intuitio of forms as time and space implies the subjectivity of are that es ourselv from distinct objects the of denying the real existence I8 represe nted as being in space and time. In the third chapter of the "Analytic of Principles," on phenom ena and noume na, Kant emphasizes that because the categories must always on, be applied to data provided by sensibility in order to provide cogniti ndentransce the by red structu are lity sensibi and because the data of of tally ideal forms of intuitio n, the categories give us knowledge only "that literally ," omena ("phen lity sensibi things as they appear with which appears"). Althou gh throug h pure unders tanding (nays in Greek) lwe may think of objects indepe ndently of their being given in sensibi ena," ("noum entities sible non-sen such as them ity, we can never cognize g literally "that which is though t") (A23S-6o/B294-31S). The meanin g meanin the but dent, self-evi is mena" "pheno of Kant's use of the term inare they as "things not means literally it since of "noume na" is not, as depend ently of appeari ng to us" but someth ing more like "things that deny to s appear Kant Yet t." though pure they are unders tood by For the human unders tanding can compre hend things in the latter way.


this reason, Kant says it is legitimate for us to speak of noumena may in 1Il"in a negative sense," meaning things as dependently of our representation of them, but not noumena "in a pure reason itive sense," which would be things known A point of the Critique is to that we ever have knowledge of things through pure reason alone, categories to pure or empirical data structured the forms of intuition. At this point in the Critique Kant has completed constructive project, showing how synthetic a retical cognition are the necessary conditions of the ap]J!ic:ation the pure categories to sensible data structured The next of his argument is the critical demc)fistratioln metaphysics consists largely of illusions soul, tlle to acquire knowledge of all things God) as they are in themselves by the use of reason regardless the limits of sensibility. The bulk of this argument is reserved "Transcendental Dialectic," but Kant makes a start on it with esting appendix that completes the "Transcendental An",h,fll,n entltled the "Amphiboly of Concepts of Reflection" (AI6o-92/B3Hr-",-u',. this appendix Kant presents his criticism of Leibniz's mc)n,ldcllogy arguing that through a confusion (or "amphiboly") Leibniz mere features of concepts through which we think things, spc:cll:1c8l11y concepts of comparison or reflection such as "same" and "different" or "inner" and "outer," which are in fact never to but only to them through more determinate concepts, as if were features of the objects themselves. Kant rejects the Leibnizian-Wolffian account of such metaphysical concepts as essence, identity, and possibility, and reinforces his own insistence that errlDlfl-' cal individual judgments of real possibility require seIISIIJle COJlIGJltlC)fiS in addition to logical intelligibility and non-contradictoriness. The "Transcendental Dialectic": the critique nllet,!pllY~;ics. The second division of the "Transcendental Logic" turns to mam destructive task of the Critique ofPure Reason, and that gives it its of name, the task of discrediting dogmatism and displaying the metaphysics. The "Transcendental Analytic" has prepared the way for this critique of traditional metaphysics and its foundations its ment that syntlletic a priori principles can be established only limited domain of sensible experience. But Kant's aim in the "Dialectic" is not only to show the failure of a metaphysics transcends the boundaries of possible experience. At the same time, he also wants to demonstrate that the questions that preoccupy metaphysics are inevitable, and that the arguments of metaphysics, although deceptive, should not be dismissed without sympathetic comprehension (as they are by the traditional skeptic). Kant argues that they tempt us for gen13

Introduction mne reasons, inherent in the nature of human reason itself, and when these grounds are properly understood they can be put to good use for the causes of both human lmowledge and human morality. This argument is the basis for Kant's theory of the regulative use of the ideas of reason in scientific inquiry, which Kant first suggests in the final appendix to the "Transcendental Dialectic" then elaborates in the Critique ofJudgment, and for his theory of the foundation of morality in the practical use of pure reason, which he first describes in the "Doctrine of Method" and elaborates in many subsequent works, but especially in the Ground7uork the Critique of

Practical Reason. The Leibnizian-Wolffian tradition, as presented in Alexander Gottlieb Baumgarten's (first edition, 1738), which Kant used as the textbook for his lectures on metaphysics for virtually his entire career, was divided into four parts: ontology, psychology, cosmology, and theology. The "Transcendental Aesthetic" and "Analytic" are Kant's critical replacement for traditional ontology. The "Transcendental Dialectic," however, is dedicated to arguing that the other three parts of the rationalist system are pseudo-sciences founded on inevitable illusions of human reason attempting to extend itself beyond the limits of sensibility. Kant does not present the three rationalistic pseudo-sciences as mere historical artifacts, but attempts to display them as inevitable products of human reason by associating them with the unconditioned use of the three traditional forms of syllogism: categorical, hypothetical, and disjunctive. Seeking the unconditioned subject to which all our thoughts relate as predicates, we generate the idea of the soul as a simple, non-empirical substance; seeking the unconditioned in respect of any of several hypothetical series arising in the world (of composition or extension, of decomposition or division, of cause and effect) leads to ideas such as that of a first event in time, an outer limit to space, a simple substance and a first cause. Finally, Kant derives the idea of a most real being or God as the ideal ground of the real properties constituting all other things. Kant's overall argument is that although these rationalist doctrines are inevitable illusions they are still pseudo-sciences, and must give way to doctrines remaining within the limits of sensibility: rational psychology gives way to empirical psychology, which Kant expounded in his lectures in the form of "anthropology"; rational cosmology gives way to the metaphysical foundations of natural science, which Kant derives by adding the sole empirical concept of motion to the principles of judgment; and rational theology gives way to what Kant will call moral theology, the doctrine that God and immortality are postulated, along with freedom of the will, solely as conditions of the possibility of human morality. The opening book of the "Transcendental Dia1<~ctic" is therefore a

Introduction derivation and even a limited defense of the as the immor tal soul, free will, and God, with dogma tic metaphysics has always been preoccu pied (A293- 338/B 349-96). Reason, traditionally though t to be the highest of our cognitive has a "logical use" in which it simply draws inferences from pnmclples, also a "real use" in which it seeks to base series ordinar y inferences, such as those from cause to effect, in ultimate, foundational pnlnclp les, such as the idea of an uncaused first cause. The ideas of ultimate principles are genera ted a priori by the faculty of reason it seeks, throug h regressive syllogistic reasoning, for what is unconditioned in respect of the objects given in experience, according to the principll~s underst anding that govern these objects. In particular, it is categories of relation when llsed withou t regard to the limits of sensibil ity that give rise to the chief ideas of metaphysics: concep t stance giving rise to the idea of the soul as the ultimat e "UltJIC:~l, concep t of causation giving rise to the idea of the W(}f!,::!-,,yl1Ole completed series of conditions, and the concep t of glVl11g rise to the idea of God as the commo n ground of all possibilities. Kant suggests that each of the three relational categories gives rise to a distinctive form of syllogistic inference, series of which can terminated by the idea of an uncond itioned ground , but also to acquire knowledge by means of the relational categories withou t sensibility gives rise directly to the idea of an uncond itioned subject, series, and set of all possibilities. The second and by far the larger book of the "Dialectic" expounds "The Dialectical Inferences of Pure Reason" in great detail. errors of rational psychology are diagnosed under the rubric of "The Paralogisms of Pure Reason," those of rational cosmology under the rubric of "The Antino my of Pure Reason," and those of theolog y under the rubric of "The Ideal of Pure Reason." The "Paralo gisms. " Rational psychology is the the "Paralogisms" (or fallacious inferences) of pure reason, which argue l11v'alI,cH v from the formal unity, simplicity, and identity of the thought of the subject of thinkin g or the "I" to the conclusion that the soul is a real and simple (hence indestr uctible ) substance that is self-identical throujgh()ut all experience (A341-66). In the first edition , the "Paralo gisms" included a fourth part, which defends the reality of external appearance in space simply by reducin g objects in space to one form of immed iate represe ntation (A366- 40S). This response to idealism appears to provide only a Pyrrhic victory over it, which provok ed charges of Berkeleianism against Kant, and was therefo re replaced in the second edition with the "Refuta tion of Idealism," which as we saw argues for the real existence of objects in space and time althoug h for the transcendental ideality of their spatial and tempor al form. In tlle second


edition, the entire chapter on the paralogisms was rewritten and simplified (B 46-22); to fill the place of the superseded fourth paralogism, Kant adds an argument that his dualism of appearance and reality undercuts the traditional dualism of mind and body, with its problem about the possibility of interaction between two fundamentally distinct kind of substances, opening the possibility that both mind and body are different appearances some single though unknown kind of substance. The "Antinomies." The longest and most painstaking part of the "Transcendental Dialectic" is the "Antinomy of Pure Reason," which deals with the topics of rational cosmology (A40S-S83/B432-6u); indeed, as we will show below, Kant originally thought that all ofthe errors of metaphysics could be diagnosed in the form of these antinomies. Here Kant argues that reason's natural illusions are not merely revealed by subtle philosophical analysis but unavoidably manifest themselves in the form of actual contradictions each side of which seems naturally plausible. Kant argues that unless we accept the transcendental idealist distinction between appearances and things in themselves, we will be committed to accepting mutually incompatible arguments, arguments both that there must be a first beginning of the world in time and that there cannot be, that there must be to the world in space and that there cannot be (the two halves of the first antinomy), both that there must be a simple substance and that there cannot be (the second antinomy), both that there must be at least one first or uncaused cause and that there cannot be (the third antinomy), and that there must be a being whose necessary existence is the ground of all contingent beings and that there can be no necessary being (the fourth antinomy). The only way of resolving these contradictions, Kant argues, is by accepting that the natural world is a realm of appearances constituted by the application of the categories to sensible intuitions, and not a realm of things in themselves. Regarding the first two antinomies, which he calls "mathematical" antinomies because they havc to do with size and duration, Kant argues that there is no fact of the matter about the size of u1.e world as a whole, because the natural world is never present in experience as a whole, but rather is given to us only through the progressive or regressive synthesis of spatiotemporal intuitions. We can always proceed indefinitely far in the progressive composition of spaces and times into ever larger or longer realms or in the regressive decomposition of space and time into ever smaller regions, but we can never reach a beginning or an end to such series, as would be possible if they were finite, nor complete any synthesis of them as infinite either. Both sides of the mathematical antinomies, therefore, turn out to be false, because both rest on the common - and false - assumption that the world is given independently of our ongoing synthesis In its representation,

Introdu ction

and that it therefo re has a determ inate magnitude, which must finite or infinite. For the and antinomies, "dynamical" because they have to do with the causation of the world and its events, Kant proposes a differe nt solution. Here he argues both sides may be true, if the denial a free cause or necessary is restrict ed to natural and sensible world and is taken to refer to what might exist in a noume nal or supersensible world of things in themselves. Just as his thinkin g the antinomies generally shaped his thinkin g about the structu re and outcom e the entire "Trans cenden tal Dialect ic," so Kant's resolut ion of the third antinom y will go on to an import ant role in his and in his ultimat e accoun t of the relation be1':WE~en th(~or'etlcal moral philosophy. The "Ideal of Pure Reason ." Rational theology, the of the metaphysical pseudo-sciences, is taken up by Kant in chapter of the "Transc endenta l Dialectic" (AS67- 64z1n 595-67 0). If an "idea" is a pure concep t generat ed by reason, then an "ideal" is the concept of an individu al thing as exemplifying an idea pure reaSOn. It would not be natural to think of the idea of the soul, for example, as ing rise to an ideal, because we natural ly think are many souls; it is natural (at least in the Judaeo -Christ ian to of idea of God as the idea of a single and thus idea of is ideal of pure reason. Kant of the idea of God as an ens or

l\ii~~=ifiii,;~~li~:1;~d~~;~1~;~d~;nearltywenty years before the puiblH:atilon

his own earlier attemp t to prove the existence of God as such an ens reexistence of God, which were already criticized in Kant's earliest philo:soJ)hical writing, the New Elucidation of the First Principles ofMetaphysical Cognition (I75S) as well as in The On~y Possible Ground. Kant organizes the traditio nal proofs of the existence of God out attemp ting to explain why there should only be these three) into the ontological proof, based solely on the concept of God, the cosmological proof, based on the sheer fact of the existence of a world, and the physico-theological proof, based on the particu lar constit ution of actual world, especially its alleged exhibit ion of purposive design. The first of these is Kant's represe ntation of the proof favored by SI. Anselm and revived by Descartes; the second is his name for an ment from conting ent existents to their necessary ground favored Wolff and his followers; and the third is what Kant calls the argume nt
alissimum as well as the other traditio nal attemp ts to prove

argume nt here

of the Critique in The Only Possible Ground ofProo ffor a Demonstration of the Existence ofGod (1763)' But now Kant subjects to witheri ng criticism

Introduction so many thinker s of the early Enligh tenmen t, from design favored especially in Britain (where Hume had already subject ed it to trenNatura l Religion, which, bechant criticism in his Dialogues Germa n, Kant had not yet into cause of the delay of their transla tion on). seen at the time he publish ed the Critique First Kant attacks the and ty proper a not is ce tological argume nt, holding that since existen therefo re not itself a perfect ion, it cannot be include d among the contents of the idea of God, and cannot be inferre d from that idea alone. ce of an object is always the presup Hl:>LCdU , Kant argues, the existen aspositio n of the truth of any assertio n about it, and cannot itself be if even that argues then sumed for the proof of such an assertion. Kant the h establis could cosmological and physico -theolo gical proofs , existence of some necessary and purposive being (which they cannot) Deity perfect ely suprem a of they still could not establish the existence ical unless the ontolog ical proof also succee ded. Since the ontolog exthe proving of ise proof is unsoun d, the entire metaphysical enterpr up given be must n istence of God - as an object of theoretical cognitio as hopeless. l Regula tive use of the ideas. The outcom e of the "Transc endenta mislead a is This e. Dialectic," therefo re, seems to be entirely negativ ing conclusion, however. In an appendix to the "Dialectic," Kant begins a limited rehabil itation of the ideas of traditio nal metaphysics by argut ing that the ideas of reason have an import ant functio n in the conduc are they if is, that vely, of natural science if they are unders tood regulati is taken to represe nt not metaphysical beings or entities whose reality inquiry of ns directio supposed to be demonstrable, but rather goals and and that mark out the ways in which our knowledge is to be sought for us tes stimula which organized. This is true of the idea of a simple soul, worldte comple a of to search for a unified psychology; of the idea ic whole, which leads us constantly to expand the domain of our scientif world the ng regardi investigations; and above all of the idea of God, for the as if it were the produc t of a highest intelligence leads us to look for orgathe for ial benefic maxim um in order and connectedness, which is nizatio n of whatev er empirical knowledge we do acquire. This arguof ment, which Kant continu es in the Critique ofJudgment, is the first if but ing mislead be can Kant's constru ctive argume nts that reason consecond Kant's wisely used is far from idle or even unnecessary. sDllctive argume nt about reason, that its ideas have a profou nd practical use for the guidance and regulat ion of conduc t, is begun in the final part of the Critique, the "Doctr ine of Method ." "The Doctri ne of Metho d." The second major division of the readCritique, the "Doctr ine of Method ," tends to be neglected by its arthe and long so is ts" ers, perhaps because the "Doctr ine of Elemen of ine "Doctr the But gumen ts already surveye d are so exhaus ting.


Method," in Kant reflects upon the the critical philosophy comparing it with other compares the method the method of philosophy with the method of theoretical philosophy with the method of and of dogmatic, emthe method of critical philosophy with the pirical, and skeptical philosophy - includes some extremely ImpOl~tallt discllssions. Its first chapter, the "Discipline of Pure '-'.L,a"',n,, Kant's most mature treatment of the difference between phl1o~,Of)hy mathematics, arguing that both provide synthetic that mathematics provides determinate answers to its nrl,hl!pn,c its objects can be constructed in pure whereas phllo:sOflhy vides only general principles because it can construct are ditions of possibility for the experience of objects, not pa]rt1C~Ul~lr (A 7I2-38/B 740-69). Then it provides an ardent of freedom of public communication as well as of open-mindedness in of metaphysical issues, arguing that the very existence depends on the free give-and-take of controversy bel:WE~en ratlona1 ings, requires the liberty to come to one's own co:ncJluslOrlS and to express them to others (A 738-691 766-97). discussion presages Kant's impassioned defense of of th()UF~ht in his political writings of the r 790S. The chapter concludes with a discussion of the contrasting roles of hypotheses in science and phllo:,Oflhy (A 769-8 21 B798-810) and then with a reflection upon his own he calls nr,nntc" philosophical argumentation, (A 782-941 B 810-22). The second chapter of the "Doctrine of " the "Canon of Pure Reason," contrasts the epistemological status ofthf>orptll'OlI tion with that of the principles and presuppositions of pnlctJlcal or morality, and in so doing provides Kant's most sy~;tel'natrc of moral philosophy prior to the Groundwork oftbe iVli,tai~bvszcs (1785) and Kant's first systematic statement of his for rational faith in God on moral grOlmds (A 79s-83 1/B82 3-59), an argument that Kant was to restate and refine in the subsequent two and to continue to work on until the end of his life. The third the "Architectonic of Pure Reason," continues the discussion of the contrast between and other forms of cognition, such as his:tOljC;l! knowledge, as as of the contrast between theoretical and practical reason (A832-sr/B860-79), the final I'hon'-pr of the "Doctrine " and of the whole the Pure Reason," orients the critical in relation to skepticism, and competing positions of dogmatism, had the (A852-S61 ferentism, the discussion of B 880-84). For all its brevity, this section has influence on subsequent conceptions of the history of phllo'sOplTY.


Reason is comple x and many-sided. Both its overall The Critique message and its meanin g for the subseq uent history of philoso phy defy as any easy summary. The Critique has perhap s most often been seen the g -avoidin while virtues, the es combin that way markin g out a third the pitfalls, of both the "ration alism" of Descar tes and Leibni z and , Critique the reading of way This Hume. and "empiricism" of Locke de, himself Kant ed suggest extent some to however, even though and pends on a simplified reading of the history of modern philoso phy weakand hs strengt the of ent assessm lete at the very least on an incomp nesses of Kant's modern predecessors. Less controversial is the obsern vation that the Critique'S main intentio n is to find a middle way betwee view theistic a bolster to ts attemp its lly especia traditio nal metaphysics, that of the world with a priori rationa l argume nts, and a skeptic ism those with along science natural modern of would underc ut the claims of religious metaphysics. critiVVe see this clearly in the way that Kant defines the positio n of and ism, skeptic ism, empiric ism, dogmat to t philoso phy in contras siga phy philoso ical theoret for out carve to indifferentism. He seeks dge knowle al empiric of that from t distinc , nificant but limited domain rated exagge the ing exclud but sense, n and the opinion s of commo the claims that have brough t metaphysics into disrepute. In this way, philoso modern in n traditio main a to s Critique of Pure Reason belong philopriori a an provide to tries that tes, Descar phy, beginn ing sophical founda tion for the met.1}ods and broad features of a modern an examination of the suitability of human scientific view of nature of knowledge of nature that modern scicognitive faculties for the time, Kant tries to save precisely what same the ence aims to achieve. At connec ting the claims of relicannot, s the dogmat ic metaphysician but to the sphere of mon1 theory of sphere the gious metaphysics not to -edition preface, by second the of words famous practice, and, in the (BXXX). But Kant faith for room make to order limitin g knowledge in an authentically in last, the lly especia goals, tries to accomplish all these l capacity rationa our to place first giving always , Enlight enment manner them, correct to ments, achieve and abilities ve to reflect on our cogniti human that so itation, self-lim to reason of ions and to subject the pretens reason itself retains ultimat e authori ty over all matters of human knowllies edge, belief, and action. The ultimat e autono my of human though t outside ty authori any to answer must nor can in the fact that it neither itself. focusing on the The originality of the Critique can be indicat ed intractable most the of two resolve to neously way it attemp ts simulta


Introduction problems of early modern philosophy, the smmltane01JS VIllldlC21t/Oln the principle of universal causality the treed;:)m will. The great idea of the Critique ofPure Reason is that the that explains the possibility of our Imowledge of the turldanH~ntal principles grounding a scientific view of nature is also the possibility of our freedom in both intention threatened the rule of causality in that natural world. Kant that the principles of the scientific can be lmown are tainty because they express the structure of our own therefore conditions of the possibility of our experience, which we impose upon raw data of sensation. Thus, there is a sense in cerbecause titude about the principles of science is possible autonomy: we are not merely passive perceivers of sensible information also cognitive agents flowing into us from external objects, structure what we perceive in accordance with the necessary COJrldJlt/cms of our active thought. Thus Kant argues that we can be certain of fundamental principles of science - above the law causation, the assumption underlying all scientific that has a cause and can therefore be explained in accordance with 2 nature - precisely because this law is a condition of the possIbIlIty thought that we must impose upon our perceptions in experience at all. In the seventeenth and eighteenth centuries, the of causation had been put into ever more successful use practicing scientists, but at the same time doubt had been cast upon it philosophers. First the principle had been supported upon theological foundations Descartes and his follower Nicolas Malebranche, then to a mere phenomenon, as by Leibniz, or finally exposed Hume as simply the result of mere custom. Kant, however,argues a geJl1UJl11e necessary connection between events is required for objective succession in time, and that the concept of causality in this connection is expressed is imposed on experience by our own as an indispensable condition of its possibility. The therefore, is the true lawgiver of nature, and the successes of modern science are due to its conduct of its inquiries in accordance with a whose ground lies a priori in the structure of xviii). At the same time, nature is to be regarded as essentially an of human sensation and thinking, and the validity of the causal pIe is to be restricted to the world as it appears under the conditions of our experience of it. In this way, the same account guarantees the certitude of the principle of causation also guarantees the of the human will, which is precisely what was typically thought to exeluded by the universality of causation. According to Kant, if we understand the principle of causality and


s e e le

Introduction scientific wc)ri<ivlew as prc)ducts of our the posslbl hty is human when action of a radical self-de termina tion of human as such works, later In itself. in is it as conside red not as it appears but within the Practical Reason (1788) and the the with the this tes comple Reason (1793), Kant Boundaries ion obligat our of ess awaren ble the inexora further argume nt own our eously spontan given which is to live up to the in the breach), can (even if reason and which we all ry conditi on of the necessa the is prove the reality of our es. Yet this furourselv upon make we possibility of the that we cannot nt argume ther argume nt presupp oses the first t at the same withou lves themse ground the princip les of natural science nces. appeara mere to time revealing that their scope is limited one stroke two of the most pressKant's bold attemp t to resolve has seldom been accepted by his ing problem s of modern qualification. Some feel that Kant's identifi cation of successors of the basic princip les of science with the fundam ental princip les specifthe in nce confide human unders tanding itself betrays too much Nev.'1:onian mechanistic physics that prevailed at his time, leaving thetoo little room for subseq uent scientific developments, such as the that felt have Others ics. ory of general relativity and quantu m mechan t is though human of laws Kant's reducti on of the laws of science to the Few . science of validity e objectiv not an adequa te accoun t of the the possibility of freedom could have felt comfor table with the idea of placing the real arena human decision making behind be defende d freea veil of ignoran ce, and many have felt that the idea that human adh throug only realized be value but that it can dom is our time, same the at Yet one. ical herenc e to law is a strange and paradox and broad elemen ts of Kant's philoso phy have become indispe nsable of frame modern the of tions therefo re often almost invisible assump a mcrely is mind human the mind. No modern thinker can believe that h althoug But "'9 nature. of r passive recorde r of external fact, or a "mirro many hold that since we have no way of steppin g outside the human our point of view, it may not be as easy as Kant though t to separat e out modern every yet nature, of subjective contrib utions to the constit ution philoso phy holds in some form or other the Kantia n thesis that human h beings make an active contrib ution to their knowledge. And althoug betion distinc n Kantia rigid a few defend human freedom throug h have tween phenom enal appeara nce and noume nal reality, even fewer es preclud science in inism determ though t that the assump tion of causal what to ng accordi ns decisio make conceiving of ourselves as agents who have seem to us to be the most rationa l princip les of value. Thus many difental fundam a is there that accepted in some form the Kantia n idea 22

Introduction ference between the standpoints of the actor this difference is crucial to Even those who reject Kant's solutions to the of gT(YllIl(11ng natural science and making sense of our moral agency must solve these find in Kant's problems and find a way to avoid what solution to In this way, all modern thinkers are c111Jdl'en whether they are or bitter about their palternity.


Critique of Pure Reason has often been re]:lre~,ented of a violent revolution in Kant's thought 1772 - a midlife crisis in which the ro]~tv-elQ'Jllt-1ve,lr-,ow UHU"J~L jected his previous adherence to the LelbrnZlan- \;\/olltlan phI1 osclphy, the systematic philosophy Christian ated out of the brilliant fragments that were all the philosophy of Gottfried Wilhe1m Leibniz and become the dominant philosophy in enlightened German universities after the 1720S. Kant himself gave rise to this legend own remarks, above all his comment in the Prolegomena to Future Metaphysics - the short work that Kant lished in 1783 to try to overcome the indifferent or hostile reception of the Critique - that "it was the recollection of David Hume that many years ago first interrupted my dogmatic an specentirely different direction to my investigations in the ulative philosophy."2I There were certainly changes in Kant's thought both before and after the publication of his m,mg:ur;ll dlSS(~rt;l tion, De mundi sensibilis atque intelligibilis forma~t flr'lnrlfJ1'" Form and Principles of the Sensible and Intelligible last publication preceding the years of intense leading up to the publication of the Critique in I 78 I. Kant has misled those who have supposed that all his work in the years preceding this point was slumbering in Wolffian and that he awoke from this slumber only through some sudden re(:olLectlon the skepticism of David Hume (17II-I776). In fact, Kant had been chipping away at tull(lalm(~ntal Leibnizian-Wolffian synthesis at least since the pu.bll.catlOn exclusively philosophical work, his M.A. thesis fln'ncztJiorU1n fJ:Vil1wr:1171l cognitionis metaphysicae nova dilucidatio New of Principles of Metaphysical Cognition) in 1755. There were rprNll,lv major developments in the content of Kant's views in the period around 1769-70 leading to the publication the dissertation, and then further developments in Kant's doctrines and his


Introduction method in the period beginn ing in 1772 concep tion of these were the Critique. and culmin ating in the publica tion in the hisand t though own Kant's in both ments revolutionary develop as well Reason, ofPure Critique the so, Even phy. philoso tern tory of\Ves have to be seen as as the further "critical" works that were to follow in the produc t of a continu ous evolution at least since 1755, a process in which Kant never fully subscribed to the Wolffian orthodo xy and and which he continu ed revisin g his positio n both substan tively . Critique the at arrived he until lly methodologica was first published, Kant's though t Moreov er, even after the ces continu ed to evolve: as we will see below, there are major differen the work (both present ed in betwee n the first and second editions their entirety in the present translation). Indeed , even after the cation of the second edition, Kant continu ed to revise and refine both his arguments, in publish ed work such as the Critique of his views he was still workin g at the Judgment and in the manuscripts on postumum).Z2 Furthe r, it Opus the as ed publish end of his life (later philosophy, as first mature Kant's that t t.1}ough be should by no means outrigh t rejectio n an nts represe Reason, ofPure expressed in the Critique original philoso the of all above ssors, predece his of of the philoso phy be though t of can phy philoso Kant's y, contrar the phy of Leibniz. On blished harpreesta the of vision 's Leibniz ize synthes as an attemp t to 3 with the gracel of les princip the and nature of les mony of the princip of substance of N ewtoni an science and the moral and political insights a ]ean-Ja cques Rousse au (1712- 1778). To the extent that Kant was not critic of the Leibniz ian-Wolffian philosophy, his criticisms came an only from Hume but even more from Wolff's Pietist critic Christi to August Crusius (IFS-I 77S). These critical forerun ners led Kant the transfo rm Leibniz's vision of a harmon ious world of monads under rule of God and Rousseau's vision of a social contrac t expressing a genaseral will into ideals of human reason, neither of which can simply be the serted to exist in well-fo unded cognitive judgme nts made within can limits of human sensibility and underst anding , but both of which ble attaina and must represe nt the ultimat e even if never comple tely t. goals of human theoret ical and practical though t and conduc We cannot offer here a full accoun t of Kant's intellectual develop pubKant comme nt briefly on a numbe r of the works ment. But we in order to point out some of the ideas that were 1770, h throug lished to incorpo rated into the Critique ofPure Reason as well as some that had will We shape. be rejected or overcome before the Critique could take then comme nt equally briefly on some of the evidence for the develop 1770 n betwee " ment of Kant's though t in the so-called "silent decade to and 1781. This discussion of the genesis of the Critique is provided


Introduction interpret the intentions of the work as well as to cast some on complexities of its organization and argumentation. Nova diluddatio (1755)' In his first treatise on metaphysics, Kant already took issue with some of the most fundamental tenets of the Leibnizian-Wolffian philosophy, while expressing his allegiance to other aspects of it. Several of the most criticisms that Kant made in this first philosophical work reappear in the Critique. The most important critical points made in the Nova dilucidahave tio are four. First, Kant rejects the assumption, to which abbeen more clearly committed than Leibniz, that there is a solutely first, universal principle of all "24 Kant argues is a logical point, that affirmative truths rest on the principle "whatever is, is" and that negative truths rest on the principle "whatever is not, is not."25 That is, he argues that the assumption that the negation of a true proposition is false is itself a substantive presupposition of a logical system and not something provable any logical system itself. This is not yet the argument that there are some truths that can be deltl10Ins'trated from adequate definitions by logic alone and others that gomg beyond logic, which will become the distinction between and synthetic judgments. But it shows that from the outset of his career Kant was dubious of the supposition that all philosophical could in ciple be derived from a single principle that lay beneath Leibniz's theory that all true propositions can be proved by the analysis of concepts. Second, Kant rejected the proof of the principle of sufficient reason offered by both Wolff and his disciple Baumgarten. According to their proof was that if it were assumed that something not have a sufficient ground, then its sufficient ground would be nothing, which aswould then mean that nothing was something;26 this is both suming precisely what is in question (that everything does a ground), and also a mere play on words. Kant's argument is that in every true proposition the subject must be determinate respect to any predicate that might be asserted of an object, so must always be something that determines whether a given predicate is true of it. 27 This is not adequate either, since it fails to see that noUung more than the properties of an object are necessary to determine what predicates should be asserted of it. But it already reveals Kant's characteristic tendency to convert ontological questions into epistemological questions - that is, the transformation of questions about sorts of things there must be into questions about the conditions under which it is possible for us to make claims to knowledge about things. The velopment of this tendency into a full-blown philosophical meUl(ld be the key to the Critique of Pure Reason, in which, as Kant is to say, "The proud name of ontology, which presumes to offer a pri-

~ 8ogazi~i Universitesi KutGphanesi


Introdu ction

things in general in a systema tic doctrin e ... must give of pure unders tanding " way to the modes t one of a mere ( A2 47/ B 303) the argume nt which he was later famously to Third, Kmt existence of God. This was the "ontolo gical" argume nt for Leibniz , Descar tes and refined proof of St. Anse1m, be inferre d from predica tes necessarily inthe existence the cluded in the concep t of God. Kant's rejectio n of it was based on it that is, that is "ideal" rather than "real": suppos ition that its canbut God in the concept of we may have unpack s concep t. 28 At this ing to answer any is that h not establis the real existence of stage, Kant offered an alterna tive argume nt to reGod must be accepte d as the ground of all possibility. He was later 9 y hostilit his of Pure Reason,2 but ject this argume nt too in the remain to his analysis of its defect were to the ontolog ical argume nt m of the ontolog ical argume nt was criticis His ged. unchan essentially distincReason's Critique anothe r precurs or of , Kant Critique the In nts. judgme ic synthet and n tion betwee , science l substan tive trut-hs in mathem atics, physica will argue that indele knowab necessa rily true and philoso phy itself, call "a he (what nce experie lar particu any to appeal penden tly of of priori"), go beyond what can be derived from the mere analysis conof cepts, and therefo re require the discovery of a whole new method ssors his predece though t beyond the method of analysis employ ed arten. Baumg and , Leibniz the Finally, in the Nova dilucidatio Kant rejects the basic princip le of arten. following him, Baumg Leibniz monad ology mainta ined a substan ce is true in virtue of true ing everyth that le princip the This is apof the inheren t nature of that substan ce itself, so that what would of ons reflecti pear to be real interac tions betwee n substan ces are only suball as the creator of God has chosen to the harmon ious ones precise ly because e possibl all of best the is that a in stances Sucit is harmon ious. Kant mainta ins what he calls the "Princi ple of as insofar cession," that "No change can happen to substan ces except ency depend they are connec ted with other substances; their recipro cal 0 used on each other determ ines their recipro cal change s of state. "3 Kant his which this princip le to argue for the system of "physic al influx," monthe teacher Martin Knutze n (1713-1 751) had employ ed against physadology. The argume nt for a system of real interac tion among all ples "princi the of ical objects in space and time was to be a crucial part . Critique the in of empiric al though t" for which Kant would argue special a ion" Further , Kant also derived from this "princi ple of success ed argume nt that all change s among perceptions would have to be explain of ce existen and thus a proof of the "real as due to change s in

ori cogmn on



bodies."31 Changed from an ontological to an epistemological key, this argument would become the basis of the "Refutation in second edition of the Critique ofPure Reason. So Kant's first piece of philosophy already contained some of Kant's most characteristic criticisms of his predecessors as well as some of the substantive conclusions of his mature work. What was was his own still a new philosophical method that could get shaky arguments for these conclusions to a totally new t011nclatJlOn them. That would take at least two more decades to discover. Before leaving the Nova dilucidatio, however, we should also mention several points at which Kant still agreed with his predecessors, above Leibniz, and that would subsequently come in serious criticism. The first point concerns Kant's early treatment of the freedom of the will, to which he devoted an extensive dialogue Nova dilucidatioY At this stage, Kant recognized only the two tradll:I01:Ial ternatives of determinism, according to which any event, a human action, is entirely determined an sequence of events, in the case of a human action may go all way to to earlier involuntary events in the agent's life or even to events that life, and indeterminism, according to which a free choice is in no way determined by any prior history. The latter position, which Kant called the "indifference of equilibrium," was represented by Crusius,33 and firmly rejected on the ground that this position undermine any reasonable conception of responsibility. he opted for Leibniz's position, which was a form of determinism now usuhuman aC1tlOns, ally known as "compatibilism": all events, admit of causal explanation, but some human actions are due to an inner rather than an outer cause or principle, and among those some are to the representation of the chosen action as what be best for the agent to do. Actions caused in this way, even though be necessary and predictable, are entitled to be called spontaneous, untary, or free)4 the time of the Critique Kant was to reject this Leibnizian conception of freedom as the of a turnspit,"35 and it was to be a fundamental task of the Critique Reason, not yet foreseen in 1755, to make way for a third altenlatlve tween traditional determinism indeterminism. Kant was to by means of his "transcendental idealism," his CW,tlJ1cl:10n bet\11een necessary appearance of things to human and how those things, including human agents themselves, might be in themselves: this would allow him to reconcile the Leibnizian and Crusian POSltlOIlS maintaining the Leibnizian position as the truth about appearances or "phenomena" while holding that the Crusian true about things in themselves or "noumena." The second concerns another retention of Leibnizian

Introduction of Coexistence," or the thesis the Kant This is not, in virtue of their existence alone, stand in a "Finite substances any interrelationship with each other, nor are they linked togethe r existence, their of le princip n commo action at all, except insofar as the harmon y of state a in them ns maintai , namely the divine underst anding 6 Even though the rejectio n of this prinin their reciprocal relations."3 recciple follows from his "Princi ple of Succession," Kant did not yet ian Leibniz of part this in mainta ognize this, and would continu e to work that though even tion, disserta metaphysics throug h the inaugural would reject fundam ental aspects of Leibniz's theory of space and time time. It would not and introdu ce Kant's own mature theory ofspace recognize that Kant Reason itself that be until the Critique ry condinecessa a is objects l thoroug hgoing interac tion among physica that the and nce, experie l empora tion of the unity of our own spatiot unity of the than ground other admits no unity of the physical major the of one be our experience; coming to this recogn ition . Critique the accomplishments of the 1770S leading up to The philoso phical works of 1762-6 4. Around the time of the Nova science that would dltuadaitlO, Kant publish ed two other works in natural These are the phy. philoso help to provide a founda tion for his later 55) and the (17 s al Natura l History and Theory of the Heaven

Univers cuius Metaphysicae cum geometria junctae usus in philosophia naturalis, Natuin yment Emplo Specimem I continet Monadologiam Physicam (The

ral Philoso phy of Metaphysics combin ed with Geome try, of which Sample I contain s the Physical Monad ology) (1756). Howev er, the next period of major philosophical publication for Kant was the years 1762 all to 1764, during which time Kant published four philosophical works Reason. Pure of Critique of which are import ant steppin g stones to the Three of these works appear to have been comple ted in the fall of 1762, possibly in this order: the False Subtlety of the Four Syllogistic Figures, published in 1762; The Only Possible Basis fir a Demonstration ofthe Exisness tence ofGod, publish ed in 1763; and the Inquiry concerning the Distinct winner -prize second the y, ofthe Principles ofNatu ral Theology and Moralit an in a compet ition held by the Berlin Academy of Sciences, in which first won 785) (1729-1 Moses Mende lssohn "Essay on Eviden ce" des prize. Finally, the Attemp t to Introduce the Concept ofNegative Magnitu 1763. of r summe the into Philosophy was comple ted and published by The essay on False Subtlety, which is primari ly concern ed to effect in a simplif ication of the many classes of syllogi sm recogn ized nce emerge the to least the ute Aristotelian logic, would seem to contrib of the Critique ofPure Reason. But in its "Concl uding Reflection" Kant as touches on one theme that will be crucial for both the formulation with dealt s problem phical well as the solutio n of virtually all the philoso in the Critique. This is the claim that the fundam ental notion in formal

Introduction logic and in the analysis of the powers of nition is the notion of judgment. Concepts, he argues, icates to one another, can become distinct only by means and inferences, which might have been thought to call upon adchtiOJ131 powers of mind beyond the power of judgment, are in fact or iterated judgments.37 Thus Kant concludes son, that is to say, the faculty of cognizing distinctly and the syllogistic reasoning, are not different fundamental faculties. Both consist in the capacity to judge ... "3 8 The recognition that judgment is the fundamental rorm of all cognitive acts be crucial to the Critique in three ways: Kant late the problem of the very possibility or philosophy as the possibility of synthetic a priori judgment, or the prc)bI,eIll of how judgments can go beyond what can be derived from the mere analysis or concepts yet also claim universal and necessary He argue that the necessary conditions for the application of categories derived from the logical forms of judgment to the spatiotemporal form human experience are the source of all those synthetic a judgments that theoretical (as contrasted to practical or moral) philosophy can actually prove. And he will argue, in the "TranscendentallJlalectI':;" of the first Critique, that the fundamental illusion metaphysics is to think that human reason gives direct theoretical insight into the constitution of things as they are in themselves instead of simply concatenating simpler judgments of the understanding into the more complex judgments we call syllogisms or inferences. Kant's insistotence on the primacy of judgment in human thought is a first ward all these critical theses. In a longer work, indeed a small book, The Only Possible Basis fir a Demonstration of the Existence of God, .!{ant's thought advanced the Critique from a different direction. The argument of the book divides into two main parts. In the first section, as the title suggests, Kant discusses proofs of the existence of God. On the one he refines to it critihis original criticism of the ontological argument, and cisms of two other traditional arguments, the argument from the contingency of the world to the necessity of its cause, which been popularized by Leibniz and which Kant was to dub the "cosmological" argument, and the argument from the order of tlle world to an intelligent author of it, or the argument from design, which was widely ular among eighteenth-century thinkers and which Kant was to call "physico-theological" argument.3 9 On the other hand, Kant refines and extends his own argument that the existence of God can be aejtllC)l1strated as an actual and necessary condition of the existence of any possibility, an argument that appeals to the premise that it be impossible to deny that anything is possible. 40 From the concept of God

Introdu ction

proceed s to derive traKant necessary ground of as d1tl011al predica tes of God such as uniqueness, simplicity, imml1tability, and indeed even the claim that the necessary being is a mindY The introdu ction of God as the ground of all possibility must have the ontological argun:Ient and thesounde r seemed to Kant than the Leibnizian conception, on which the ologically more the anthe universe is constrained creatio n power of God in of Critique the teceden t existence of determ inate possible worlds. But in the as well as nt to reject this argume Pure Reason Rant was tes three traditio nal ones, and to argue that both the existence and predica bel practica as be demons trated on moral grounds, of God could ical dogmas (A 8 I 0- 16 lE 838-44 ; A 8 2 8-9 / theoret than liefs rather of Rant's argument, that a B 856-7). Neverth eless, the underly ing idea demons trating that genuine or "real possibility" is not established just must have some sort of affirmaa concep t is free from contrad iction tive ground in actual existence, was remark ably deep-se ated in Kant's of though t, and would manife st itself again not just in the structu re at crucial points in his practical philosRant's theoret ical ophy as well. Possible Basis shows Kant's early the The second main section and concern to find a proper characterization of scientific laws of nature, which reveals that Kant's comple x view of teleology, or final causes, on seems to be a late accretion to the Critique of Pure Reason, touched 704/ only in the append ix to the "Trans cenden tal Dialect ic" (A642in the Critique of.Yudgment, was acB670-732) and fully developed the tually a longsta nding part of his though t. Against the backgro und of argues debate betwee n occasionalism and preestablished harmony, Rant gthat God's purpose s for the world would be expressed through unchan any h throug ing natural laws valid throug hout its entire history, and not miracu lous episodic interve ntions: "VVhere nature operate s in accorthe dance with necessary laws, there will be no need for God to correct of ty necessi direct intervention; for, in virtue of the course of events which that the effects that occur in accordance with the order of nature, the is displeasing to God cannot occur."4 Thus Rant argues "That in is it if even proced ure of purified philosophy there prevails a rule which, in that ... not formally stated, is nonethe less always observed in practice n investigating the causes of certain effects one must pay careful attentio of nature as far as possible."43 to maintai ning the Here Kant defined an ideal of human knowle dge that was to be ceneven tral to the Critique ofPure Reason and all of his subseq uent works, more ever as its theolog ical founda tion in a concep tion of God became beattenua ted. To have knowledge of the events of an objective world causal under yond one's own consciousness is to subsum e those events laws laws, and to have knowledge of causal laws is to conceive of those


Introdu ction

as themselves of a system if not actllalJlv {'lrp51tPfj God, can nevertheless be conceived us as if ated an intellig ence like but more than ours. 44 Kant not yet see how much effort his task in Critique of Pure Reason and subseq uent be pn~ciseJly show that knowledge of the "unity of nature" or of constan t ture is the necessary conditi on of the unity of our own eXj)erlenlce, to explain how knowledge of such laws of nature is po:ssil)1e. Kant's though t about the problem of causal laws would be advanced further in the last of the four key works of 1762-6 3, the essay on Negative Magnitudes. But before we turn to that, we different steps in the directio n of the Critique that Kant took in the third of these works, the concerning the Distinctness the Principles of Natura l Theology and Morality. Kant wrote in the late fall of 1762 and submit ted it to the of Sciences in Berlin by I Januar y 1763, deadline for on the questio n of whethe r metaphysics, conceived to l11clucle uaUl"a' theology and ethics, had the same prospects ics and could use the same method . The still domina ted Wolffians, preferr ed Moses Mendelssohn's elegant restate ment of fundamental tenets of 'Volffianism for the first prize, recognized the merits of Kant's essay with an honora ble mentio n and publicaltion along with Mendelssohn's essay (which did not take place until In the rationa list tradition, Mende lssohn argued for the Sl11111~lflty the method s of mathem atics and philoso phy with a the suggestion that the certitud e of metaphysics is even greater that of mathematics. In an accoun t of the epistemology of mathematics that would still be acceptable to many philosophers, he argued that the proof of mathem atical theorem s from their premises on the application of logical principles to mathematical concepts, the truth of mathem atical propos itions is an matter, depend ing upon the inconte stable but still observational fact basic concepts of our mathem atics fit our experience. then that metaphysical argume ntation proceeds for the most part the same lines as mathem atical proof, with the one difference key cases the connec tion of the formal system of proof to not have to be made empirically but is also secured on tual grounds. These two cases are the metaphysics Kant would later label "rational psychology") where the Cartesian cogito proves the existence of the soul in a non-em pirical way, and the metaphysics of God (or "ration al theology"), where accepted the ontological argume nt as proving existence of from the mere concep t of God. Since in these two paradig matic losophy existence claims could be proved withou t recourse even to

Introduction to have the judged most secure observation, 45 potenti al for even greater certain ty than mathematics. Althou gh he wrote withou t prior knowledge of Mende lssohn' s essay, Kant was of course familiar with the Wolffian backgr ound on which in criticizing the method ologica l asMende lssohn was drawing, sumpti ons of Wolffianism more firmly than he had ever done before, Kant wrote an essay diametrically oppose d to that of his competitor. This essay takes major steps toward the positio n of the Critique ofPure deReason, althoug h crucial differences still remain. Kant's most radical the toward step t parture from prevail ing orthod oxy and his bigges of Instead ty. certain Critiqu e comes in his accoun t of mathem atical analyzof process the two-fro nt holding that mathem atics proceed s ing concepts on the one hand and confinn ing the results of those analyses by compar ison with our experience on the other hand, Kant argues that in mathem atics definitions of concepts, no matter how similar they may seem to those current in ordinar y use, are artificially constru cted by a process which he for the first time calls "synthesis," and that mathematical thinkin g gives itself objects "in concreto" for these definitions, or conslructs objects for its own concep ts from their definitions. Thus, diswhatev er exactly the concep t of a cone might signify in ordinar y arthe of t course, in mathem atics the concep t a cone "is the produc on rotated bitrary represe ntation of a right-a ngled triangle which is 6 Thus, we can have certain knowle dge of the definione of its sides."4 tion because we ourselves constru ct it; and we can have certain knowltrue edge that the definition correct ly applies to its objects because the r howeve cted, objects of mathem atics are nothin g but objects constru have es ourselv that may be, in accordance with the definitions that we constru cted. In philosophy, however, things are quite different. Philoso phy does not begin from self-constructed and well-defined definitions, but from are also given in a confused manconcepts, which are already given ner. Compl ete definitions of philosophical concepts come, if they come at all, at the end of philosophical inquiry. In fact, Kant insists, the goal the of definin g concep ts - so central to the academ ic philoso phy of the es compar time - is not the goal of philoso phy at all. Instead, Kant "inmethod the proper method for philoso phy to what he takes to be not troduc ed by Newto n into natural science": obtaini ng certain ty that marks about comple te definitions but about "those characteristic and are certainly to be found in the concep t of any general propert y" tely comple and can lead to "judgm ents about the object that are true certain." The certain ty of such judgme nts has to be ground ed in someithing other than definitions, in the case of metaphysics in "an immed e evidenc of ate and self-evident inner consciousness."47 Such sources while so tions, then have to be careful ly analyze d for their implica

Introduction means of synthesis. "geometers acquire their concepts phers can acquire their concepts only by means of IJ.'VU'lI.'"'''' completely changes the method of thought."4 8 from the definitions introduced into mathematics determinate objects can be constructed, this is not the case in philosophy, where the objects knowledge are not our own constructs, where our concepts give us deonly abstract and indeterminate knowledge of objects terminate and concrete objects themselves. Thus "in the object is considered under sensible signs in concreto, whereas in ophy the object is only ever considered in unlVETS:l] alJstracted cepts."49 So mathematical knowledge is certain because it is gnmflded on definitions of our own construction and because concrete objects can be constructed from those definitions, whereas on philosophical knowledge is less certain because it is analysis of given concepts and less determinate because it yields general judgments about objects. Kant illustrates the differences between m~lthenlaj:lc;ll sophical method with three examples. First, following Crusius, argues that metaphysics depends not only on two distinct formal or logical principles (as Kant had already argued in 17SS), also on many "first material principles of human reason" are "lDldemc)l1strable," such as "a body is compound."5 0 Second, he reiterates his argument of the Only Possible Basis that from the argument for the existence of God as the ground of all possibility other predicates of God can be derived - this is supposed to show how from a certain though incomplete consciousness of some of a thing's characteristics other certain judgments can be derived - but also adds that in further judgments, about God's justice and goodness, only an "approximation to rptt"innr" is possibleY Finally, about morality Kant argues we may easily be able to identifY some formal principles of obligation, such as "1 ought to advance the total greatest perfection," such are useless without material principles of obligation, which us the extension of an abstract concept like perfection actually is - what courses of action actually contribute to perfection - and such material pIes are themselves indemonstrableY Kant is here clearly working his way toward several of the central ideas of the Critique ofPure Reason. Although he does not yet analytic or synthetic judgments, his distinction between ",","'yu,,synthetic methods is leading in that direction: whereas this contrast between methods was merely a contrast between direction in causal or syllogistic inference,53 for Kant the difference has become one between constructing concepts or their definitions (the syrlth,etlc method) and unpacking concepts to get to definitions method). This will lead to the distinction between judgments that con33

Introdu ction

is given (synthetic judgments) amplifying struct fuller concepts showing what predits concep given te and those that merely explica (see A6-7/B IO-II). nts) judgme cates they already contain depend and ysics metaph that nt Further , Kant's argume logical or just not and les, princip l upon indemo nstrabl e materia of his tenet ental fundam the for way the ng principles, is clearly prepari itions propos basic the that al practic mature theoret ical judgments. But Kant's concep tion of of both are synthet ic yet a has not yet caught up to this recogthe in philosophical method we know these "indem onstrab le" how to nition: he is at a loss considered to be anaof principles when the mathem atics. Before of like lytic, rather than synthet ic to discover a philohave would he , written be Kant's mature work could ic judgme nts. synthet or ial" "mater yield could sophical method that would finally that 1770S the of work phical be the philoso This Reason. Pure of pave the way for the n Once Kant takes this further step, however, the contras t betwee rebe to have will the in d mathem atics and philoso phy provide no vised. The differe nce betwee n mathem atics and philoso phy will latthe and method ic synthet the uses longer simply be that the former atter the analytical method . On Kant's mature account, both mathem mean not does This . method ic synthet a ics and philoso phy must use but that the accoun t of the Inquiry will be comple tely surrend ered, mathof ctions constru e concret the n betwee ce rather that the differen will have to be recast as a ematics and the abstrac t results of use of the synthe tic The : method tic differe nce within the synthe results about decertain yet ic synthet yield method in mathem atics will in philoso phy method ic synthet the of use the terminate objects, whereas of objects, or ce experien the for es principl will yield synthet ic yet certain tandunders the of ts concep pure the of ata" what Kant will call "schem a with ts] concep [these ng providi for ons ing, "the true and sole conditi key s contain already the Thus 185). relation to objects" (AI46/B see aspects of Kant's mature theory of mathematics, but does not yet s. method ic synthet use must phy philoso and that both mathem atics the n betwee ion distinct s InquilY' the then r, Once Kant sees this, howeve philoso phy concret e results of mathem atics and the abstrac t results determ iof ction constru the n betwee ce can be retaine d as the differen principhical philoso of ction constru the and nate mathem atical objects 54 . general in objects of nce experie the of ples for the possibility The last of the essays of 1762-6 3, the Attempt to Introduce the Concept of Negative Magnitudes into Philosophy, focuses on a substantive rather than a method ologica l issue. Kant considers a variety of relationships icthat must be constru ed as real opposit ion rather than logical contrad ns, directio te opposi in motion rs, numbe e tion: positiv e and negativ

Introduction pleasure and pain. Asserting a proposition and its cOlltradlctclry in a contradiction, which asserts nothing at all. Lc)mbuung tions in opposite directions does not result in a logical but in a state of rest that is a real state of affairs. So all sorts of sciences need room for the concept of positive and negative not logical notion of contradiction. Kant's undel~lYJng th()U~~ht hinted at in the last part of the Inquiry, is of identity and contradiction are not sufficient for knovvledg"e of the objective world, and that philosophy must find room for material principles. He concludes noting that the relation cause and effect, although it is not a relation of opposition, is than a logical relation, and cannot be justified mere analV'SIS concepts showing that the consequence is in the hilJUlLIU. This raises the fundamental question, "How am I to ull.derSl:aUld fact that, because something is, something else is?"55 of understanding real opposition, real causation, and more geJt1el-aHy real relations becomes the fundamental substantive nrl,hlptTl ical philosophy. Kant rejects Crusius's attempt to and makes no mention of Hume's formulation of an errLplrlc:al to this problem, which was already available to translation of the first Enquiry (1755)' But he COllcIilld(~s prophetic words:
something is, something else is

Let us see whether we can offer a distinct explanation of how it is and whether we can say anjJl:hing than I have already said on the matter, namely that it does not in virtue of the law of contradiction. I have reflected upon the nature our cognition with respect to our judgment concerning grounds and consequences, and one day I shall present a detailed account of the fruits of my reflections.'?

This day was not to come until the publication of Reason in May 1781; Kant had identified a problem to yet possess a solution. But he clearly was not waiting for a rec:olllecnon of Hume to awake him from dogmatic slumbers. Kant published three more significant works during the [760s: the Observations on the Beautiful and Sublime in 1764; Dreams ofa Spirit-Seer in 1766, a devastating critique of the pretensions Swedenborgian spiritualism as an extreme example of metaphysics that also contained some interesting anticipations of his later moral theory; and a short essay, On the Differentiation of Directions in Space, in 1768, the existence of incongruent counterparts (for example, right- and lefthanded gloves or screws) to argue for a Newtonian conception of absolute space against a Leibnizian conception of space as a representation of a system of relations among objects that could in prmc:lpJle be captured by purely conceptual rel::Jtions, which would SUT)pC)se,dly

Introdu ction

betwee n otherw ise identic al objects leave out differences of difsuch as gloves or screws. Once again, Rant was worryin g about the did he essay brief this in but s, relation ference betwee n logical and real ablike ing someth know could we how of not yet have his own theory any general philosophical conclusions from this solute space, or specific issue about the nature of space. The Inaugu ral Disser tation ("1770). This was to change in Kant's benext work, also the last of his publications on the way to the Critique disserral inaugu Kant's was This I770s. fore the "silent decade" of the Forma et Principiis (On the tation, De Aiundi Sensibilis ible V\lorld), defend ed Intellig e the Sensibl Form and Principles aited ascension to long-aw Kant's and publish ed in August 1770, after March 3 I of that on berg Konigs in the chair of logic and metaphysics n March and betwee written been year. The work is presum ed to have of writing lity possibi the n mentio begun to August, althoug h Kant as 1765, early as ysics metaph for a systematic work on new foundations Proper The on book ming forthco and his publish er had even listed a s8 year. that of ue catalog fair Method ofl'vIetaphysics in the autumn book , But whatev er plan he may have had at that time had come to naught anwrote Kant that 1770 in it occasion demand ed and it was not ce other systematic work, though as it turned out an essay on the substan rather than the method of metaphysics. of This work is a milesto ne in Kant's progres s toward the Critique n betwee ion distinct ental fundam the Pure Reason because it introdu ces on , capacity the mind, the of ies the sensible and the intellectual capacit the one hand, to have singular and immed iate represe ntation s of parmeans of the senses, which Kant hencef orth calls "inticular objects and tuition";s9 and, on the other hand, the capacity to form abstrac t , Further t. intellec the of means ts, general representations, or concep and n intuitio for ies capacit our that as his title suggests, Kant argues les, conceptualization each have their own characteristic forms, princip of basis the ute constit which or laws, which can be known by us and docthe cing introdu argues, Kant metaphysical cognition. Moreov er, intrine that he will later name "transc endenta l idealism," the "laws of means by things of ntation represe the tuitive cognition, "60 or the laws of not of the senses, characterize how things necessarily appear to us, but 61 By contras t, at this stage, alhow they actually are in themselves. though not later, Rant holds that intellectual represe ntation s things, t or concepts, present things "as they are." Thus, sensibility and intellec things mena," "pheno : objects of ts presen t us with two differen t accoun are as they appe3f to the senses, and "noume n3," things as they really (nous).62 t intellec the and are known to be On this account, sensibility and the intellec t oper.ate essentially indedispenden tly of one another . The fundam ental stimulus to this radical

Introduction tinction seems to have been Kant's discovery, perhaps that several paradoxes about the infinite (long known discussed by a number of eighteenth-century philo!sophen;),63 such as the conflict between the supposition that time appears to have no beginning yet any object and thus any universe of objects must a beginning, could be resolved by distinguishing between the intuition as forms of appearance, on the one hand, thought as the forms of reality, on the other: thus it could be for example, that there is no contradiction between the sensible appearance the intellect, that time has no beginning and the reality, known all existence must have some beginning, for sensibility not present the same things. In the Critique ofPure Reason, Kant was to call the set of such paradoxes, to be resolved by the between phenomena and noumena, the antinomies of pure reason. However, there is also a crucial difference between Kant's treatment of the antinomies in 1770 and his eventual treatment of them in 178 L This is connected with an equally fundamental difference in Kant's conception of the relation between the two basic mental capacities intuition and conceptualization in the inaugural dissertation and the Critique. In the dissertation, Kant supposes that the intellect reveals the true nature of reality, and that the antinomies are to be resolved by preventing any limits inherent in the laws of from realbeing misconstrued as limits on purely intellectual knowledge ity. But he has in fact no adequate account of the role of concepts in knowledge of ordinary objects in space and time, and once he realizes - as he after 1772 - that con,eepts of the must be used in conjunction with the intuitions or data supplied sensibility to account for the possibility of such knowledge, not mdlep,cndently, then he will also have to revise his account of the antmOl1me:s. He will have to revise his resolution of them by arguing that be no knowledge of any spatiotemporal reality at beyond of sensibility, although in cases where concepts of the unde.rstan<:1mg can be used to formulate coherent conceptions of no!n-sp~Iti()te:m]poral entities, above all God, there may be coherent belief, even if not any knowledge. In sum, in the inaugural dissertation Kant introduces his fundamental distinction between intuitions and concepts, and uses distinction for a resolution of the antinomies, but does not yet realize knowledge can arise only from the conjoint use of intuitions and concepts to yield a unified experience. Once he comes to that realization, he will have to transform his resolution of the antinomies, surrendering the view that sensibility gives us knowledge of appearances intellect metaphysical knowledge of things as they are in themselves. Only then will the way be open for Kant's fully mature position the


room for certain beliefs cannot become on practical grounds. 64 can We will describe the contents of the inaugural dissertation in some detail, since it will be in reading the to see exacdy what Kant retain~ trom earlier work to be fundamenrevised. Kant signals the importance of problem of the antinomies trom the outset, opening the work with statement "just as analysis does not come to an end until a is reached which is not a whole, that is to say a so likewise synthesis does not come to an end we reach a which is not a that is to say a argues that since the of appearances is WORLD."6S He given with space and time as its and space and time are continuous quantities, there can be "no limit" in analysis or the "regression trom the whole to parts" nor in synthesis or composition, "the progression from the parts to the given "66 and thus no satisfaction of the opening definition of a simple and a world; but since dle pure concepts of the intellect give us access to a realm of things with their own principles of form, where parts are not spatiotemporal regions and the principle of composition is not that of spatiotemporal extension, but where instead the parts are substances and the principle of composition is the common dependence of substances upon God, the conditions for metaphysical knowledge of both simples and a single of them can be satisfied. The remainder of the work is then divided into a fuller statement of the distinctions between intuition and concept and phenomena and noumena (Section 2); separate expositions of the fundamental forms of intuition or sensibility (Section 3) and of the laws of understanding (Section and the concluding argument that the limits of sensibility must not be mistaken to preclude metaphysical knowledge through the intellect (Section 5). Section 3 is taken over into the Critique of Pure Reason without essential modification, but Section 4 will be radically revised by the mature theory of the function of the understanding in the Critique, and once that revision is made there must also be fundamental revision in the treatment of the antinomies in Section 5. In Section 2, Kant first introduces his distinction between sensibility, which is characterized as the "recepti7Jity of the subject in virtue of which it is possible for the subject's own representative state to be afthe presence of some object," and what he fected in a definite way here calls "intelligence rationality)," "the faculty of a subject in virtue of which it has the power to represent things which cannot by their own quality come before the senses";67 he also calls this faculty "intellect" Next, he argues "that things which are thought sensitively are representations of things as they appear, while things which are intellectual are representations of things as they are."69 Kant's reasons for this momentous claim are far from clear. He suggests two reasons:

limits of

Introduction first, that "whatever in cognition is sensitive" as "dependent upon the subject insofar as is or that modification by the presence of objects," where it is assumed that different subjects may be modified or to the same in different ways, and thus cannot all represent the as are; and second, that "objects do not strike the senses in virtue or aspect," but in virtue of their matter, thus form of ... representation ... is not an outline or any kind of schema of the object, but only a certain which is inherent in the mind and means of which it coordinates for itself that which is sensed presence of the object."70 a "leVT1"<:lI" Next, Kant argues that there are two uses of use in which it subordinates concepts, "no matter whence given," to one another in accord witl1 logical rules (e.g., of contradiction"), and a "real" use, in which concepts thi~mseJves, "whether of things or relations," are given. Kant suggests that the logical use of intellect, or "the reflective cognition, which arises several appearances are compared by the intellect" to D[()(1l1ce em:Di)"iCt1! concepts, is sufficient to transform mere appearance into eXl~erteJu:e. Finally, he argues that in its real use the intellect produces concepts, such as "possibility, existence, necessity, substance, cause, etc.," "never enter into any sensory representation as parts," but that can instead be used "dogmatically" to lead to a "paradigm" of "NOU1VIENAL PERFECTION," which in the theoretical context is God and in the practical context is moral perfectionY Thus in its use, intellect supplies no unique concepts of its own, and nizes data supplied by the senses into experience or empirical edge; in its real use, it does supply original concepts of its own, and uses them to know a non-sensible reality as it really is oito a non-sensible goal for our action. This series of claims throws light on doctrines of the sulbs(;qllerlt Critique, but also raises problems that the later work will need to First, the characterization of sensibility as a passive power of the mind and intellect as active will remain central to many arguments in Critique;73 but Kant will also subsume sensibility under the faculties" (Erkenntnisve771ziigcn) generally, and since the term "tc,,,,,'ln." (jacultas, for which Kant's German equivalent is l/p'''111nme n\ iU'}yiiv0 tivity, this means that there is an active element in sensibility as which fits Kant's claim that the fOrm of sensibility is in fact SUI)plled the mind. So it will be important to see that even sensibility passive and an active element: our senses are acted upon by external jects, but we act upon the sensations so induced to give them Further, the two arguments that Kant here gives his claim sensibility represents the mere appearance of things - his eventU:1J

Introduction problematic. His first argume nt is - are "transc endenta l outer objects in differen t ways; nt represe that differe nt subjects might subjects represe nt objects those all that follow not but from this it does one sort of subject who is there maybe are actually they other and maybe indeed that not, do others while ly correct represe nts objects that the form of the is nt argume second His us. one sort of subject is as they are in objects the nt represe cannot represe ntation of objects in the mind." t inheren "law a nts represe themselves because this form unargu ed and d unstate an is there first, here: But there are two issues nt a represe also cannot mind" the in t inheren "law a assump tion that contual intellec since , second lves; themse form inheren t in objects inheren t in the mind used to give form to our reprecepts also are give sentati ons of things, it would seem to follow that they too really they as not and us to appear they as knowledge of objects see that Kant supplies further argume nts for transce ndenare. We later in the dissertation and in the Critique; whethe r tal idealism these argume nts are indepe ndent of the initial assumptions that whatever is receptive and whatev er is formal are inheren tly subjective rather than objectively valid will be an import ant question. Finally, there are major questions about Kant's charact erizatio n of the l" "intellect" here. As we saw, he supposes that we need only the "logica of out nce experie and ts concep al empiric use of the intellec t to generat e mere appearance, and the "real" use of the intellect, in which it generates non-em pirical concepts, is sufficient to furnish knowledge of nonthe empirical objects. Both assumptions will be rejected after 1772. On ted genera ts concep pirical non-em that ze one hand, Kant will recogni ng includi ts concep pirical non-em of list a by the intellec t - in fact, subty, necessi ce, existen ility, "possib as those mentio ned here such sensibility in stance, cause, etc." - must be applied to the data given abstraction mere dge; knowle al empiric or nce order to arrive at experie the other On . purpose this for suffice not and reflective compar ison will cannot lves themse by ts concep those that e hand, Kant will also conclud sense, not do we objects about dge knowle be used to obtain theoret ical nt cohere form to used be ely ultimat can such as God, althoug h they s. ground moral on d validate be can that concep tions of such objects These profoun d revisions in Kant's though t will call for terminological revisions as well. Here Kant speaks of a single faculty, "intelligence" , or "intellect," which has both a real and a logical use. In the Critique or parts two as reason and anding underst n Kant will distinguish betwee the higher cognitive faculties of the mind.74 perhap s better aspects Unders tanding will be the source of non-em pirical categories or "pure by concepts of the underst anding " that must be applied to data furnished only but use real a have thus and dge, knowle al the senses to yield empiric for empirical objects; further, since Kant continues to believe that sensi40

Introduction bility furnishes mere appearance, the real use also be confined to appearance. Reason a has a legitimate logical use insofar as it links judgments constituted with concepts of the understanding into more complex, inferential structures, but has a mistaken real use if it is thought that either means of inference or the use of concepts of the understanding accompanying data from sensibility it can obtain knowledge of non-(~mlPlr!cal objects such as God. The only legitimate real use of reason will be to mulate conceptions of non-empirical objects that be vaIHJate:C! moral considerations; that is, reason has a real use as pnlctical son. Thus, reason will be denied the power of introducing a p~lra{::l1Rn1u of "noumenal perfection" on theoretical grounds, though it retain the power of introducing the practical paradigm of "moral peJrtel:tH)n" ideas and will be able to justify a certain non-cognitive use of as what Kant will come to call "postulates of reason."75 The few paragraphs of Section 2, introduce iUJ1d:lmental sumptions of the Critique ofPure Reason as well as positions radically revised. The three paragraphs of Section 3, contrast, sent a treatment of the forms of intuition, space, time, that carried over into the Critique largely unaltered, in the second edition of the Critique) somewhat amplified. Here Kant claims that the principle of form of the world as appearance or nomenon is "a fixed law of the mind, in virtue of which it is necessary that all the things that can be objects of the senses ... are seen as necesare in sarily belonging to the same whole. "76 He then argues fact two such laws or principles, time, the form of all we sense, whether inner or outer, and space, the form of our outer sense, or our sensory perception of objects we take to be distinct from ours(~lves. Kant argues that space and timc are both the purefgrms of all inl:uitions, intuor ''formal principle[s] of the sensible world, "77 and themselves itions: 78 They are the forms in which particular objects are to us by the senses, but also themselves unique particulars we can have a priori knowledge, the basis of our a priori knowledge of both mathematics and physics.7 9 But the embrace of space and time "is limited to actual things, insofar as they are thought capable offalling under the senses" - we have no ground for asserting that space and time acterize things that we are incapable of sensing. 8o Kant makes the following claims about time: 8r "7Z?e idea of time does not arise from but is presupposed by the senses": this is because any concepts we can form from our experience of things already presupposes that we can represent them as either simultaneous or successive. "The idea oftime is singular and not general": this is because all Da:rm:ular times, say two particular years, are thought of as part of a single are larger time, in which they each occupy a determinate position,

"The idea a similar type. not just unrelat ed tokens singuly " precise because it is tuition, " and indeed a it an makes which nce, experie lar and immed iately given to us in an our ted abstrac than rather osed intuitio n, but also given to us as presupp be will claims these of pure. it makes our experience, from the h althoug , revision t withou of Pure Reason reiterat ed in the 82 somew hat amplified. exposition of them made in tI~e initial disNext, Kant asserts a claim that is not in a numbe r of later osed but is presupp cussion of time in the is a continuous mag"Time (4) that claim import ant parts of the that betwee n instead parts no simple nitude," or that it consists , smaller inanother always is any two times, no matter how small, there in Section given already terval of time. Then Kant adds to the reasons nor is it a real, and objective ng 2 for the claim Lftat (5) "Time is not somethi see to ant import is It ." substance, nor an accident, nor a relation posThe (5). claim this to there is both a positive and a negative aspect betime of n intuitio pure a we must have itive side is the argume nt or objects lar particu any our percep tion of cause it is presupp osed now Kant which nt states as simultaneous or successive, the argume we must have a pure represe ntation of time reiterates. This implies percept ion, but does not imply indepe ndent of any particu lar real," that is, nothing but a form of that time is not also "objective sorts represe ntation . For that further, negative claim Kant suggests two "the and n Newto against of reasons : a metaph ysical reason, aimed or a ce substan a as time e English philosophers," that the idea of absolut an absurd; is dei) m sensoriu proper ty of any substance (such as of ing conceiv that , Leibniz epistem ologica l argume nt, aimed against would objects of s relation ed time as someth ing we abstrac t from perceiv etely render our knowledge of it merely empirical and therefo re "compl and atics mathem of rules ental destroy" all the certitud e of the fundam r, howeve nt, argume l ologica premises of this epistem physics. The hinted are there even and are not spelled out before the "time, posited in itself and abthat at. 83 Finally, Kant eless, as "the universal neverth solutely, would be an imagin ary being," is "to the highest deit outer, or form of phenom ena," whethe r inner sensible world." (7) "an gree true" 84 He claims space. about claims Kant makes a series outer sensations," because I can ofspace is not "The repreme outside "only conceive of someth ing as place the diversoJ .. . loco from which is senting it as in a t of concep the t abstrac cannot I , in other words, in which I am canI because from distinct space from my experience of as m withou t already represe nting not experience them as " singular a is ofspace space. (2) Like that of time, "the concept

Introduction because regions space are represented as parts a single, UU1JU(1less space rather than as instances of some general sort. As Kant infers from these two arguments that (3) "The concept ofspace is thus a pure mt:luti'on," an intuition because it is singular and pure because it is not "compounded from sensations" but presupposed "outer sensation" or experience of objects as distinct from ourselves. Here Kant skips an argument that space is a continuous though he also assume that in the Critique, and instead inserts the argument from 1768 about incongruent counterparts, using it now to show that since features of directionality such as a right- and left-handedness are not inferable from the concepts of objects they must be "apprehended a certain pure intuition." (This argument be omitted from the as in the case of time, Kant infers from these results that "Space is objective and real, nor is it a substance, nor an accident, nor a re1aul:m; is rather, subjective and ideal; it issues from the nature of " Again, he infers this from the prior arguments that it is scheme ... for coordinating everything that it senses externally" also two additional claims, the metaphysical claim made English" that the idea of "an absolute and boundless rec,?pttlcle ble things is absurd" and the epistemological argument Leibniz that conceiving of the propositions of geometry, are taken to describe space, as merely abstracted from an experience of relau()lls among objects would "cast geometry down from the summit of certainty, and thrust it back into the rank of those sciences of which the (5) even principles are empirical." Finally, Kant again concludes though "the concept ofspace as some objective and being or rw,,"n,~rtv be imaginary, nonetheless, relatively to all sensible things whatever, it is not only a concept that is in the highest degree true, it is also of all truth in outer sensibility." This is as good a statement of the docintrine of transcendental idealism as we will find in the Critique sisting on both the subjectivity yet also universality necessity of space as a form of representation. 85 This account of space and time as the forms and of the sensible world, as we have said, remains essentially unchanged in Critique. In Section 4 of the dissertation, however, Kant gives an account of the "principle of the form of the intelligible world" that is largely unchanged from his earliest work but will disappear from the Critique. The content of this section is basically just the argument that a multitude of substances can constitute a single world in virtue of their common dependence on a single cause. This ar~:unlerlt is based on the thoroughly Leibnizian premise that "the existence of each [necessary] substance is fully established without appealing to any dependence on anything else whatsoever,"86 and the further inference that contingent substances, the only kind which might therefore con43

Introduction their are charact erized precisely stitute an interre lated in virtue single a ute constit re therefo penden ce on a cause, 87 le reconci to t attemp Kant's cause. n commo a of their depend ence on l physica of theory the to on attracti nding longsta his this argume nt with 88 ing. unavail is , objects t distinc n betwee tion influx, or real interac also the underly ing assump tion Howev er, not only this argume nt as the concep t of substance, can such t, pure concep ts of the intellec dge of things as they are in knowle be used on their own to provide . This particu lar argume nt Critique the from themselves will disappear tion among physical obinterac that nt argume "vill be replaced by the them as simultaneously ncing experie for on jects is a necessary conditi a single space (Kant's in ns positio inate occupy ing differen t yet determ 9 metaphysics will be ing underly the and import ant "Third Analogy"),8 ies of the undercategor pure that n positio Kant's critical replace d for theoret ical used if illusory are b.~at standin g lead to ideas of reason tes of practipostula as serve can they knowle dge on their own, though cal reason. in The same transfo rmatio n awaits Kant's treatm ent of "metho d tion. disserta al inaugur the of 5 Section metaphysics" in the conclud ing arguing that philoso phy has no special method to preKant begins only scribe to ordinar y science, because here the use of the intellec t is inthe by d provide lves themse not are logical, organiz ing concep ts that metaof case the In nce. experie from ted tellect but are instead abstrac ng physics, however, where the intellec t does have a real use, supplyi 0 metaof method The 9 science." all precedes original concepts, "method care physics, Kant then maintains, "amoun ts to this prescription: great transn cognitio sensitive to native are that es must be taken lest the principl "9 ' The fungress their limits, and affect what belongs to the comes from is, that ysics, metaph in ss damen tal obstacl e to progre of sensibillimits t inheren and ons conditi ry assumi ng that the necessa well. Kant as dge knowle tual intellec of ity are limits on the possibility These unon. confusi this from arise that lists three "subrep tic axioms" warran ted assumptions are:


also the conditi on of the possibility itself of the objectY In other words, at this stage Kant holds that it is a mistake to assume and that the characteristic forms and limits of sensible represe ntation s

the compare what is given so as to form a concept ofthe understanding of object. the object, is also a conditi on of the possibility itself of 3. The same sensitive conditi on, under which alone some object met with can be subsumed under a given concept of the understanding, is

The same sensitive conditi on, under which alone the intuitio n of an object is possible, is a conditi on of the possibility itself of the object. The same sensitive conditi on, under which alone it is possible to


Introduction the conditions for the application of concepts to sensible rerlre"entations our metaphysical cogniti on of objects as are. gives examples of the errors that arise from this assumption: It is an error to assume that whatev er exists is located in space and time;93 it is an error to assume that" every actual multiplici~y can be given " as multiplicities given in space and time are, and that "every magnitude is finite;"94 and it is a mistake to assume that may be an empirical criterio n for the application of a concep t, as non-ex istence at some time is a sensible criterio n for the modal concep t of COjGtlngency, is actually a necessary feature for any use of the concep t at allY5 1nlplication of Kant's argume nt is that paradoxes may arise in to derive metaph ysical knowle dge from the conditi ons of serlsltllhl:y, One such paradox is that if the world is represe nted as existing in space and time, then the world must be both finite and infinite. Now Kant's argument further implies that such paradoxes can be avoided bec:aw,e we Gm have intellectual knowledge of reality of the concepts of space and time as conditi ons of "sensitive cognition." Finally, Kant concludes the section by mentio ning, as an terthought, that there are certain "principles of convenience" convenientiae) that are not principles of sensitive cogniti on rules by means of which "it seems to the intellect itself and cal to deploy its own perspicacity." These are the "all things in the universe take place in accordance with the order ofnature," that "principles are not to be multiplied what is absolutely necessfll:y," and that "nothing materia l at all comes into being or passes away."96 This is a striking list, because it includes two principles - the principle sal causation and the princip le of conservation of substance - that Kant will later identif y as "consti tutive" or necessa ry conditions of the possibility of the experience of at but another one - the principle traditionally called "Ockha m's razor" is more like what he will later identify as a "regulative" which is not a necessary conditi on of the possibility of any experience at all an assumption we make for various subjective reasons. The fact that Kant could indiscriminately mix what he would later distinguish as constitutive and regulative principles shows that he did not yet have a clear concep tion of the function of the former as necessary conditions of the possibility of experience, a consequence fact that he did not yet have a dear unders tanding that the pure concepts of the unders tanding (such as the concepts of causation and substance mentio ned in these principles) can yield knowledge applied to data furnish ed by the faculty of sensibility. Likewise, that he could argue at this stage that metaphysical illusion can be avoided not letting the conditions of sensibility limit the use of concepts intellect shows that he did not yet see that the concepts of the under45

Introduction to "c.,""'U"Hy standin g a practical have can its therefo re to the tion disserta ral inaugu the use. Before he could progres s from tion concep new a to have would Kant Critique the of use e sensibl among ions distinct of the use relithe and reason, pure use of underst anding , the reason. able use pure


of the dissertation, Kant fell into a proa few minor essays97 and a series of letlonged silence ters to his student l\1arcus Herz. Herz had particip ated in the was now in Berlin, studyin g medidefense of Kant's disserta tion98 philoso phers of the capital. cine but also in contact with the these letters, our primary from Aside from what little can be gleaned these years comes from in t Kant's though source of inform ation these are only a surviving margin alia and notes, though and have to be period this fragme nt of what Kant actually wrote during materials these r, howeve are, 99 used with caution. Fragme ntary as they ant import most the of some of cast considerable light on the emerge nce troumost its some explain also new argume nts of the Critique and blesom e obscurities. In the fall of I770, Herz went off to Berlin with copies of the disserh tation for leading intellectuals such as Mende lssohn, Johann Heinric acand 779), (I720-I Sulzer Georg Lambe rt (1728- I777) and the letter to Lambe rt survives. In this compan ying letters, of which promis e of collaboration, letter Kant apologizes for the lapse of a work on the metaphysics a of tion publica makes a promis e for the to be redeem ed for anbegin even of morals (a promis e that would not continu ing commi this evinces ise otherw other fifteen years)IOO dissertation. IOI By the in ted enuncia ment to the view of metaphysics with letters conreplied had Christm as, all three Berlin philoso phers hold time to be Kant could how n: taining essentially the same objectio is the form of time when reality e a mere appeara nce with no objectiv s in inner change of nce experie iate have immed inner sense and we to might we ance signific l sense regardless of whatev er externa 02 those changin g interna l senses?I of Lambe rt initially raises a questio n about whethe r Kant's "two ways sepately comple so "are t, intellec the knowin g," from the senses only rated that they never come togethe r," ' 03 but then discusses in detail sinis time nts argume Kant's ng Kant's treatme nt of time, accepti

1770-7 2. Mter the


Introduction gular, C011tulucms, and the object of a pure intuition Kant's idealism time: objectmg- to

All changes are bound to time and are inconceivable without time. real, then time is real, whatever it may be. is then no real. I think, though, that even an idealist must grant at least that cn:dIli,es exist and occur in his representations, for example, their beginning Thus time cannot be regarded as something unreal. I04 Sulzer's briefer letter also raises a problem about position that duration must have "a true reality" even if u~e concept of time is some sort of abstraction from our experience of real tion; ID 5 and Mendelssohn too objects that For several reasons I cannot convince myself that time is something subjective. Succession is after all at least a necessary condition of the representations that finite minds have.... Since we have to grant the reality of succession in a representing creature and in its alternations, why not also in the sensible objects, which are the models and prototypes of representations in the world?I06 Kant made no immediate to this objection, as we letter to Herz of 7 June 177 1. ID7 He merely asked Herz to ap()log'lZ:e his correspondents saying that their letters set him on a series of investigations, and then Herz he was now Occulpie:d with a work that "under the title The Bounds of Sensibility and Reason' would work out in some detail the relationship of the concepts and laws determined for the sensible world together with the outline of what the nature of the theory of taste, metaphysics, and contain."ID8 In his next pledge, Kant said that he expected to co:rnrllet:e plan of the work shortly. Kant does not appear to have written to Herz again until 2 I 1772, when he wrote what has become his 1T1OSt letter. lIere Kant reviewed his plan for the work mentioned the previous stating that it was to consist of "two parts, a theoretical a " the first of which in turn consist of a general phenonle11ology and (2) metaphysics, but this only with regard to its nature and mc~ttlIOd," while the second part was to deal with "(I) the universal princ:lpl(~s of feeling, taste, and sensuous desire and (2) the basic principles of ity." I0 9 However, Kant says, as he thought about the theoretical part - where the "phenomenology" was to have the of sensitive cognition before the purely intellectual of metaphysics were expounded - "I noticed that I still something essential, something tlut in my long metaphysical studies I, as well as others, had failed to pay attention to and that, in fact, constitutes the key to the whole secret of hitherto still obscure metaphysics." But the fundamental problem that Kant now announced had notlling to objec-


Introduction raised; insavants tion to his idealism regardi ng time n, lIO he objectio that ledge acknow deed, althoug h Kant time. of ideality the about would in no way rethink his L'1stead, Kant raises a comple tely differe nt question: "What is the the ground the relation of that in us which we call represe ntation to of ship relation the of case the object?"1II This is a puzzle precisely in exsensible by ed present pure concepts of the unders tanding to objects ntaperience. It is not a puzzle in the case of entirely empirical represe case the in nor , objects l their externa tions, which are merely caused obthe where ns), intentio human we may of divine archetypes now Kant But, . ntation represe the anteced ent ject is merely caused must . holds, "the pure concepts of the unders tanding ... though they ted formula are they because have their origin in the nature of the soul" senof objects to apply yet by us and known "compl etely a priori," must the sible experience even though they are neither caused by nor cause questhis over passed tely latter. II2 Kant now admits that he had comple that tion in the inaugural dissertation because he there failed to realize the to applied be must our pure concepts as well as forms of intuitio n exbe now must what same objects, the objects of our experience. Thus nce experie ... which plained is "the possibility of such concepts, with of must be in exact agreem ent and which nevertheless are indepe ndent protanding unders the experience." The idea that the pure concepts of senvide knowledge of entities other than the spatiot empora l objects of sibility suddenly disappears. Kant did not describe how the possibility and necessity of the agreebe ment of experience with pure concepts of the unders tanding is to these of cation classifi atic explained, beyond sugges ting that a system can "conce pts belong ing to comple te pure reason" or "catego ries" ." anding underst the of laws be reached by "following a few fundam ental to ready be would he that nt In spite of this obscurity, Kant was confide the first time entitled a Critique of publish the work, which he now Pure Reason, in only three months !"3 In fact, it would be almost nine was years before the work with that title appeared. Much of this delay catethe why of idea clear due to the fact that Kant did not yet have a gories necessarily apply to objects of experience. will As Kant thinks further about this problem , a problem about time but time of reality the about play a key role, though not the problem about nts judgme inate rather a problem about how we can make determ in the order of objective states of affairs or even our own experiences sevthe in n attentio Kant's of time. This problem will become the focus will eral years following the letter to Herz, especially in 1774-75, and remain central in the Critique. Kant's next report on his progress is in anothe r letter to Herz, this "be one written toward the end of 1773. "4 Kant writes that he will not

Introduction seduced any author's itch into seeking " suggests working on "a principle that will completely been a riddle and that will bring the misleading of the selfalienating understanding under certain and easily applied mles," but nevertheless promises that he will have his book, which he continues to call "a critique of pure reason," ready by the following Easter or after, I I 5 that is, in spring of 1774. In Kant's next letter to Herz, however, written three years later in November 1776, we find him suggesting that he has been held up by difficulties SUITOIJncIing the fundamental principle of his new position, he says that he made progress with it the previous September once promises the completed book by the following Easter. August finds Kant reporting "a stone in the of the Pure Reason," though once again he is optimistic that can get this obstacle during the following winter (1778). Bur April 1778 finds Kant writing that the mmor that some pages of his book are at the press is premature, and in August of that year Kant is "still working indefatigably" on his "handbook." So for at least five years the completion of the pnJmlsed that Kant has not tinues to be put off, and there are repeated found the fundamental principle he needs, the one that would answer the fundamental question of 1772 . .From the to Herz, the only one of his known correspondents in this to whom Kant says anything at all about his planned book, it might seem as if Kant was making no progress at all. But our other sources reveal he was indeed working "indefatigably" on the Critique this period, and that beginning by April 1774 - in other words, in VICl11ity of his first promised Easter completion date - Kant did begin to explore a solution to his puzzle about why a priori concepts of the understanding should necessarily apply to the data presented to us sensibility and not have any constitutive, theoretical use outside that application. 1774-75. Using a letter sent to him on 28 1774 as scrap paper, Kant wrote a series of notes that were clearly part of his work on the Critique. Much of the material goes over claims about space and time ala line ready established in the inaugural dissertation, but Kant now of thought that had not previously appeared. He says that the of time implies the unity of the self and the determinate position of jects in time; even more explicitly that the unity of space on unity of the subject and on the ability of the subject to assign representations of objects determinate positions in space; and then suggests the concepts of the understanding are necessary precisely to acJ!ue:ve such unification of and order among the intuitions of objects pn;sejGte:d in the form of time and space. In his words, he asserts first:



Introduction in mythis: I can intuit all objects Tim e is unique [einig]. INhich means cts of obje ible poss own subject, and all self and in representations found in form ial spec the with ce in accordan my intuition stand in relation to each of this intuition ... their determinate position in time. For 4. All things and all states of things have to must have their determinate relation through the unity of inner sense they . Il 7 all other putative objects of intuition our abo ut spac e - spac e is not He the n mak es para llel clai ms unialso but s, elve ours to cts exte rnal uniq ue form for repr esen ting obje pos imus t be assi gne d a dete rmi nate ct obje ry eve that e fied in the sens othe rs in it: tion in rela tion to thus mere form even without given matter, Space is nothing but the intuition of y of unit the of unt acco on n tatio esen repr pure intuition. It is a single can cts obje r in which all representations of oute the the subject be placed next to one another. uS and ing or the use of con cep ts of the und erst Finally, Kan t sugg ests g repgnin assi of on nec essa ry con diti rule s asso ciat ed with them is the ied unif a in ns itio r dete rmi nate pos rese ntat ions or thei r obje cts thei spac e and !or time : inthe senses; thus no other concepts can We have no intuitions except r orde and on ositi disp the to ain e which pert habit the understanding except thos and l, ersa epts mus t contain what is univ among these intuitions. The se conc the the learned understanding; in concreto: acto: abstr rules. The faculty of rules in all in ce eren pref has ing mon understand common understanding. The com re whe but s; case the from da cases, where the rules must be abstracte 9 it does not obtain at all." they have their origin a priori, there cep ts are used only in 3pp lica tion Thi s rem ark pres upp oses that con that was t had not yet seen in 177 0 to intu itio ns, the thes is that Kan famous its Reason, with of the to bec ome the with ns itio intu with out con tent are stat eme nt that icupart S)' It furt her sugg ests that the out con cep ts are blin d" (AS rlB7 to is play cep ts of the und erst and ing lar func tion that the a priori con s ition intu ng pos itio n and ord er" amo serv e as rule s for esta blis hing "dis be uld sho ts con cep of obje cts, thou gh Kan t doe s not as rule s for this tion func ts cep con how nec essa ry for this pur pos e or use of abst ract ion Kan t sug gest s that eve n the purp ose. n the use of the emp irica l con cep ts dep end s upo for the pro duc tion for the esta blis hme nt of this "dis a priori con cep ts of the und erst and ing seem may ts con cep " eve n tho ugh thes e a pos itio n and because it nt orta imp an is s Thi on." "lea rned " rath er than "co mm in the out con cep ts to be of a imp lies that the ry of theo a , bee n seen Reason is not, as it has som etim es


Introduction the founda tions of natural science conside red as separat e life, rather a theory of the founda tions of science as continu ous all of our knowledge. The following series of notes shows that Kant spent much of his time in the next several years trying to work out his that the categories can be shown to be a priori yet necessary conditi ons our knowledge of objects by showin g their use is necessary cOJtldltlon all determ inate "dispos ition and order" of intuitio ns. These notes are assigned to the year 1775 because one of them is written on letter to Kant dated 20 May 1775. Although, as we saw, Kant had been moving toward the idea of a fundam ental contras t bel:WE~en real relations throug hout the 1760s, it is only in these notes clearly links his fundam ental philosophical problem with the distinct ion between judgme nts that are analytic those that are SV1[lttletlc Kant asks under what conditi ons a predica te b can be pr(~dlcalted of an object x that is also subsum ed under anothe r predica te a. In some cases, b can be predica ted of any x of which a is pn~dlcal:ed bec:aw,e predicate b is already to or contain ed in a, we have no to consider or experience any particu lar x in order to see cases, a propos ition of the form x's that are a are true in virtue of "the princip le of identit y and contra(llctlC)n, "merely logical " "princi ple of form rather conten t," that is, it would be 120 however, the predicates a and b can be related to eaeh other only throug h x, then the judgme nt is synthetic: "If I refer both predicates to x and only thereby to each other, then it is " and the predicates are in that case "not in a but in a tion."'2' Kant also says that "In synthet ic the relation tween the concep ts is not immed iate this happ~enls case of propositions), rather it is represe nted the conditi ons of their concret e represe ntation in the subjeet."I22 Kant does not say so explicitly, but he is clearly already assumi ng that pr<>pc)sHi0I1S a5iseltm g that concep ts to the objects of into this class of synthetic judgme nts expressing real relations. Kant's next step is to argue that there are three differe nt ways in which synthet ic judgme nts may be made. The x means of which we link predica tes a and b may be constru cted in pure intuil::10 ]rl, it may be given in empirical intuitio n or or it may be "~l-te sensible of subject within a is to be assigned its position."'2 3 Or, in anothe r passage, writes: x is therefore the determinable (object) that I think through the concept a, and b is its determination or the way in which it is determined. In x is the construction in experience it is the and with to an inherent representation or thought in general x is the function 111 general in the subject. "4

Intr odu ctio n

the first two opti ons . In mat heKan t mea ns It is dea r eno ugh inte rior angles as "Th e sum of matics, syn thet ic judg men ts - such con firm ed by or e mad are t angles" of a plan e tria ngle equals two righ tria ngle ") in ane ("pl e icat satisfying the first pred con stru ctin g an the seco nd fies satis n ctio the con stru pur e intu itio n, and then seei ng that n yields ctio stru con a such t angles"); pred icat e as well ("equals two righ 179 or not , rees deg 180 righ t angles con tain a dete rmi nate answ er ds a yiel it but ct, obje ar n of a part icul 181) bec ause it is the con stru ctio form the n, itio intu e pur in because it takes place resu lt that is a ial all possible trian gles or othe r spat ctur e stru the that dete rmi nes synhes blis esta tion erva experience, obs figures or objects. In prop osit ions con ting ent or a but nate thet ic and dete rmi n like itio pos pro a to part icul ar experience: because of the ("w orn tion rma info s add is wor n and dog -ear ed" cop y of the obje ct the of ion ript desc the goes and dog -ear ed") be only can tion rma info but that add itio nal cop y of the ing noth has it use beca , obje ct that is observed asse rted of the t mea n of app eara nce . But wha t does Kan form l to do witll any essentia sen"the as ses phra by such obs cure his thir d case, refe rred to "? eral gen in king thin func tion of sible con diti on of a sub ject " or that ely nam 4, 177 in is wha t he hint ed at \Vh at Kan t has in repof er" ord and n itio the "dis pos ther e are cert ain rules necessary for g to a unified self and occupying ngin belo as of ed rese ntat ions con ceiv es its e and tim e in whi ch that self plac dete rmi nate pos ition s in the spac conthe to ons diti con eral that these rule s add gen repr esen tatio ns, ar icul part the ond bey go erie nce that cep t of any possible obje ct of exp of ns mea and erve obs hap pen to feat ures of such obje cts we may of s step ther toge gs to refe r to them . He brin whi ch we may age: this argu men t thus far in this pass a, in pert ains prop erly to the conc ept

[h] In analytic judg men ts the pred icate is not the concept, since the pred icate [h] of ct obje the to ts men synt heti c judg ept conc a to , the obje ct that corr espo nds cont aine d in the concept. However conin tion posi its i.e., ication of this concept, has cert ain cond ition s for the appl is also epts is sensible; thus, if the conc ept conc all of ition creto ... Now the cond in its gle trian a e.g., eto, concr in ed cons ider sensible, but universal, it mus t be then l, irica emp signify pure intu ition , but cons truc tion . If the conc ept does not the i.e., , time and e spac relative posi tion (a) in x cont ains the cond ition of the I25 . them in ing som ethi ng cond ition for universally dete rmin

ts wha t Kan t is saying is that judgmen Thi s is still som ewh at obs cure , but be ly universal, that is, a priori, can that are syn thet ic but also gen uine judg such s, atic hem mat of the case gro und ed in one of two ways: in the ctio n of a mat hem atic al object; in men ts are gro und ed in the con stru rmin dete of on diti con he in.t und ed othe r case, such judg men ts are gro ct in space and tim e to others. ing the relative pos itio n of one obje


Introduction Kant also puts this point saying that

principles of the exposition ofappearances,

means pn~ciselv assignment of each representation to a determinate in the unified space and time that is the framework for all the representations belonging to a unified self. Kant introduces this concept when he writes: The principium of the exposition of appearances is the general of tt1.:le exposition of that which is given. The exposition of that which is thc)Ug:htcie1perrds merely on consciousness, but the exposition of that which is given, if one regards the matter as undetermined, depends on the ground of all rclation and on of the linkage [Verkettung] of representations (sensations).... The appearances is therefore the determination of the ground on which the nexus of sensations depends. 126 But perhaps a dearer statement of Kant's strategy is this: of disposition as well as of affection. The apThere is in the soul a of pearances can have no other order and do not otherwise belong to the the power of representation except insofar as they are amenable to the common of disposition. For all appearance with its thoroughgoing determination must still have unity in t.he mind, consequently be subjected to those conditions through which the unity of representations is possible. that which is uisite for the unity of representations belongs to the objective conditions. unity of apprehension is necessarily connected with the unity of the intuition space and time, for without this the latter would give no real representation. the The principles of exposition must be determined on t.he one side laws of apprehension, on the other side through the of the power of understanding. They are the standard for observation and are not derived from perceptions, but are the ground of those in their entirety. "7 Kant's argument is that although particular representations are given to the mind in temporal form, and all representations of outer objects are given to the mind as spatial representations, these rq)re:seJI1taltiCll1s cannot be linked to each other in the kind of unified order the mind demands, in which each object in space and time has a determinate tion to any other, except by means of certain principles that are mJtWJrerlt in the mind and that the mind brings to bear on the appearances it experiences. These principles will be, or be derived the pure concepts of the understanding that have a subjective origin yet necessarily apply to all the objects of our experience, those concepts not have any determinate use except in the exposition of appearances. is the theory that will answer the puzzle Kant raised in his letter to Herz of February 1772, and that will eventually allow to write the

But how exactly will the categories be shown to be the necessary conditions for the exposition of appearances? This has by no means made clear in anything cited thus far. Kant throws out a number tan53


in it was taliz ing ions gest sug e of thes cho osin g betw een as e befo re 1779. One shap l fina from the that app eara nce s in the of the thin g that Kan t sugg ests is that ss ima unif ied min d or self -con scio usne ord erly ain cert a is e ther app eara nce s bec ause pose s cert ain prin cipl es on t wha or d con ceiv e of a unif ied min way in whi ch it is nec essa ry to to is ver hate - itself. He stat es that "W Kan t now calls stan ds und er a rule of app erce pas an be ther e is a "thr eefo ld then clai ms or self -per cep tion , of app erce pbeca use ther e are "thr ee dim ensi on e of our ceiv con the way in whi ch we tion " or thre e "ex pon ents " ing " I. hav as ts of our own tho ugh selves: we nec essa rily con ceiv e r, and othe each of succ essi on amo ng 2. a rela tion to a cate e sam e " and we ther efor e imp ose thes 12 3. [com pris ing] a - on 9 tion rela as the cate gori es of Kan t will gori es says t Kan t, this argu men the obje cts of our repr esen tatio ns. in cts obje of " that is, I con ceiv e that "I am the orig inal of I3 0 Alte relf. mys I mus t con ceiv e ana logy with the way in ceiv e of objects con ily ssar nece we natively, Kan t som etim es sugg ests a to whi ch both a succ essi on and usin g the cate gori es of a sub ject the and es selv then con ceiv e of our who le of prop erti es belo ngs , and the way we nece ssar ily thin k of obwith unit y of our tho ugh t in ana logy to a argu es that "All exis tenc e belo ngs jects. Thu s, in ano ther note he ing ryth eve es; s is a mem ber of a seri subs tanc e; eve ryth ing that hap pen rdete ally proc who le who se part s reci that is sim ulta neo us belo ngs to a e ceiv con we ch ests that the way in whi min e each othe r," and then sugg e thes to nds of exp erie nce, corr espo of ours elve s, as subj ecti ve orde rs of obje cts. '3' tmld :lme ntal ways for con ceiv ing that rem arks , however, Kan t sugg ests g isin In som e of his mos t prom cerof use the of win g the nece ssity ther e may be dire ct argu men ts sho ions inat term ing for cert ain tim e-de tain cate gori es of the und erst and een the way in whi ch we conceive betw s to anal ogie with out any s, Kan t argu es that asof obje cts in eith er dire ctio n. Thu of the self eeven ts in time pres upp oses a fram sign ing dete rmi nate pos itio ns to r othe the in that sam e cate gori es wor k of prin cipl es emp loyi ng the ct: obje an of or con cep t of a sub ject passag'es he has asso ciat ed with the

of perception), precede an occu rren ce (con ditio n Som ethi ng mus t e ther e is one ede an occu rren ce, but amo ng thes ~AJl sort s of thin gs can prec from whi ch it always follows. it) to in time and that whic h dete rmin es a A reality is always attached ed rmin dete ugh whi ch the poin t in time is som ethi ng acco mpa nyin g it, thro (con ditio n of perc epti on).


Introduction All sorts of things can accompany, but among them there is that is always there. Witt'l regard to that which is simultaneous there is a connection tion of perception). But it can be accompanied with all sorts of things; what is to be considered as objectively connected is a mutual determination of the manifol d one another. If there were not something that always was, thus sOlmethlngthere would be no firm point or determination of the ception, i.e., determination of something in time. If there were not something that always preceded an occurrence, then among the many things that precede there would be nothing with which that which occurs belongs in a series, it would have no determinate in the series. Through the rules of perception the objects of tt'le senses are determinable in time; in intuition they are merely given as appearances. In accordance with those rules there is found an entirely different series than that in which the object was given. '3 2 Here Kant suggest s that he has previou sly the of appeara nces" is the determ ination of a and PO:Sltlon occurre nces in time. He does not say whethe r the occurre nces are resenta tions in a subject or states of objects , but in either case to them in time is to determ ine whethe r at some or in time such occurre nces succeed one anothe r in a specific order or are simulta neous with each other. In order to determ ine this, Kant we have to posit the existen ce of objects that endure time - substan ces - and the existen ce determ inate of causation and interac tion among them. we need to use the fundam ental categor ies of substan ce, causati on, and interac tion for timedeterm ination or the "exposi tion of appeara nces." Kant does not explain in any detail why we must use these categor ies to accomp lish this end - a fuller explana tion of will await section of the publish ed Critique called the "Analo gies of Expelnel [1C(;" (A 176-2 I8/E 218-65). In the Critique, the "Analo gies" a se~)ar:ate argume nt for the univers al and necessa ry of the categor ies certain more abstrac t concep tions of both objects and apperCt~ pt1011, which he calls the "Transc endenta l Deduct ion of the Pure of the Unders tanding " (A84-q o and El 16-69). Since in Kant's ori,gma l sketches of the central argume nt of his planned Critique there is no separation betwee n the discuss ion of apperce ption, objects , the exposition of appear ances, and the origina l discuss ion of the relatio n between apperce ption and objects already has the form of an it is an endurin g questio n for the interpr etation of the Critique or not these two section s have render ed asunde r conside rations that should have remain ed joined. This is a questio n any reader of the


Intr odu ctio n

L'f ,/[1 (.,f UI::

betw een to con side r in tryi ng to ana lyze "Analogies." "Tr ansc end enta l Ded ucti on" be as far as Kan t had gott en to seem 177 6-7 7. The se thou ghts s from arou nd 177 6-77 , we find for 177 5 In several furt her extensive note ine for a whole inte nde d book. In the the first time wha t looks like an outl und er four headings: "Dialectic first of these notes, Kan t divides his and ing - Transcendental The ory of of Sensibility"; "Dialectic of Und erst ory of App eara nce - Rea lity and Mag nitu de" ; "Tr ans cen den tal The ory of Experience."I33 Thi s fourfold Neg atio n"; and "Transcendental The as elaborate a conception of the indivision does not, however, headings all cover use the first tend ed wor k as it mig ht seem to, beca space and tim e as already theo ry the same grou nd, namely, Kant's e. The fourth part adds to this a stat stated in the inaugural dissertation subof s cept con the g erience involvin men t of the thre e principles of exp were first clearly listed in R468I. that ion ract inte stance, causation, d first thre e sections all have the wor Further, in spite of the fact that the icexpl t Ran that in the four-ch section "dialectic" in thei r titles, it is only es of the kind that we tlnd in the thes anti itly states bot h theses and in the thou gh he also hint s at anti nom ies "Dialectic" of the trea tme nt of space and time. organization of his Ran t is still exp erim enti ng with the At this y he here envisages incl udin g is fairl plan ned work. But the con tent that s time and e spac "All he maintains that clear: First, abo ut space and time, them are time and e "All parts of spac are parts ofla rge r ones," and that s part ple sim no are lies that ther e selves spaces" and times. '34 Thi s imp that space and are continuous, e tim and e in space and time, that spac of space or on regi a ter how large time are intlnite yet unit ary (no mat lies that imp t Kan er space or time).'35 tim e is, it is always part of one larg space that me assu to ms we also have in ord er to und erst and t..hese clai 6 Und er the title of "Di alec tic of Und erand tim e "are noth ing real."I3 of Mag nitu de" he furt her states that stan ding - Tra nsce nde ntal The ory time implies the tatio n of space alth oug h the natu re of our repr esen time, neverand e or division of spac infinitude of the possible extension nsible" [unehe mpr past tim e are inco theless "Inf init e space and infinite re of the natu the flic t betw een begreiflichJ."37 Thi s sug gest s a con cept or con tual llec inte natu re of an intu itio n of space and tim e and the flict is con this how t does not explain com preh ensi on of them; but Ran to the only ng belo e "Space and tim to be resolved bey ond asserting that 38 ld."" wor the ond ld and not bey appearances and ther efor e to the wor den tal The ory of Experience." The n Kan t turn s to the "Tr ans cen Her e he asserts thre e theses:

of er, neit her comes into nor goes out Som ethi ng as substance, that is matt in nihil o nihil (ex nal eter is ter mat ing, i.e., existence, from noth ing comes noth alth oug h depe nden t.

Int rod uct ion

ov er in ex ten sio n are nd s spo rre co wh eth er or no t spa ce d the secon are ts en ev all er eth wh wh eth er or no t wh ich debates the in no to the thi rd a first cause tha t has se or wh eth er the re is the lat er fou rth have an an tec ed en t cau own; an d the thi rd ev en ts in of an tec ed en t cause of its series the necessary. '4 it antiI1LOrny, wh ich s ke ma an external gro un d of is co nti ng en t or s u~i nt draws from Ka t tha ion lus nc co sts sugge However, u~e . He wh at he will lat er argue later the in ses the an tin om ies is no t the the be t en nd sce ut tha t the "tr an the un de rst an din g wi tho ep ts nc co ing us m fro se ult an tin om ies ) ari ne nt principles" res itions, while the "im ma space an d time as cond to space an d tim e and ts of the un de rst an din g ep nc co the from the str uc tur e of ou r rep nditions im po sed co ousi ng as "principles of the exp time - usi ng t" use of res en tat ion s of space t rej ect the "tr an sce nd en no es do he t Bu " es. nc ll seems to sit ion of appeara On the contrary, he sti the is a legitimate the concepts of dissertation, tha t the re as he did in the res tri cte d by un g an din concepts of the un de rst ee subrepthr tra ns ce nd en t use of the the s us he reiterate Th d an ce spa of the conditions Dialectic": tation as thr ee "Rules of tic axioms of the disser

h rules ances in acc ord anc e wit s no t bel on g to appear Do no t jud ge wh at doe e. and tim d wi th [rules of] space rit, to its of appearance, e.g., Go ter appearance, e.g., spi ou to g on bel t no s doe at wh t jec sub t no 2. Do rep res ent ed in conditions. can no t be conceived and at wh le sib pos im be 3. Do no t ho ld to te division. the infinite or of infini int uit ion , the tot ali ty of ity of reason" tha t can

iples of the absolute un these three Th en he lists fou r "princ we do no t violate any of as g lon as ed ain int ma ap pa ren tly be rules:
nk ing subject. a. Simplicity of the thi s. dit ion of rat ion al action ns in con the as b. Fre edo m tio n of one's representatio nec con all of tum stra sub the as s c. En ance a whole. of the wo rld in accord n tio tric res the e fus '4 ' d. Do no t con ten t wi th its lim ita tio n wit h its origin and con

the Critique (I) an if Ka nt envisioned for as ms see it n, the , int At this po paralleling that in tur e of space and tim e uc str d an e tur na the concepts of the ac co un t of un t of the use of a priori co ac w ne a , ion tat nt principles for the disser ich the y yield "im ma ne wh to ing ord acc g, din un de rst an applied to the conrst an din g" on ly wh en de un the of use l ica "exposition of the empir en tat ion to achieve an res rep l ora mp ote ati sp ditions of

Introduction appearances," but (3) continued adherence to view of tion that these concepts can also yield transcendent or m{;taph'vsical knowledge when freed of the restriction of the forms of seI1lsib ihty. Perhaps this last point was only a momentary lapse, however, the next preserved note Kant says that "The transcendent are principles of the subjective unity of cognition through reason, i.e. of the agreement of reason with itself"; "Objective principles are principl'=s a possible empirical use."'4 This suggests that exactly use of the transcendent principles of pure reason is, it is not to obtain any be achieved the knowledge of external objects, which can empirical use of the concepts of understanding, their to representations in space and time for the exposition of Kant continues with this thought in the following note, he out four conflicts between "principles of the exposition of appearances," or principles applied to "appearances" for the "unity of experience," on one hand, and "principles of rationality or comprehension" on the other. These conflicts correspond precisely to the four antinomies the Critique. The first set of principles is:


no absolute totality in composition, hence infinite /Jn1i!1',;SSU:s, no absolute totality of decomposition, hence nothing ab~;ohlte]ly simple, 3. no absolute totality of the series of generation, no unconditioned spontaneity, 4. no absolute necessity.

The opposing set of principles of rationality is: Unconditioned totality of the dependent whole, Unconditioned simple, 3. Unconditioned spontaneity of action, 4. Unconditioned necessary being.


Kant says that the latter "propositions are as principles of the use of reason in the whole of cognition: the whole of the manifold of cognition of the understanding. are regard to ..."'43 He does not finish the practically necessary thought, or explain the necessity of the of reason. But he is clearly drawing back from the that reason furnishes metaphysical cognition of real objects m<1q)erld{;nt own thought. 1777 Kant Summing up our results thus far, then, it looks as come this far in planning the Critique: it would include the account of space and time as transcendentally ideal pure forms of intuition already reached in 1770; (2) a derivation of three concepts of standing - substance, causation, and interaction - and

Introduction appear ances gIven princip les - as necessa ry for the exposit ion only in throug h the forms of space and time and as objectively y pitantinom rt a present ation in the form of a four-pa context, and against nces, appeara of ting those principles, valid for the exposit ion four oppose d transce ndent princip les, using the concep ts of underthe forms of sensibility which have restrict ion standin g but unino objective validity but can be used in an unspecified way for the pracified fication of empirical ~'1owledge and for some equally unspec tical purpose. Such a Critique would basically have consist ed of a theory au antinom y of pure reason. of sensibility, a theory of experience, n Clearly Kant needed more time to unders tand the positive functio the not is this But at in these notes. is of pure reason, the Critique that we can constru ct for the only way that the outline period around 1777 differs from the work as finally publish ed. There are several other glaring differences. First, the "transc endenta l theory unof experience," or theory of the "imma nent use" of the concep ts of the of ion deduct derstan ding, is not yet divided into a transce ndental exthe in categories and a derivat ion of the princip les of judgme nt used all , Second positio n of appearances, as it will be in t.,~e publish ed work. ed exhaust is of these notes suggest ti~at the conten t of the "Dialec tic" Critique ed the four antinom ies of pure reason, wherea s in the publish the Dialectic is divided into three parts, the "Paralo gism," "Antinomy," to and "Ideal of Pure Reason ." Can we learn anythin g about what led form the on these further divisions of the Critique before it finally took Kant gave it in 1779 and 178o? 1778-8 0. Fortuna tely, some notes assigned to the period 1776-78 rather than 1775-77 survive and throw light on the final develop ment d of Kant's concep tion of the Critique. In one note that has been assigne there that s to the later part of this period, Kant for the first time suggest may be a deducti on of the categories as necessary conditi ons of apperof consciousness that does not depend upon the ception or the eartempor al charact er of the data to be unified. Since this may be the sepas of ed liest surviving sketch of a transce ndental deducti on conceiv as arate from and antece dent to the argume nt to the categor ies what or nces, conditi ons of the possibility of the exposition of appeara Kant would come to call the "Analo gies of Experie nce," it is worth quoting this passage in full:
In everything passive or what is given, apprehension must not merely

be found, the indii.e., given, as it t represen to order in ated necessit be but it must also the vidual apprehension must be determined by the universal. The universal is from ished distingu being By state. relation to the others and to the whole of the the the arbitrary is it considered as given, and only by being subsumed under acin ted represen be e therefor must It ng. somethi as red conside it categories is the which by and nce cordance with a rule by which appearance becomes eX1Jerie 60

Introduction mind comprehends it as one of its actions of self-consciousness, within as in space and time, all data are to be encountered. The of the mind is the condition of thinking, and the subordination of every under the universal is the condition of the possibility of associating a given representation with others through an action. Even if the rule is not immediately obvious, nevertheless one must represent the object as amenable to a rule in order to conceive it as that which represents something, i.e., something which has a determinate position and function among the other determinations. ... '44 This note, which is very similar to a cmcial passage in the version of tile "Trans cenden tal Deduc tion" publish ed in the first edition of the Critique (A 108), is notable for two reasons. On the one hand, it clearly suggests that there must be mles for the unity of consciousness that can be charact erized In,c!elpeJlldently of specific rules for time-de termina tion, althoug h the way remains open for a further inferen ce that once the tempor al of consciousness is considered, then these general mles may rise to further mles which are themselves tempor al in content . separation betwee n the most general form of mles for the sciousness and the specific mles for the unity of a consciousness that is temporal in character, along with the necessity of relation between the two forms of rules, will be central to of the Critique ofPure Reason, where Kant will offer: a transce ndental deducti on of the pure concep ts of the unders tanding as conditi ons of the possibility of any unity of consciousness in general, under the of an "Analytic of Concep ts"; (2) a derivation from those general rules of more specific rules for time-de termina tion, under the of a "Schematism of the Pure Concep ts of the Unders tanding ,"' 45 is in turn part of (3) the "Analytic of Principles," in Kant argues for specific principles involving the tempor ally categories, as the principles of the conserv ation of substan ce and causation, as necessary conditi ons of objective time-de termina tion. 146 introduction of the concep t of schematism, which Kant first records in a note from 1778-7 9 with the statem ent that "We must subject all of our pure concep ts of the unders tanding to a schema, a way of nnf';Tn,Y the manifold togethe r in space and timc,"' 47 is require d precisely the explicit separat ion betwee n the transce ndental deducti on of the categories and the analogi es of experie nce (and related argume nts) means of which Kant had previously derived the categories. On the other hand, this note also reveals a fundam ental ambivalence about exactly how the categories are to be derived from the general idea of the "unity of consciousness," an ambigu ity continu ing one <llr'p<lrhr found in the materials from I 77 5. In one strategy, rules are necessary to distinguish an arbitra ry series of represe ntation s from the orderly or rule-governed series of represe ntation s by means of which a detenn i61

Introduction account, the is pres ente d to consciousness; on nate of consciousness characteristic of consciousness" wou ld mea n the Kan t suggests that rules are the pres enta tion of an object. -ccmsi;i01lsness, of consciousness as a form of set)~ the necessary they may r tatio ns, whateve the reco gnit ion that various repr esen le mind. sing a to g of belo ngin the or may not repr esen t, a means gest sug sepa rate these two strategies, nor Kan t does not mpt s atte t's Kan of ty will plague all for con nec ting them . Thi s amb igui runs It es. gori cate ded ucti on of the to find a definitive form for the s lead then and ," tion of the "De duc thro ugh out the first-edition version on, ucti ded the for pro per form Kan t to con tinu e to exp erim ent with he completely in the of ion not mer ely in the second edit in which he od, the inte rven ing peri rewrites the "De duc tion ," but in Future to in the Prolegomena trie s to resolve the amb igui ty Science of Foundations ili1t?talJbyslCS (1783),'48 the S as 1790 the 5 into on g draf ts,' 0 (1786),'49 and a num ber of survivin for ts draf his in the ded ucti on whe re he con tinu ed to tink er and niz Leib of e tapbysics from the Tim an essay on the Real Progress of ]I/Ie rpre tati on of the tran scen den tal deinte e Arriving at a definitiv lthe mos t difficult task for Kan t scho duc tion of the categories has bee n ing erly und this th cen tury , arsh ip thro ugh out the twe ntie large par t of the a is tegy stra its of tion ambivalence in Kant's con cep reas on for this prob lem . '5 fundamental stra tegy of the deKan t never resolved the issue of the con tent and struch else abo ut duc tion of the categories, but muc 177 8-7 9. Several had dea rly bee n reso lved ture of only arrived at not this peri od show that Kan t had extensive drafts also that he but ," end enta l Analytic the final organization of the "Tr ansc cen den tal ans "Tr atio n of the now arri ved at the final org aniz he had mes sche the plic ated Dialectic," whi ch is also mor e com s from note the 5-77 . Wh erea s in bee n con side ring in the peri od 177 le set sing a as tic" eria l of the "Dialec that peri od Kan t pres ente d the mat , parts n mai e thre the material into of antinomies, now he has divided by ed erat gen ion" scen den tal illus the diagnosis of "thr ee kinds of tran 3 Thu s, at this poi nt Kan t envie."'5 renc infe l "thr ee kinds of rati ona sioned the following argu men t. the boo k would consist of two main The constructive argu men t be bro ken into two furt her parts: parts. The first of these would in turn and time that had bee n in place since first would be the acc oun t of space lly enti tle this the "Transcendental 1770; in the Critique Kan two uld fina lmd er the title of "Transcendental Aes thet ic." The n in the seco nd, t, '54 Kan t would make the argumen Analytic" that he now introduces, a on niti cog ic can have synthet based on the principle that "W e consists of principles of the possibilif nce, abo ut objects of experie


Introduction of experience"'55 he now explicitly formulates, a transcendental deduction the categories tion of their role in empirical time-determination intervening schematism of those categories. This an:;UJnent, ShOVlTU1l[ that the categories must be applied to representations and time in order to yield unity of consciousness ence of objects, have the consequence nize only objects of the senses," thus that the categories "do not to the supersensible." It would then be the burden of the second main which Kant had already been referring to as a years, to show that "Even though the concepts not standing] extend to all objects of thought in general," "they any amplification of theoretical cognition," but may nevertheless is a "practical-dogmatic" role in a "practical regard, where condition of their use."'5 6 Now Kant divides critical of the work into three divisions. He argues that it is characteristic reason to assume as a or "postulate" the that "All conditioned cognition not only stands under conditions, but under one which is itself unconditioned," or that "If the is then the entire series of all its conditions is also given."!57 He now argues that because there are three kinds of rational inference, from a property to its subject, from a property to another property, and from a property to its ground, there must be three dialectical inferences back to an unconditioned or absolute substance, an unconditioned or abreason solute whole, and an unconditioned or absolute ground. unpostulates "the unconditioned subjective conditions of thinking, conditioned (objective) condition of appearances, and the unconditioned objective condition of all things in gener.d."I5 8 These thrce inferences, which Kant discuss in the Critique under "Paralogism," the "Antinomy," and the "Ideal of Reason," will diagnosed as theoretically unjustified, because the underlying pnnc!pl(~, that whenever the conditioned is given so is its IS theoretically unjustified. Nevertheless these three ideas of the uncondItioned will be useful in a practical context. Even in the Critique Kant will retain the argument that the forms of "transcendental arise from three forms of inference/ 59 but he also suggests both in these notes and in the publ1:she:d work that they arise directly from an unwarranted reification of the three concepts of a subject, a series, and a ground,'60 and it is easier to understand his diagnosis in these terms. Thus, the three tUlld:unental errors of metaphysics are the assumptions (1) that because we assign all of our thoughts to our selves as subjects, we have of the soul as an absolute subject; (2) that because we place all appearances in se63

Introduction ever dec reas ing spaces and times, ries of ever incr easi ng spaces ent of con ting ents necessarily dep end times, of causes and effects, and in ns nsio exte ed plet com wledge of upo n som ethi ng else, we have kno a of and e, caus first a of , e and time space and time , of simp1es in spac k thin t mus we use beca that ; and necessary gro und for all con ting ents knowledge of an abso lute e hav we ty, ibili poss any of som e gro und for possibilities. In Kant's words: gro und of t re[the idea that] the understanding mus The idea of the soul is grounded on only the as this me assu ns to the self and late all thoughts and inner perceptio permanent subject. in conditions in appearance is grounded The idea of the unconditioned for all unthe of ition completeness of all cogn reason as the prescription to seek the . ation rdin subo derstanding in [series of] is all objects of thought in an ens entiu m of y unit d ione ndit The idea of the unco 161 ... hip among all possible [things] necessary in order to seek the relations that ther e for us to form thes e ideas, it is Kan t suggests inte rpre t to take mis a it is do so, is eve n a subjective necessity to could that kind a of dge of objects them as offe ring theo reti cal knowle the senses. nev er be pres ente d ions conhis diagnosis of metaphysical illus Wn at led Kan t to l psyiona (rat s part e and God into these thre cern ing the self, the ly ious prev n whe ), logy rati ona l theo chology, rati ona l cosmology, and d thir and nd seco the of es ply instanc the claims abo ut the soul wer e sim ity plic sim of e anc inst an soul was just anti nom ies (the simplicity of the self just an instance of abso lute spon the of dom free in gen eral , and the the was ents ting con all y gro und of taneity), and an abso lute ly necessar con tent s of the thir d par t of the The 162 y? nom anti suh ject of the fou rth , the "Ideal of Pur e Reason," sug"Di alec tic" in the pub lish ed Critique ona l theo logy into a sepaKan t elevated the discussion of rati gest a too muc h mat eria l to trea t it as rate sect ion simply because he had coscal, logi onto the of s his crit ique single anti nom y - he reca pitu late argu men ts from the Only Possible ical olog -the sico mol ogic al, and phy ng his own positive argu men t from Ground of 1763 as well as criticizi argu men ts abo ut God that constithat work, even while reta inin g the ies in the Critique. Kan t wou ld also tute the thir d and fou rth anti nom positive acco unt of the necessary rahave bee n hard put to inte grat e his reas on ("Tr ansc end enta l Dialectic," tion al genesis of an ideal of pur e any discusSec tion 2; A 57 I -8 3/ B 599 -6 II) into Cha pter Boo k takes the form of an antinomy. sion , logy in the "Pa ralo gism ," however The criticism of rati ona l psy cho first the for 9 8-7 177 of s appears in thes e note is som ethi ng new, Kant's reure that the new "Pa ralo gism " is ject con can time. Her e one gor iescate the of on ucti den tal ded spo nse to his own new tran scen

Introduction because he has claimed that the unity ofconsciousness is an sity from which we can deduce the validity of the categories, now also has to tell us to be careful what not to infer from this unity of consciousness, namely any metaphysical claims about the soul, claims that the subject or bearer of consciousness is a unitary, simple, and eternal substance. Such a "paralogism of pure reason" be "a transcendental subreption," an illusion in "the which is subjective, would be taken for the of the as a thing." '6 3 We find no such warning in Kant before we find introduction of a separate transcendental deduction of the categories from the unity of consciousness; so we can assume the expansion the "Dialectic" to include paralogisms of pure reason separate from second and third antinomies was a cautionary response to the new deduction, Kant's own warning about what not to read into tion. Then once the structure of the "Dialectic" had so eXlpal1Cltod, it would not havc been unnatural for Kant to add a treatment about theoretically unjustified though morally useful conceptions of God as well. One last note, written on a matriculation record from March 1780, or a and thus either a last draft for the about to be written of this outmemo written during its composition, recapitulates line and then adds a reference to one final section of the Critique: To the Canon: the end of the whole of metaphysics is God and the future and the end of these [in] our conduct, not as though morality must be arranged in accordance with these, but because without these morality would be without consequences. 164 This is cryptic, and can only be understood in light of the argument that Kant develops, over all three Critiques, that the highest good conor maximization of both virtue and happiness, which we can ceive of as being made possible by an intelligent and beJlev'oItont of the world prepared to give us the time necessary to perfect our virtue and to make the world suitable for the achievement of our ends, is not the motivation for virtuous action but is presupposed its raiJOJlal1ty. This is the practical use to which Kant will put the theoretical illusions of metaphysics. Conceiving of a "canon" of pure reason as as its critique - that is, a doctrine of its positive practical use as well as negative criticism of its misguided theoretical use - was the final stage in conceiving of the structure and content of the Critique, where this "canon" would be expanded into a "Doctrine of Method" would accompany the "Doctrine of Elements," into the "Transcendental Aesthetic," "Transcendental Analytic," and "Transcendental Dialectic" would be placed. INith all of this in place by 1779 or 178o, Kant was to

Introduction 1781, afte r a to ann oun ce to Her z on I writ e the ter eme nts, that "In the cur ren t Eas dec ade of apo logi es and pos tpon Pure k of min e, enti tled Critique boo k fair ther e will app ear a boo ,6 Ten days te he wro to Her z thes e lines: Reason. " 5 but bring about a complete change cannot work, may it stand or e ledge [metaphysics], a part of knowledg of thinking in this part of human know mite part I have nowhere sought to crea that concerns us so earnestly. For my have ts in order to patch up my system; I men rages or to advance specious argu ld wou that ht insig hed finis a I might get to rather let years pass by, in order that find now I that so ed; arriv fact in I have satisfy me completely and at which of theory (something I could never say main the in ge chan to t wan I nothing ficlari and s tion here and there little addi any of my previous writings), though 166 . rable cations would be desi

Reason was nev er inte nde d to be For Kan t himself, the and systematic met aph ysic s of natu re mor e than a prop aed euti c to the n ntio nde d to writ e, and his own inte of mor als that he had long inte ctly mus t have bee n to pro cee d dire upo n the com plet ion of the Critique stan tial oso phic al system. He mad e sub to thes e two part s of his phil of ing the Metaphysical Foundations pro gres s in this dire ctio n, pub lish tion lica trie d to sho w that the app Natural Science in 1786, in whi ch he to the empirical con cep t of mot of his gen eral prin cipl es of judg men N ewt onia n physics, and the Groundtion yields the basic prin cipl es of n 1785, inte nde d to be the intr odu ctio work ofthe lWetaphysics ofNIorals in s ysic aph wou ld con stitu te the met to the deta iled system of duti es that real the initi not in fact app ear unti l 1797). But of mor als (and Kan t's expecd inte ppo disa ly sore on cep tion of the Critique ofPure Reas ge lp but brin g abo ut a com plet e chan tati on that the wor k cou ld not "he the of Kant's effo rt dur ing the dec ade of thin king ," and a grea t deal of en task of clarifying the critical foun 1780s was dev oted to the unfo rese ed plet that he tho ugh t he had com dati ons of his syst em of phil oso phy of diff eren t forms: the publicaber in Ma y 1781. Thi s wor k took a num in ed pop ular izat ion of the Critique tion of a brie f defense and atte mpt the on k re lvletaphysics; con tinu ed wor 1783, the Prolegomena to any Futu propriv ate note s dur ing 178 3-84 ; a tran scen den tal ded ucti on in his the in es tal ded ucti on of the cate gori pos ed revision of the tran scen den a ce; Scien l sical Foundations of Natura intr odu ctio n to the 178 6 lvletaphy ion edit nd ofPure Reason for its seco sub stan tial revision of the Critique of two furt her critiques, the Critique in 1787; and finally the pub lica tion Critique ofJudgment (1790), which of Practical Reason (1788) and the tion plan ned at the time of the publica wer e clearly not wor ks Kan t had

Introduction of the Critique ofPure Reason which instead grew out struggle to clarify the foundations of his critical phJllm;ophy. comment on all this material hcrc; instead, after some in the on the revisions to the Critique of Pure Reason that are Prolegomena and Metapbysical Foundations Science, we will conclude this introduction by outlining the main changes in second edition of the first Critique. After a year of silence, broken only two tri,eniily '.L' __ ._ reviews published in Frankfurt and Greifswald, the ceived its first serious review in the first supp!ero.ell1t2lry Gijttingiscben Anzeigen von gelehrten Sachen for 1782. Gottingen, had been founded in 1737 by George I of .GlJgl"mu in his continuing capacity as Georg August, elector of t-hmClVer, home to a group of empiricist philosophers J. G. H. (1740-1820). The review, apparently abridged from a much longer and more sympathetic draft the Berlin moral philosopher Christian Garve (1742-1798), was dismissive. r67 The version of the review published by Feder omitted Garve's tion of much of Kant's arguments and his quite IllSIgtlthll irlterplretation of Kant's justification of the possibility of synthetic a priori III general, and in mathematics in particular, to focus on three ob]ectlons. First, it charged that Kant's "system of the higher or ... transcendental reducidealism" was nothing but a restatement of Berkeley's ing all objects to our own sensations and leaving the real existence of any objects beyond our own representations unknown. Second, it argued tlIat on Kant's account there could be no "between the actual and the imagined, the merely possible," belJWc~en actual and "mere visions and fantasies." Third, it charged that Kant's argument that the unsound theoretical use of pure reason can and must be replaced by a sound practical use was entirely unnecessary, since morality already has a sound foundation in common sense. Kant had apparently already formulated the intention to write a shorter and more popular presentation of his critical philosophy almost as soon as the Critique was published, but the hostile review galvanized him, and he included explicit answers to some of its charges in the pages of the Prolegomena to any Future Metaphysics that he lished in August 1783. Specifically, he differentiated his PO:SltlOn Berkeleian idealism by arguing that he denied existence of space and time and the spatiotemporal properties of objects, but not the real existence of objects themselves distinct from our representations, and for this reason he proposed renaming his transcendental idealism with the more informative name of "formal" or "critical idealism," making it clear that his idealism concerned the form but not the existence of external objects. r68 Further, he argued that his theory of the undel:st~ll1dlf ' __


Introdu ction

idealism, offered deusual ing and its principles, of veridical experice coheren b.~e hing termina te principles for establis s, '6 9 and that for fantasie and dreams rent ence as contras ted to incohe , an this reason it should not be considered a form of "higher " idealism ness, fanciful of tion implica ive pejorat a d detecte expression in which he but rather a philoso phy firmly rooted in the "fruitful bathos of experiy's ence."'7 Finally, Kant rejected any compar ison of his view to Berke1e space of dge knowle all leaves ism empiric on the ground that Berke1ey's and time a posteriori and conting ent, whereas only Kant's own formal knowledge of space and time as the uniidealism can explain our n. I71 intuitio necessary forms versal ndental idealism can explain our transce Empha sizing that only his pure physics while at the same and atics a priori knowledge of mathem it is entirely compat ible idealism formal as time demon strating that objecbe objects l with the real existence of externa Vindi. edition second its for Critique tives in Kant's revisions of be cating his view that the illusory theoret ical use of pure reason must the ged challen point last the use, replace d by its sound practic al Gottin gen review althoug h not replied to in the Prolegomena, would also be an aim of those revisions. But, as had been the case before with d in the critical response to the inaugural dissertation, Kant also reveale cona , namely raised: not had critic his the Prolegomena a concer n that cern about the adequacy of the transce ndental deducti on of the catethe gories itself. Kant expressed this worry about the deducti on (and comis he that says he could: he as mildly as associated paralogisms) but pletely satisfied with the "conten t, order, and doctrin e" of his work of parts some in ation present the with d that he is "not entirely satisfie L~e or ies categor the of ion deduct the e.g., the Doctrin e of Elemen ts, be paralogisms of pure reason."'7 In fact, both of those chapters would reto part in , Critique the of edition second comple tely rewritt en in the spond to the challen ge to Kant's variety of idealism raised by about s concern own Kant's to d respon to Gottin gen review but also their persuasiveness. Indeed , Kant had already begun to manife st his concer n about the he adequacy of the deducti on in the Prolegomena itself. Follow ing what the than rather mena Prolego the of method ic" claims to be the "analyt be "synthe tic" method of the Critique '73 - the difference is supposed to uted undisp of ns positio presup the s analyze betwee n a method that knowledge-claims and one that determ ines the consequences of fundathe mental claims about the human cognitive faculties,'74 but in fact s concern works two the in nt argume Kant's n major difference betwee Kant d analyze are ons conditi whose is it ims which knowle dge-cla s replaces the transce ndental deducti on of the categories, which purport transcen the of lity possibi the of ons conditi ry to analyze the necessa



dental unity of apperception, with an analysis of the necessary COlldltlon life universally and necessarily valid judgments in ence that makes no use of the concept of apperception at alL Kant argues that while mere "judgments of perception," which make no claim to necessary objective validity or the agreement of others at but only report how things seem to a single subject, use the logical forms of judgment, "judgments of experience," which do to objective validity necessary for can only derive universal and necessary validity from their use of a priori categories to make the otherwise indeterminate use of the forms of judgment determinate. '75 This approach is pursued even further in the Preface to the M,eta,phi'slc,a! Foundations afNatural Science three years later, where Kant suggests the categories can be derived as the necessary conditions of making the use of the logical forms of judgment determinate even explicit reference to the alleged distinction between judgments of perception and of experience. 176 But although this strategy avoids the obscurity of some of Kant's claims about the transcendental unity of apperception, empiricists it is open to the charge of begging the question against and skeptics, proving that the categories are necessary only byaccepting an interpretation of ordinary scientific knowledge-claims as universally and necessarily true that neither a skeptic nor an enlplnclst would dream of accepting. In any case, Kant's notes from the period 1783-84 show that he continued to experiment with both the unity of apperception as well as the concept of objectively valid judgment as possible bases for the tion of the categories. '77 However, when Kant came to rewrite the chapter on the transcendental deduction for the second edition, he returned to his original strategy of trying to combine the conditions of possibility of the unity of apperception with those of judgment of objects to create an unshakable foundation for the objective "" i!irlinr of the categories. 1786 that a lNhen Kant was first notified by his publisher in new edition of the Critique would be needed, he apparently contemplated a drastic revision that would include an extensive discussion of practical reason as well as a restatement of his work on theoretical reason. At the same time, he also assumed the rectorship of his university. At some point during the year he must have decided on the more modest iliough still extensive revisions iliat we have, enough of were completed by January 1787 for typesetting of the new edition to begin, and all of which were apparently completed by tllat April, just a year after ilie new edition was first requested. '78 (At some point bet:we:en 1781 and 1787 Kant made the annotations in his own copy of the edition of ilie Critique that we reproduce throughout our tramslatlOll, but as iliese notes are not closely matched by ilie changes in the 69

Introduction these note s were mad e dur eve of 1787, ther e is no reas on to beli new edition.) on the ing 178 6-8 7 as grow ing part ly out of Kant's ion, The mai n changes in the second edit and part ly out of his own concerns, response to the criticism of the first Kan t replaced the preface as we have just described, are as the mos t general term s abo ut speaks in to the first edition, a aphysics on a secure footing, the nee d to place the science of met inbot h the s in muc h mor e considerably long er one that describe odu ces the intr t Kan that here is it novations of Kant's critical met hod anth rop oce ntri c pro ced ure in famous com pari son betw een his own revo luti on in astr ono my and Cop erni cus 's heli oce ntri c has a positive role pur e reas on (Bxvi) - and his pos itio n reti cal use (Bxx1v-xxxvii). The latonly in its practical rath er than theo ond to the dismissive remarks of ter emphasis is clearly mea nt to resp ect. '79 The new pref ace concludes the Got ting en review on this subj ges in the new edition, and then with with a brie f com men t on the chan yet furt her the new "Re futa tion of a long foot note (BJoox-xli) revising mos t imp orta nt of thos e changes. Idealism" that is one of y expanded. Its mai n changes are, (2) The intr odu ctio n is considerabl the dist inct ion betw een a priori and first, a mor e detailed discussion of then an exthe first edit ion had included, and a posteriori cog niti on e mathepur of ns nitio cog ri ic a prio tend ed argu men t that the synthet idealism, tal den scen tran his be explained matics and physics can wit hou t cha nge from the Prolegowhi ch are in fact lifte d virt uall y es shows that he is still very conmena. ,80 Kant's inclusion of these pag bet\veen Berkeley's idealism and his cern ed to emphasize the difference one of to explain a priori knowledge was own, since Berkeley's ena. Kant's chie f charges in the Prolegom ic" is also considerably expanded. thet Aes tal den cen ans "Tr (3) The to buttress the have bee n Kant's aim in its revision seems to necessity of his tran scen den tal ide(anti-Berkeleian) argu men t for the t cog niti on, rath er than the argumen alism to explain syn thet ic a priori exisble with knowledge of the real that his form of idealism is compati s dom inat e his revisions in later part tenc e of external objects, which will of ons previously undivided discussi of the work. Thu s, Kan t divides his nscalls the "Metaphysical" and "Tra space and time into wha t he now s ume re the first of these titles subs cen den tal Exposition" of each, whe intu are pur e and a priori forms of the argu men ts that space and time sepa thei r own right, and the second itio n as well as pur e intu itio ns in of n our synthetic a priori cognitio rate ly expounds u~e argu men t that can only be explained by transcen mat hem atic s (especially geometry) with es of the "Aesthetic" conclud den tal idealism. The revised version beh alf of tran scen den tal idealism in a num ber of add itio nal argu men ts ion. that wer e not pres ent in the first edit

Introduction (4) next major change comes in the of the categories, Kant rewrote almost COll1pletely edition introductory sections arc left largely but the rest is completely rewritten, thirty-five pages in the first edition replaced with forty completely new ones in the second). To characteran ize the nature of the changes that Kant made in any interpretative venture inappropriate for a points can be noted. First, in spite of his experiments with an apperception-free deduction in 1783 and 1786, Kant in fact tried to gremUlQ the entire deduction more clearly on the starting-point of the he in 178 I. At the same trying to apperception his experiments of the intervening years, he also tried to connect the unity of apperception more unequivocally with the idea of the obJec:tne validity of judgment than he had in the earlier version. Second, Kant tried to prepare the way for the coming new "Refutation by stressing that the cognitive subject must be regarded as determJ'ni:ng as it the structure and order of its own self-consciousness just as does to the representation of external objects ( 23-5). continuing the stress on the necessity of the representation of space that was part of the Prolegomena's response to the charge of idealism, Kant stresses that the synthetic unity of consciousness, in the first edition had been associated exclusively with the of time, is responsible for the of both space and time, and indeed that the representation of determinate spatial relations is a necessary concliis an tion for the representation of a determinate temporal undeniable feature of any conceivable self-consciousness (see B 156). (5) The argument that while time is the form of all sense, the representation of space is itself the necessary condition for the rej)reSelltation of determinate order in time, which continues Kanes of the charge of Berkeleian idealism, is the chief theme of all of the revisions in the "Analytic of Principles." These revisions take the form restatements of the several principles of judgment, and of additional paragraphs at the start of each of the proofs; but Kant's most lIT1nn,rr<,nr addition to this part of the book is the new "Refutation of Idealism" that is inserted into the discussion of actuality in the "Postulates of Empirical Thought" (B 274-9). This may seem like an m~mSplClOl!S cation for such an addition, Kant's intention in choosing it can have been to show that empirically meaningful judgments the modalities of possibility and necessity all depend upon connection to the actual in perception, and then to show what he means the actual in perception: that which we judge to exist independently of our representation of it, even if we also know that the form in which we represent the independence of such objects is itself dependent upon the constitu" in other tion of our own sensibility. The "Refutation of

Introduction forIS his prov e words, is Kant's ulti mat e atte mpt to the subjective realism Berkeley. mal idealism rath er of the mos t imp orta nt of Ran t's The "Re futa tion of Idealism" is one the fact that before the new edit ion additions to the seco nd edition, but revision in the new he was already revi sing was even t was to be rew ritte n) shows that Kan preface (presumably the last par t " is tion men t. In fact, the new "Re futa hard ly satisfied with his new argu ght thou long -con side red process of not so muc h the culm inat ion of a a doz en or mor e furt her versions and as the beg inni ng of a new one , k to show that Kan t con tinu ed to wor from the peri od 178 8-9 survive alhad ique nd edit ion of the Crit on this argu men t even afte r the seco 181 read y appeared. sions in the cha pter on the dis(6) Kan t also und erto ok major revi in nou men a. His prim ary con cern tinc tion betw een phe nom ena and t cep con erence betw een usin g the thes e revisions was to clarify the diff rded rega positive sense. Thi s can be of a nou men on in a negative and a trin e that whereas pur e reas on has as a step tow ard clarifying his doc a does have a positive practical use, only a negative theo reti cal use it the had bee n cha llen ged bot h doc trin e the clar ity of whi ch Garve's original draft. Got ting en review and ofId eali sm, " Kan t had no choice (7) Hav ing added a new "Re futa tion ch paralogism of the first edition, whi but to rewrite at least the four th of rge cha n just ific atio n to the abo ve all oth er passages had give of cts obje could be as certain of the Berkeleianism by insisting that we noth are e because objects in space oute r sense as of thos e of inne r sens r inne of n alongside repr esen tatio ns ing but one species of repr esen tatio com a replaced this argu men t with sense (see especially A370)' Kan t puzno be men t that ther e sho uld pletely different, anti -Ca rtes ian argu betw een min d and bod y because ion zle abo ut the possibility of inte ract s s that Des cart es and his follower the differences in thei r appearance than e mor inte ract ion mig ht be no had assumed to stan d in the way of of und erly ing reality (B427-8). sort diff eren t appearances of a single self to this change, but took the opHowever, Kan t did not confine him whole cha pter on the paralogisms. por tuni ty to rew rite and simplify the in the fou rth paralogism, this is the Exc ept for his substantive change s up to his pret ense of mer ely imonly par t of his revisions that live (Bxxxvii). prov ing his man ner of exposition son," Kan t mad e no furt her sigBeyond the "Paralogisms of Pur e Rea ion. We do not kno w if this means nificant changes for the second edit , tent with the rem aind er of the book that he rem aine d com plet ely con teresta ing patience. His con tinu or only that he ran out of time and thir d Critiques of man y imporand nd men t and refi nem ent in the seco rem aind er .of the book, such as his tant doctrines touc hed upo n in the

Introduction theory of the postulates of practical reason and the regulative use ideas of reason, sugges t the latter rather the In sum, the bulk of Kant's changes in the second grew out of his desire to refine and defend his transce ndental idealism showing that only it could explain our a priori knowledge at same time arguing that it was comple tely compatible existence of external objects. Beyond this, Kant wanted to emphasize positive role of reason in the practical rather than theoretical and he continu ed to try to find a clear and deducti on of the categories. These concerns led him to revise his introdu ction, the "Transc endenta l Aesthetic," and the chapter on phenOlllerla and noumen a, as well as to revise completely his preface, "Transcendental Deduct ion" of the categories, and the "Paralogisms of Pure Reason."
Note on translation

This is an entirely new transla tion of the Critique ofPure Reason. Our intention in produc ing this transla tion has been to to give the of the translation an experience as close as possible to of the reader of the Germa n original. The criterio n for success in this intentio n is that as much interpr etative work be left for the reader of the transJ'atlon as is left for the reader of the original. This intentio n has dictated a number of our choices. Obviously it has require d as much consist ency as possible in translation ofKant 's terminology; to the extent possible, we have used the same English word for any philosophically significant Germa n word, and where a single English word has had to stand for several German words, we have noted this fact. This anses when Kant uses both a german ic and a latinate word that would be translated into English by the same word, e.g., "Gegenstand" and "Object," both of which are translated into English as "object." In some such cases it may be a matter of interpr etation whethe r Kant means precisely the same thing or not, so we have preserv ed the inform ation about his usage by markin g the Latinat e membe r of the pair in the notes, but have not impose d any interpr etation of the distinction in the text. Other obvious consequences of our underly ing intentio n include the preservation of Kant's sentences as wholes, even where considerations of readability might have suggested breakin g them up, the preservation of ambiguous and obscure constructions in Kant's original text wherever possible. The latter decision means that we have refraine d from accepting emenda tions to the Germa n text as long as we believe


Introduction men ded original, even if a prop osed som e sense can be mad e the une of a given passage. In thos e cases eme nda tion makes easier sense out ier we have not cited auth orit ies earl whe re we do accept eme nda tion s, ion the Critique in the Aka dem ie edit than Ben no Erd man n's edit ion of mea ns that we have not repr odu ced (19 II), cite d as "Er dma nn." Thi s g bac k to nine teen th-c entu ry edithe ascriptions of eme nda tion s goin RaY1mmd Sch mid t (1926, decorate the pages of the edit ion tors man Kem p Smith's Eng lish transla1930), wllich was the basis for Nor mea ns that whe re Kant's location tion (1929, 1933). Our decision also whi ch he always trea ts as a Lat in borof the adverbial phrase "a priori," us latin ate Ger man term , is ambiguo row ing rath er than a an adjectival modificaof a verb betw een an adverbial modification e it ambiguous, alth oug h we could tion of a nou n, we have trie d to leav not always do so. however, was how to pres ent the The bigg est issue that we faced, and second (1787 or "B") edior variations betw een the first (1781 eventually dicHer e too our und erly ing inte ntio n tions of the Erd man n or er eith that ado pted tate d a diff eren t app roac h from two separate as B and A n trea ted Sch mid t and Kem p Smith. Erd man its entirety, in B ion edit the works, pub lish ing in Volume 3 of the point ugh thro up A of 4 followed by a separate edit ion in s of Pure gism ralo "Pa (the s revision after whi ch Kan t mad e no furt her compare to er read the for t makes it difficul Reason"). Thi s rded rega also th Smi p Kem and mid t part icul ar passages in A and B. Sch d a single text that always follows B as the definitive text, but pres ente or the page and relegates modified the text of B on the mai n part of r note s, exc ept whe re Kan t rewrote dele ted passages from A to thei verin thei r entirety, in whi ch case the cha pter s or sections of the B. from ion vers the wed follo text sion from A was pres ente d in the es read er to follow the text, and mak Thi s ofte n makes it difficult for the er to get a clear sense of how the first it particularly difficult for the read prob lem , we have pres ente d both edit ion read. In ord er to avoid this k that Kan t rew rote extensively as versions of thos e sections of the boo , ent two versions of the introduction well as completely: thus, we pres ben the cha pter on the "Di stin ctio the "Tr ansc end enta l Aesthetic," and as well as two versions of the pref twe en Phe nom ena and Nou men a" to r " and "Paralogisms." But in orde ace, "Tr ansc end enta l e editions easier than Erd man n mad make com pari son betw een the two editions for all passages the pag inat ion it, we have also the first edition, even from ct Kan t preserved inta ct or largely inta ly extensively alth oug h not complete in thos e cha pter s that he rew rote footmad e in our have note d the changes that Kan t for the second, ion, we have folsect a in s nge cha min or notes. 'h ere Kan t mad e pref erri ng B in our and Kem p Smi th low ed the practice of

Introduction main text noting divergences in A in our footnotes way, we to inserted in B is enclosed in angled brackets). In make it easy for the reader to remain dearly aware of the differences between the two editions without treating them as if were two unrelated works, as Erdmann's approach does. Our view that we should avoid imposing our own ml:erpnotat10 n the Critique as as possible has not meant that we them in the interreferring our readers to materials that might pretation of the text. Instead, we have provided two sorts of references that may help in the interpretation of the text. The first sort of material is Kant's notes in his own copy of the first the which were published by Benno in 1881 \~'.HLJIU l::rdJmann; Nachtrdge zu Kants Kritik der reinen [Kid: Lipsius & r881]). These notes range from mere cross-outs to changes in words or phrases to extensive comments or paragraphs. ~cJ1mJdt Smith noted those places where Kant had ch;ln~~'ed but omitted all the rest. We have presented of Erdmann recorded in our footnotes, following Erdmann's deSO"lptlOn of the location of the notes as closely as possible. In this way, the reader can have the experience not merely of reading Kant's own Kant's original text of the first edition but that copy of that edition. (No annotated copy of the second edition has ever been known to exist.) These notes are cited thus: "E" Erdmann), followed by Erdmann's roman numeral and the page numtler in his edition; then the volume and page number of their appearance in theAkademie edition. Second, we have provided cross-references to remany of Kant's notes in the Handschriftliche Nachlafl mains") transcribed in volumes through 23 of the Akademie edition. Obviously we eould not index of these notes, tried to give references to those that throw light on specific passages in the Critique, especially those that seem to be either or subsequent reworkings of specific passages. Since does not appear in the original editions of the Critique or Kant's own copy of the first edition, we have not referred to it, let alone rej)f(ldllce:d in our footnotes on Kant's pages, but have put the references to it in our endnotes. Our translation has not been produced from any single German edition. As do most contemporary scholars, we began working from the edition of the Critique by Raymund Schmidt in the Philosophische Bibliothek. As we worked on the translation, however, we realized Schmidt's edition is the least conservative twentieth-century Kant's text, not only modernizing spelling and punctuation more others do but also accepting the largest number of editorial emendations to the text. We thus began to check our translation against the


Introduction text, namely those of three other main twentie th-cent ury editions Weisch edel in the Inse1 of Erdma nn in the Verlag Studienausgabe (1956), and of Verlag, subseq uently Ingebo rg Heidem ann in the Reclam Studienausgabe (r966). Of these, the Heidem ann edition appears closest to the origina l editions, though tion it does modern ize spelling. Finally, we have checke d the transla against facsimiles of the origina l editions. Here we can add a word about our choice of typography. The origiinnal edition s were set primari ly h"'l Fraktur (gothic type). Latin words, well as ri" posterio "a and cluding such frequen tly used words as "a not regard as natural Kant which " and as into Germa n, were set in roman type. Empha sis was indicat ed, not n by the modem English method of italics nor by the modem Germa thicker and method of Sperrdruck (spaced type), but by the use of larger try to Fraktur type than is used elsewhere (boldface or Fettdruck). To bold used rc recreat e the appeara nce of Kant's pages, we have u~erefo had type for emphas is and italics for the foreign words that Kant was printed in roman type. In the original, a range of Fettdruck sizes t difficul used, which makes it sometim es quite easy and sometim es very disof to tell whethe r a word is being emphas ized - this is a source which words should be emphaagreem ent in modem edition s this range of type sizes. We uce reprod to tried not have sized. We to should also note that Kant sometim es but not always uses Fettdruck used. than indicat e that a word or sentenc e is being mentio ned rather type; where he does not, we have introWhere he does so, we use duced quotati on marks. Now for a word about our use of previous English translations. We of have followed Kemp Smith in many of his choices for transla tion Smith Kemp Kant's technic al terminology, for the simple reason that not always) adopte d the wise proced ure of letting Kant's usually the own Latin equivalents for his Germa n technic al terms determ ine phrase of doubt many of Kemp Smith's turns English translat ion. minds after years of using his translat ion.) our in rated reverbe also Nevert heless, the presen t work is by no means a revisio n of Kemp Smith, and it departs from his transla tion systematically and consistently throug hout on many points. We have always worked directly also from Germa n texts, consult ing Kemp Smith from time to time but found we consult ing the earlier English translat ions as well. Of these, D. that by Friedri chMax Miiller (r88r) more helpful than that by J. M. the only of Meikel john (r855)' Of surpris ing help was a full translat ion This second edition done by Frands Haywo od (second edition , r848). to able been have is the earliest English transla tion of the Critique we clearly od discover, and often proved helpful because, like lJ.S, Haywo made literalness in transla tion his primar y objective.


In this bibliography, we list the German editions of the text of the Critique ofPure Reason that we have consulted, the earlier English translations we have consulted, and a selection of scholarly works mc:lu(lmg discussion of the genesis and text of the Critique. The last selection is not intended as even a selective guide to philosophically and useful works on the interpretation of the Critique.
German texts



Critik der reincn Riga: Johann Friedrich Hartknoch, 1781. Facsimile edition: London: Routledge/Thoemmes Press, 1994 KANT, IMMANUEL. Critik der reincn Zweyte hin und wieder verbesserte Auflage. Riga: Johann Friedrich Hartknoch, 1787. Facsimile edition: London: Routledge/Thoemmes Press, 1994KANT, IMMANUEL. Kritik der reinC17 Zweite Auflage 1787. Benno Erdmann.] Kant's Schrit'ten, herausgegeben von der K6niglich Preu6ischen der Wissenschaften, Band HI. Berlin: Georg Reimer, 1911. RANT, IMMANUEL. Kritik der reinen VenzzmJt (1. Benno Schriften, herausgegeben von der K(j'ni~~!i(:h Erdmann.] Kant's PreuJ3ischen der Wissenschaften, Band IV. Berlin: Georg Reimer, 1911. Pp. 1-252. KANT, IMMANUEL. Kritik der reinen Nach der ersten und z7veiten Original-Ausgabe. Herausgegeben von Raymund Schmidt. Dritte Auflage, mit einer Bibliographie von Heiner Klemme. Hamburg: Felix Meiner Verlag, 1990. (First and second editions, 1926, 1930.) RANT, TMMANUEL. Kritik der reinen HerausgegebcnvonWilhelm Weischedel. Werkausgabe, Bande Iv. Frankfurt am Main: Suhrlzanl1p, 1974- (Originally Wiesbaden: lnsel Verlag, 1956.) KANT, IMMANUEL. Kritik der reinen Herausgegeben von lngeborg Heidemann. Stuttgart: Philipp ReclamJun., 1966. English translations



Critick of Pure Reason. Second edition with notes and explanation of terms. Translated by Francis Haywood. London: William Pickering, 1848. KANT, IMMANUEL. Critique ofPure Reason. Translated by J. M. D. Nl,elkelJ'DhJl, introduction A. D. Lindsay. London: J. M. Dent, 1934. (Original edition: 1855.) KANT, IMMANUEL. Critique of Pure Reason. A revised and translation based on Meikeljolm. Edited by Vasilis Politis. London: J. M. Dent, 1993. RANT, IMMANUEL. Critique of Pure Reason. Translated by EMaxMuller.


Intr odu ctio n don : M:lcili1illan, '-JI1F ;1""" edition: Lon Gar den City: Anc hor 1881.) Nor man Kem p Smith. Reason. Translated Ir117nanuel !<:ant's Lon don : Macrni!1arl, r933. (First ediSecond impression with corrections.

3. Selected sernn,"iar"\1 sources


in the Antinomies. Oxford: The Oxf ord University Press, I972. and Kant's Transcendental Idealism: An ALL ISON , HEN RY E. 1983. New Haven: Yale University Press, An AME RIK S, KAR L. Kant's 1982. Reason. Oxford: Clar endo n Press, der Kritik der und die re dusse "Die L. EMI ARN OLD T, Berlin: d Gesammmelte Schriften, Ban IV " In reinen II9- 225 . Bru no Cassirer, 1908. olis: Bobbsin the Pb.ilo5"OfJl7y of Kant. Indianap ies Stud ITE. WH

196 5. ors. Cambridge, Mass.: Harvard an Philol:of).h,,: !<:ant and his Predecess Germ y Earl University Press, 1969' !<:ant to sophy c. The Fate of Reason: German Philo BEIS ER, FRE DER ICK . 1987 s, Pres y ersit Univ Fichte. Cambridge, Mass.: Har vard Cambridge: Cam brid ge University t's !<:an F. AN ATH JON T, BEN NET y Press, 1974. Cambridge: Cam brid ge Universit Outline of One Central An e: ledg of Know t's BIR D, GRA HAM . Kan an Paul, Reason. Lon don : Rou tled ge & Keg Ar{J1!i'l7f11t in the Critique
The BRA NDT , REIN HAR D. d slate Tran . B92 -IOI Critique of Pure Reason A 67-76; Table th Ame rica n Kan t Society Nor . kins Eric Wat Atascadero, Cal.: Ridgeview, 1995 Studies in Philosophy, Volume 4. ge brid Cam ge: brid Cam Introduction. Ed. C. Lewy. BRO AD, c. D. Kant: An University Press, 1978. n del" !<:ant: Die Ent"Wiiife zu einer Deduktio CAR L, WOL FGA NG. Del' . 1989 t, denh oeck & Rup rech KIlJ'rmwipnvor 1781. Got ting en: Van 2nd. ed. Berlin: F. Dummler, der rie Theo ts COH EN, HER MAN N. !<:an 1885. in iVIetaphysics. London: !<:ant's Solution for DRY ER, DOU GLA S P. Allen & Unw in, 1966. ten Auflage del" t's Kriticismus in del' ente n und zwei ERD MAN N, BEN NO. !<:an zig: Leopold Leip Eine historische Kritik del' reincn Voss, 1878. Kiel: Lipsius & Tischer, 188r. Nachtrdge zu !<:ants Kritik del' reinm l1 des Textes von Kants Krit ik der rei17C Beitriige zur Geschichte und Re7!ision . V,'1"1711nft Berlin, 1900 on the TransCfndcnt's Intuitionism: A !<:an NE. LOR IN, STE FAL KEN Tor onto Press, 1995. tal Aesthetic. Toro nto: University of

Press, Kant's

1962 .


Introdu ction

University Press, 1992. University Press, 1987.


Kant and the Exact Science,l. C2Im1bridge,Mass.: Harvard


Kant and the Claims of Knowledge. Cambri dge; Der

1775. Tiibing en, 19 10 .


und Kants Kriti':.is771l177ls um

Mohr (Paul Siebeck), 1960.


Der 07,ttO!'OfD;scl'e Gottesbe7ueis. 2nd. ed.

tib1rlgen: J. Richard

c. B.

of Reason: Essays on Kant's Philosophy. Edited Cambridge, Mass.; Harvard Univers ity Press, 1994.

Kants zur Tr,an,rzendem'a0'Jhilosopl'Jie: Kant. Stuttgar t: Kohlham mer, 1970. HOLZHE Y, HELMU T. Kants Eifah17mgsbegriff: (j7,letl:enQ:es'cb;icbt1i(:be und bedeut1Ili'j!,Sa77tli'yt"c!Je Un:terSUc.hu77gcn. Basel: Schwabe, 1970. KlTCHE R, PATRIC IA. Kant's New York: Oxford UniHINSKE , NORBER T.

versity Press, 1990.


Kants Philosophie des c)7!.DjCI'i:iS: Sy:rte?1uJti\i:he und ennvicklungs geschichtliche Untersuc!nmgcn zum Verhiiltnis von und Selbstcrkeml tn is. Kant-Fo rschung en, Band 7. Felix

199 6 .

In'CW<;;l ,


der reinen

Materialen zu Kants Kritik

Frankfu rt a.M.: Suhrkamp, 1975.

Kant - Der Durchbruch von 1769. Koln: Rezen.lio77C7i zur Kantischen PhiloSGlphi'e. Band I: de



1]81-17 87. Bebra: Albert Landau, 1990. LONG UENE ss E, B EATRI CE. Kant et la Universitaires de France, 1993.

Paris: Presses

cago Press, 1973.


Kant'sAnalogiesofExpcrience. Chicago: Urllv,ersitvofChi-

Space, Time and Thought in Kant. Dordrec ht: Kluwer, 1989' Kant's ofExperience. 2 vols. London : AlIen &

193 6 .


Kneller and Michael Losonsky. Stanford: Stanford Univers ity Press, 1992. "Bausteine zur Entsteh ungsges chichte der Kritik der reinen Vernunft." Kant-Stu dien 78 (r987): 153-69. SCHULT Z, JOHANN . An Exposition ofKant's Critique ofPure Reason. Translated by]ames C. Morriso n. Ottawa: Univers ity of Ottawa Press, 1995. SMITH, A. H. Kantian Studies. Oxford: Clarend on Press, 1947.

bei Kant. Berlin: WaIter de Gruyter , 197I. The Completeness of Kant's Table ofJudgm ents. Trans.

Second edition. London : Macmil lan, 1923. London: Methue n, 1966.


to Kant's "Critique


The Bound, ofSense: An Essay on Kant's Critique ofPure Reason. Kant und das Problem der Gesetzmiissigkeit der Natu7:

Berlin: WaIter de Gruyter , 199I.

Kant's Critique of Pure Reason within the Tradition of


Introdu ction
Modern Logic. Edited

David H. Chandler. Hildesheim: Georg Olms,

Problc77Ic der "K:ritik der reinen . Kant-


Stuttgart: v AI H I N G ER, HAN s. Km71771entar zur Kritik der reinen Vernunft. 2 vols. 1892. aft, esellsch Verlagsg e Deutsch W Spemann, 1881, and Union


1981. Berlin: WaIter de Gruyter, 1984.

La dMuction transcendcntale dans l'oeuvre

on, and de Kant. 3 vols. Antwerp, Paris, The Hague: De Sikkel, Champi Martinu s Nijhoff, 1934-37 . A. R. C. Duncan. EdinThought. Translated The 1962. Nelson, burgh: Thos. Kant: Der Mann und das Tl'Crk. 2 vols. Leipzig: VORLAN DER, KARL. FelixMe iner, 1924.



Kant's Criticism

Edinburgh: University of Edin-

burgh Press, 1975.


"The Second Edition of the Critique: Toward an Unders tanding of its Nature and Genesis ." Kant-St udien 66 (1975):


Ill.: Open Kant: The Architectonic and Df'(!el'OiJ7uel7t of His Philosophy. LaSalle, Court, 1980.

w. Florida, 198r.


Kant's Silent Decade. Tallahasee: University Presses of

Die Vol'lst,iin,'iIp,'ke:ltder kantischen Urteltsta.tet. Frankfu rt am on the Activity: A Kant's Theory Mass.: dge, Cambri Reason. Pure of of the Critique

Main: Klostermann, 1995.



Harvard University Press, 1963.


Pur e Rea son

Professor in Konigsberg


Published by Johan n Friedrich Hartk noch

17 81

Cri tiqu e

Pur e Rea son

Immanuel Kant
Professor in Konigsberg Membe r of the Royal Academy of Sciences in Berlin

Second edition, improv ed here and there


Johan n Friedrich Hartk noch


Motto (added in the second edition) Dedication (as in the first edition of 178 I) (as in the second edition of 1787) Preface (to the first edition) Preface to the second edition Table of Content s (as in the first edition) Introduction (as in the first edition) 1. The idea of transcen dental philosop hy On the difference between analytic and syntheti c judgmen ts. Il. Division of transcen dental philosophy. Introduction (as in the second edition) 1. On the difference between pure and empirical cognitio n. H. We are in possession of certain a priori cognitions, even the commo n underst anding is not without them. Ill. Philosophy needs a science that determi nes the possibility, the principlesb and the domain of all a priori cognitions. IV On the difference between analytic and syntheti c judgmen ts. V Synthetic a priori judgmen ts are containe d as principles c in all theoreti cal sciences of reason. VI. The general problem d of pure reason.

page .91

95 97 99


12 7



" This Table of Contents is the editors' expansion of the less detailed one provided Kant in the first edition. The second edition containe d no Table of Contents at all. A translation of Kant's own first-edition Table of Contents follows the two versions of the preface, correspo nding to its original location.


Con tent s a special science und er the VII. The idea and the divisions of nam e of a criti que of pure reason. ents I. Tra nsce nde ntal doc trin e of elem first edition) Tran scen dent al aest heti c (as in the


153 155

Firs t Par t. [Intr oduc tion .] Firs t section. On space. Seco nd section. On time.


(as in the seco nd edition) Firs t Part . Tran scen dent al aest hetic Intr odu ctio n. < I> Firs t section. On space. < 2-3> Seco nd section. On time. < 4-7> al aesthetic. < 8> Gen eral rem arks on the tran scen dent c Sec ond Part . Tran scen dent al logi dent al logic scen tran Intr odu ctio n. The idea of a 1. On logic in general. H. On tran scen dent al logic. into analytic and dialectic. IH. On the division of gene ral logic al logi c into the tran scen N On the division of tran scen dent dent al analytic and dialectic. ytic Div isio n one . Tran scen dent al anal epts conc of lytic Boo k I. Ana overy of all pure conc epts Cha pter I. On the clue to the disc of the und ersta ndin g the und erst andi ng in Firs t section. On the logical use of general. tion of the nd Seco section. On the logical func 9> < und erst andi ng in judg men ts. the of epts Thi rd section. On the pure conc 2> 10-1 < und erst andi ng or categories. of the epts conc pure the Cha pter H. On the dedu ctio n of und erst andi ng a a tran scen dent al Firs t section. On the principles of dedu ctio n in general. < 13> dedu ctio n of the Tran sitio n to the tran scen dent al categories. < 14> possibility ion. On the a priori grou nds for the sect nd Seco on) editi first of experience. (as in the


174 178


195 197
201 202









Contents Third section. On the relationa of the understanding to objects in general and the possibility of cognizing these a p'rioTi. (as in the first edition) Second Section. TI-anscendental deduction of the concepts of the understanding. (as in the second edition) I5- 2 7 Book n. Analytic of principles Introduction. On the transcendental power of judgment in genera.l Chapter I. On the schernatism of pure concepts of the understanding Chapter H. System of all principles of pure understanding Section 1. On the supreme principle of all analytic judgments. Section H. On u~e supren1e principle of all synthetic judgnlents. Section Ill. Systematic representation of all synthetic principles of pure understanding. I. Axioms of intuition 2. i\nticipations of perception 3 Analogies of experience A First analogy: principle of persistence of substance. B. Second analogy: principle of temporal succession according to the law of causality. C. Third analogy: principle of simultaneity according to the law of reciprocity or conul1unity. 4 The postulates of ell1pirical thought in general Refutation of idealism (added in the second edition) General note on the system of principles (added in the second edition) Chapter IH. On the ground of the distinction of all objects in general and nOU771C7W (as in the first edition) Chapter Ill. On the ground of the distinction of all objects in general into phenmuena and n01l711C1Ul (as in the second edition) Appendix: On the amphiboly of concepts of reflection Renlark to the amphiboly of concepts of reflection Division two. Transcendental dialectic Introduction. 1. Transcendental illusion


267 268

278 279
283 286

295 299

316 321 326 334


354 366

384 384



Con tent s scendental illusion IT. On pure reason as the seat of tran A. On reason in general. B. On the logical use of reason. C. On the pure use of reason. on Book I. On the concepts of pure reas ral. gene in s idea Section I. On the ideas. Section Il. On the transcendental dental ideas. scen tran of em Section Ill. The syst reason pure of es renc infe l Boo k II. The dialectica on reas pure of sms logi para Cha pter 1. The on) editi first the (as in Firs t paralogism of substantiality. Second paralogism of simplicity. Thi rd paralogism of personality. Fou rth paralogism ofideality. doct rine of the soul Obs erva tion on the sum of the pure reason Cha pter 1. The paralogisms of pure on) editi nd seco (as in the f of the persistence of Refutation of Mendelssohn's proo the soul. sitio n from rational Gen eral rem ark conc erni ng the tran psychology to rational cosmology. reason Cha pter II. The anti nom y of pure ical ideas. olog cosm of em syst Section I. The on. reas pure of ic thet Section Il. The anti Firs t conflict Second conflict Thi rd conflict Fou rth conflict on in these conflicts. Section Ill. On the inte rest of reas prob lem sa of pure reason, Section IV: On the transcendental capable of a solution. insofar as they absolutely mus t be of the cosmological Section V Skeptical repr esen tatio n dental ideas. questions raised by all four transcen as the key to solving Section VI. Tran scen dent al idealism the cosmological dialectic. cosmological conflict of Section VII. Critical decision of the reason with itself. e b of pure reason in Section VIII. The regulative principl rega rd to the cosmological ideas. 387 387 389 390 394 395 399 405 409 41 I

415 417 422 425 432 445


456 459 460 467 470 476 484 490 496

503 508 SI!




Content s Section IX. The empirical use of the regulative of reason in regard to the cosmological ideas. I. Resolution of the cosmologic~l idea of totality of the composition a of the appearances into a world-whole. II. Resolution of the cosmological idea of totality of division of a given whole in intuition. Conclud ing remark on the resolution of the mathem atical-transcendental ideas and preamble to the solution of the dynamical transcendental ideas. HI. Resolution of the cosmological idea of the in the derivation of occurrences in the world from their causes. The possibility of causality through freedom. Clarification of the cosmological idea of freedom. IV Resolution of the cosmological idea of the of the dependence of appearances regarding their existence in general. Concluding remark to the entire antinomy of pure reason. Chapter Ill. The ideal of pure reason Section I. The ideal in general. Section H. The transcendental ideal (prototypo71 transccndentale).b Section HI. The grOlmds of proof of speculative reason inferrin g the existence of a highest being. Section IV On the impossibility of an ontological of God's existence. Section V On the impossibility of a cosmological of God's existence. Discovery and explanation of the dialectical illusion in all transcendental proofs of the existence of a necessary being. Section VI. On the impossibility of the physicotheological proof. Section VII. Critique of all theology from principles' of reason. Appendix to the transcendental dialectic On the regulative use of the ideas of pure reason. On the final aim of the natural dialectic of human reason.

52 5


53 2 535


55 1

55 1 553


.5 I

transcendental prototyp e


Con tent s

hod ll. Tra nsce nde ntal doc trin e of met

627 Intr odu ctio n. reason Cha pter I. The discipline of pure reason in dogmatic use. pure Section 1. The discipline of reason with rega rd to its pure Section H. The discipline of polemical use. satisfaction pure reas on On the impossibility of a skeptical that is divided against itself. reason with regard to Section fir. The discipline of pure hypotheses. reason with regard to its Section IV The discipline of pure proofs. The cano n of pure reason fi. pter Cha the pure use of our reason. Section 1. On the ultim ate end of est good. high Sect ion II. On the ideal of the ing, and believing. know Sect ion Ill. On having an opinion, on reas pure Cha pter HI. The architeetonic of on reas pure of ry histo The IV Cha pter

630 643 652 658 665 672 673 676 684 691 702


Baco de Verulam


Instauratio Magna. Praefatio

De nobis ipsis silemus: De re autem, quae agitur, ut ho,m:m(~s earn non Opinio nem, sed Opus esse cogitent; ac pro certo non Sectae nos alicuius, aut Placiti, sed utilitatis et humana e fundamenta moliri. Deinde ut suis commo dis aequi . . . in commu ne consulant ... et ipsi in partem veniant. Praeter ea ut bene sperent , Instaur ationem nostram ut quidda m infinitu m et mortale tinlgant, et animo concipiant; quum revera sit infiniti erroris finis et terminu s legitimus. a

This motto was added in the second edition: Bacon of Verulam The Great Instaurat ion. Preface Of our own person we will say nothing. But as to the subject matter with which we are concerned, we ask that men think of it not as an opinion but as a work; and consider it erected not for any sect of ours, or for our good pleasure, but as the foundation of human utility and dignity. Each individual equally, then, may reflect on it himself ... for his own part ... in the common interest. Further, each may well hope from our instauration that it claims nothing infinite, and nothing beyond what is mortal; for in truth it prescribes only the end of infinite errors, and this is a legitimat e end.


To his Excellency, the Royal Minister of State, Baron van Zedlitz I


Gracious Lord,


To further for one's own part the growth of the sciences is to your Excellency's own interest; for the former is most imvVa:rdly up with the latter, not only through the exalted post as a protector of the sciences, but also through the more intimate relationshipb of a lover and an enlightened connoisseur. On this account, I myself of only means within my capacity to show my gratitude gracious trust with which your Excellency honors me, as though that could contribute something to this aim. For someone who enjoys the life of speculation the "n,~rnM,,1 enlightened and competent judge is, given his modest ful encouragement to toils whose utility is great, but distant, it is wholly misjudged by vulgar eyes. To such a judge and to his gracious attention, I now dedicate this piece of writing; to his protection I commend all the remaining business of my literary vocation; and with deepest reverence I am, Your Excellency's humtlle, most obedient servant



Immanuel Kant
Konigsberg: the 29th of March, 1781
, As in the first edition. b vertrmlterc Verbii!t17is; this last word was added later, according to Kant's letter to Biester of 8 June 1781.


a<Gracious Lord, To further for one's own the growth of the sciences is to your Excellency's own interest; for the former is most im;\T3Jrdlv up with the latter, not only throug h the exalted post as a protect or of the sciences, but also throug h the more intimat e relationshipb of a lover and an enlight ened connoisseur. On this account, I avail of only means within my capacity to show my gratimd e for trust with which your Excellency honors me, as though tribute someth ing to this aim. To the same gracious attentio n with which Your Excellency has dignified the first edition of this work, I dedicate also this second one, and at the same time all the remain ing business of my literary vocation; and with deepest reverence I am, Your Excellency's humLl!e, most obedie nt servant, Konigsberg, the 23rd of April, 1787>


Imman ud Rant

b vc;1mufcrc

, As in the second edition. Verhdltni"; this last word was added later, according to Kant's letter to Biester of8]une 1781.




Human reason has the peculiar fate in one species of its cognitions that it is burden ed with questions which it cannot dismiss, since are given to it as problem s b by the nature of reason itself, it also cannot answer, since they transce nd every capacit y of reason. Reason falls into this perplexity throug h no fault of its own. It begins from principles whose use is lmavoidable in the course of experie nce and at the same time sufficiently warran ted by it. With these pnncJlpl (~s it rises (as its nature also requires) ever higher, to more remote con(Htions. But since it becomes aware in tl1is way that its business must a1ways remain incomp lete because the questions never cease, reason sees itself necessitated to take refuge in principles that overstep all possibl e use in experience, and yet seem so unsuspicious that even or,ClU13I Y common sense agrees with them. But it thereby falls into obsCllnl:y contradictions, from which it can indeed surmise that it must somewhere be procee ding on the ground of hidden errors; but it cannot discover them, for the principles on which it is proceed ing, since surpass the bounds of all experience, no longer recognize any stone of experience. The battlefield of these endless controversies is called metaph ysics. There was a time when metaphysics was called the queen of sciences, and if the will be taken for the deed, it cieserved this title of honor, on accoun t of the preemi nent importa nce ofits object. Now, in ') accordance with the fashion of tl1e age, the queen proves despised on all sides; and the matron , outcast and forsaken, mourns Hecuba : Modo maxima rerum, tot generis natisque potens - nunc trahor exul, inops - Ovid, lVletam07phoses. d In the begilming, under the admini stration of the dogma tists,2 her rule was despot ic. Yet because her legislation still retaine d traces of ancient barbarism, this rule gradual1y degene rated throug h interna l wars into complete anarch y; and the skeptics,3 a kind of nomad s who all permanent cultivation of the soil, shatter ed civil unity from time to
" As in the first edition. Kant wrote a new preface for the second edition, given below.
b aut!(c!!:cben

A viii


d "Greatest of all by race and birth, I now am cast out, powerles s" (Ovid, Mt'trm10rjlhw'cs

13:508- ro).


Preface <ID a few of them , they could time. But since ther e wer e con tinu ally atte mpt ing to rebuild, not prev ent the dog mat ists from una nim ous ly acce pted amo ng them thou gh nev er acco rdin g to a be ld wou end an gh seem ed as thou selves. Onc e in rece nt time s it even a the com peti ng lawfulness of the es, ersi put to all thes e con trov phy siol ogy of ain ded, thro ugh a cert claims wou ld be com plet ely deci ed out that turn it famous Locke);4 but the hum an und erst and ing (by the rabb le of the to ted que en was trac ed alth oug h the birt h of the pur por e bee n hav e efor ther ensi ons wou ld com mon exp erie nce and her pret claims, her rted asse eles s she still righ tly rend ered suspicious, nev erth met athus ly; false attr ibut ed to her because in fact this gen ealo gy was thus and , ism mat wor m-e aten dog physics fell bac k into the sam e old AX have to was nce scie pt out of whi ch the into the sam e pos itio n of con tem s (as we pers uad e ourselves) have path all r afte bee n extricated. Now 5 um and com plet e indi ffer enti sm, hee n trie d in vain, wha t rule s is tedi also time e sam the nig ht in the sciences, but at the mot her of chaos and , of thei r inci pien t tran sfor mat ion the orig in, or at leas t the prel ude obome bec e hav pplied effo rt they enli ghte nme nt, whe n thro ugh ill-a scure, confused, and useless. ren ce with resp ect to such inFor it is poin tles s to affect ind iffe enatu re can not be ind iffe ren t. Mor quiries, to who se obje ct hum an ble niza cog unre es k to mak e themselv over, how eve r muc h they may thin soschools for a pop ular style, thes e the of e by exc han ging the lang uag alan, at g thin any k nt that they thin called ind iffe ren tist s, to the exte yet they ch whi ns, aph ysic al assertio ways una void ably fall bac k into met eles s this indifference, occurring erth Nev ise. prof esse d so muc h to desp , and dire cted prec isel y at thos e sciami d the flou rish ing of all sciences b be had at all) we cou ld leas t do with enc es who se results (if such are to is s Thi on. ecti refl and our atte ntio n out, is a phe nom eno n dese rvin g AXl ghtl essn ess of our age, but of its thou the of evid entl y the effe ct not ilch will no long er be put off with ripe ned pow er of jud gme nt,* whi age's way plaints about the superficiality of our * Now and again one hears com not see do I Yet ce. scien ed t the decay of well-ground of thinking, and abou well laid, such as mathematics, physics, that those sciences whose grounds are er, they maintain their old reputation etc., in the least deserve this charge; rath of natural science, even surpass it. This for well-groundedness, and in the case tive in other species of cognition if same spirit would also prove itself effec ect their principles. In the absence of only care had first been taken to corr t criticism are rather proofs of a wellthis, indifference, doubt, and finally stric the genuine age of criticism, to which grounded way of thinking. Our age is

Rechtmdssigkeit KC717rtni'se

, Principicn


Preface <A> lusory knowledge, and which demands that reason take on anew to the most difficult of all its tasks, namely, that of self-knowledge,a institute a court of justice, by which reason may secure its claims while dismissing all its groundless pretensions, and this not mere decrees but according to its own eternal and unchangeable and this court is none other than the critique of pure reason itself. 6 Yet by this I do not understand a critique of books systems, a critique of the faculty of reason in general, in respect of all the cognitions after which reasonb might strive independently of all experience, and hence the decision about the possibility or of a metaphysics in general, and the determination of its sources, as as its extent and boundaries, all, however, from principles.c It is on this path, the only one left, that I have set and I flatter myself that in following it I have succeeded in removing all those errors that have so far put reason into dissension with itself in its nonexperiincaential use. I have not avoided reason's questions by pleading pacity of human reason as an excuse; rather I have completely specified these questions according to principles,d and after discovering the where reason has misunderstood itself, I have resolved them to reason's full satisfaction. To be sure, the answer to these questions has not turned out just as dogmatically enthusiastic lust for knowledge have expected; for the latter could not be satisfied except through magical powers in which I am not an expert. Yet this was also not the intent of our reason's natural vocation; and the duty of philosophy was to ish the semblance arising from misinterpretation, even if many and beloved delusions have to be destroyed in the process. In this business I have made comprehensiveness my chief aim in view, and I make bold to say that there cannot be a single metaphysical problem that not been solved here, or at least to the solution of which the key has not been provided. In fact pure reason is such a perfect unity that if its ciple' were insufficient for even a single one of the questions that are set everything must submit. Religion through its holiness and legislation through its majesty commonly seek to exempt themselves from it. But in this way they excite a just suspicion against themselves, and cannot lay claim to that unfeigned respect that reason grants only to that which has been able to withstand its free and public examination.
, SelhterkC7777tnis
b sie. To agree with "faculty of reason" (das Vel~nU1"ftz;,C77li'ii!len) the pronoun should have



been neuter; perhaps Kant was taking the antecedent to ( Principien


"reason" (die


, Princip


Preface <Po d, [principle] mig ht as wen be discarde for it sque er oth not be up to answering any of the because then it also tions with com plet e reliability.7 perceive in the face of the reader Wh ile I am saying this I believe I pt at claims that are app aren tly so an indi gna tion mixed with con tem A XlV they are inco mpa rabl y mor e mod pret enti ous and immodest; and yet s end pret com mon est pro gram who erat e than thos e of any auth or of the l or the necessity of a first beg insou the to prov e the simple natu re of auth or pledges him self to extend nin g of the wor ld. For such an which I bou nds of possible experience, of hum an cog niti on bey ond have to I ead asses my capacity; inst hum bly adm it that this wholly surp ive acaust pur e thinking; to gain exh do mer ely with reas on itse lf and its it is in ause far bey ond myself, bec qua inta nce with them I nee d not seek s me give also com mon logic already mys elf that I enc oun ter them , and tema syst of reas on may be fully and 3n example of how the simple acts may I h que stio n is r3ised how muc ically enu mer ated ; only here the if all the m3terial and assistance of hop e to settle with these simple acts experience are take n away from me. re3ching eac h of the ends, and for So muc h for the com plet ene ss in of them toge ther , which ends the com pre hen sive nes s in reac hing set up for us by the natu re of cogare not prop osed arbitrarily, but are critical investigation. niti on itself, as the ma tter of our ity, two thin gs that con cern the clar Fur ther mor e cer tain ty and AXV viewed as essential demands, which form of the investigation, are to be an or who ven ture s upo n so slippery may righ tly be mad e on the auth und erta king . have myself pro nou nce d the judgAs far as cert ain ty is concerned, I is in no way 3Bowed to hol d opin men t th3t in this kind of inqu iry it n idde forb a is s like an hypothesis ion s, and that any thin g that even look for sale even 3t the lowest price up put commodity, whi ch sho uld not be n it is discovered. For every cognitio but mus t be confiscated as soo n 3S held be to ri proclaims that it W3ntS that is supposed to be cert3in a prio e is this true of 3 determin3tion mor n for absolutely necess3ry, and eve ch is to be the stan dard and thus even of all pur e cog niti ons a priori, whi phical) certainty. Wh ethe r I have the example of all apodictic (philoso in that resp ect rem ains wholly to perf orm ed wh3 t I have just pledged to it is 3pp ropr iate for an 3uth or only the judg men t of the reader, since es. judg his e abo ut thei r effect on pres ent the grou nds , but not to judg entl y be the cause of weakenvert imd But in ord er th3t he sho uld not self or may be perm itte d to note him ing his own arguments, the auth AXV l of end tal den pert ain only to the inci those places that , even thou gh they may he that som e mistrust, in ord er the work, may be the occasion for influence that even the reader's the in 3 time ly man ner cou nter act

Preface <A> slightest reserva tion on this point may have on his over chief end. I am acquainted with no investigations more for to the bottom of that faculty we call the underst anding , and at same time for the determ ination of the rules and boundaries of its use, those I have underta ken in the second chapter of the Analytic, under the title Deduc tion of the Pure Conce pts of the Unders tandin g; they are also the investigations have cost me the most, but I hope not unrewa rded, effort. This which goes rather deep, has two sides. One side refers to the objects of the pure understanding, and is supposed to demon strate and make compre hensibl e the objective validity of its concepts a pTiori; thus it belongs essentia lly to my ends. The other side deals with the pure unders tanding itself, concerning its possibility and the p;;'wers of cogniti on on which it itself rests; thus it considers it in a s~bjective relation, and althoug h this exposition is of great importa nce in respect of my chief end, it does not belong essentially to it; because the chief questio n always remains : "What and how much can unders tanding and reason cognize free of experience?" and not: "How is the faculty of thinkin g itself possible ?"8 Since the latter question is someth ing like the search for cause a given effect, and is therefo re someth ing like a hypothesis (although, as I will elsewhere take the opportu nity to show, this is not in fact how matters stand), it appears as if I am taking the liberty in this case of expressing an opinio n, and that the reader might therefo re be free to another opinio n. In view of this I must remind reader in advance that even in case my subjective deducti on does not produc e complete conviction that I expect, the objective deducti on that is my mary concern would come into its full strengt h, on which what is said at pages [AJ 92-3 should even be sufficient by itself: Finally, as regards clarity, a the reader has a right to demand first cursive (logical) clarity, throug h concep ts, but then also intuitiv e (aesthetic) clarity, throug h intuitio ns, that is, throug h examples or other illustrations in concreto. I have taken sufficient care for the former. That was essential to my underta king but was also the conting ent cause of the fact that I could not satisfY the second demand, which is less strict but still fair. In the progress of my labor I have been almost COIlst~ll 1tly undecided how to deal with this matter. Examples and illustrations ways appeared necessary to me, and hence actually appeared in their proper place in my first draft. But then I looked at the size of my task and the many objects with which I would have to do, and I became aware that this alone, treated in a dry, merely schola stic manner,
, Dcutlich!,cit


A xviii


Preface <A>




it suffice to fill an extensive work; thus I found it inadvisable to swell a for only ry further with examples and illustrations, which are necessa for suitable made popula r aim, especially since this work could never be popula r use, and real experts in this science do not have so much need be for things to be made easy for them; althoug h this would always rcounte ing agreeable, here it could also have brough t with it someth meais book a of productive. The Abbe Terrass on says that if the size and sured not by L~e numbe r of pages but by the time needed to underst r shorte much be it, then it can be said of many a book that it would view our direct we if if it were not so short.9 But on the other hand, is toward the intelligibility of a whole of speculative cogniti on that equal with wide-ra nging and yet is connec ted in princip le,' we could had right say that many a book would have been much dearer if it parts the in helpb not been made quite so dear. For the aids to clarity whole, since the reader cannot quickly enough but often confuse in and attain a survey of the whole; and all their bright colors paint over which make unreco gnizabl e the articulation or structu re of the system, 10 yet matters most when it comes to judging its unity and soundness. to reader It can, as it seems to me, be no small induce ment for the carof ct unite his effort with that of the author, when he has the prospe nt rying out, according to the outline given above, a great and importa metapiece of work, and that in a comple te and lasting way. Now one physics, according to the concepts we will give of it here, is the only and effort, of all the sciences that may promis e that little but unified nothing that indeed in a short time, will comple te it in such a way that inremains to posteri ty except to adapt it in a didacti c manne r to its it For tentions, yet withou t being able to add to its conten t in the least. or, is nothing but the invent ory of all we possess throug h pure reason dered systematically. Nothin g here can escape us, because what reason t to brings forth entirely out of itself cannot be hidden, but cis brough disbeen light by reason itself as soon as reason's commo n princip le has it covered. The perfect unity of this kind of cognition, and the fact that exwould arises solely out of pure concepts withou t any influence that tend or increase it from experience or even particu lar intuitio n, which would lead to a determ inate experience, make this uncond itioned completeness not only feasible but also necessary. Tecum habita, et naris quam d sit tibi curta supellex. - Persius. Such a system of pure (speculative) reason I hope myself to deliver


helfen. Kant's text reads "fihlen" (are missing). We follow Erdmann , reading

Princip possessions are" "Dwell in your own house, and you will know how simple your 2). 5 4: Satires (Persius,


Preface <A> under the title Metaphysics of Nature, tensive will be incomparably richer in content this crItique, which had first to display the sources and conditions of its possibility, and needed to clear and level a ground that was completely overgrown. Here I expect from my reader the patience a but there I will expect the cooperative spirit assistance a worker; for however completely the prindplesa of the system may expounded in the critique, the comprehensiveness of the system itself requires also that no derivative concepts should be lacking, however, cannot be estimated a priori in one leap, but must be tiJ,ciUIJ<llq sought out; likewise, just as in the former the whole synthesis cepts has been exhausted, so in the latter it would be ad(.lltlon:aHy manded that the same thing should take place in respect of analysis, which would be easy arfd more entertainment than labor. I have only a few more things to remark with respect to the book's printing. Since the beginning of the printing was somewhat delayed, I was able to see only about half the proof sheets, in which I come upon a few printing errors, though none that confuse the sense except the one occurring at page [A] 379, fourth line from the bottom, where specific should be read in place of skeptical. The Antinomy of Pure Reason, from page [A] 425 to page [A] 461, is arranged in the manner of a table, so that everything belonging to the thesis always continues on the left side and what belongs to the antithesis on the right side, which I did in order to make it easier to compare proposition and counter-proposition with one another.
, Principicn




Preface to

second edition a

the cog niti ons belo ngin g to the con Wh eth er or not the trea tme nt of ch whi ng ethi rse of a science is som cern of reason travels the secure cou r man y preliminaries and prepaafte If can soo n be judg ed by its success. k as soo n as it approaches its end, rations are made, a science gets stuc a mus t ofte n go bac k and set out on or if in ord er to reac h this end it orkco-w t impossible for the diff eren b new path ; or likewise if it proves in whi ch they sho uld pursue way the to as ity nim una eve achi ers to be sure that such a stud y is merely thei r com mon aim; then we may having ente red upo n the secure gro ping about, that it is still far from a service to reas on if we can possicourse of a science; and it is already t have to give up as futile muc h of wha bly find that path for it, even if we . form ed with out deliberation was included in the end previously has traveled this secure course can Tha t from the earliest times log ic BVll l tim e of Aristotle it has not had to be seen from the fact that since the we cou nt the abo litio n of a few disgo a single step backwards, unless inct dete rmi nati on of its presentapensable subtleties or the mor e dist mor e to the elegance than to the tion , which imp rove men ts belo ng her remarkable abo ut logic is that secu rity of that science. "Vh at is furt to take a single step forward, and unti l now it has also bee n unable plete. For if appearance to be finished and com ther efor e seems to cho logi psy rge it by inte rpo latin g som e mod ems have tho ugh t to enla gination, ima nitive powers (abo ut cal chapters abo ut our diff eren t cog the difor on ut the orig in of cog niti wit), or met aph ysic al cha pter s abo objects' of rdance with the diversity fere nt kinds of cert aint y in acco about s pter or anth rop olo gica l cha (abo ut idealism, skepticism, etc.), only eeds remedies), then this proc our prejudice (abo ut thei r causes and an not is It ! r natu re of this science. from thei r igno ranc e of the peculia ies ndar bou r the sciences whe n thei imp rov eme nt but a defo rma tion of ther ; the boundaries of logic, howano are allowed to run over into one y by the fact that logic is the science ever, are dete rmi ned quit e precisel tly proves not hin g but the formal tllat exhaustively pres ents and stric B lX
on. ces the preface from the first editi This new preface, so entitled, repla here because it sense e mak not does h whic lgt" (result or ensue), b Kant 's tcc"'{t reads "erfiJ o in read ing verfolgt. is an intransitive verb; we follow Grill , Objecte


Preface to the second edition <B>

rules of all thinking (whether this thinking be or a whatever origin or object" it may have, and whatever contingent or natural obstacles it may meet with in our minds). For the advantage that has made it so successful logic has solely its own limitation to thank, since it is thereby justified in abstracting - is indeed obliged to abstract - from all objectsb of cognition all distinctions between them; and in logic, therefore, the undejrst:mclmg has to do with nothing further than itself and its own form. How more difficult, naturally, must it be for reason to enter upon the secure path of a science if it does not have to do merely with itself, but has to deal with objectsC too; hence logic as a propadeutic constitutes only outer courtyard, as it were, to the sciences; and when it comes to mation, a logic may indeed be presupposed in judging about the latter, but its acquisition must be sOllgl1t in the sciences properly and tively so called. Insofar as there is to be reason in these sciences, something in them must be cognized a priori, and this cognition can relate to its object in either of two ways, either merely detennining the object and its concepf(which must be given from elsewhere), or else also making the obactual. The former is theoretical, the latter practical cognition of reason. In both the pure part, the part in which reason determines its .Object! wholly a priori, must be expounded all by itself, however much or little it may contain, and that part that comes from other sources must not be mixed up with it; for it is bad economy to spend whatever comes in without being able later, when the economy comes to a standstill, to distinguish the part of the revenue that can cover the expenses from the part that must be cut. Mathematics and physics are the two theoretical cognitions of reason that are supposed to determine their objectse a priori, the former entirely purely, the latter at least in part purely but also following standards of sources of cognition other than reason. Mathematics has, from the earliest times to which the history of human reason reaches, in that admirable people the Greeks, traveled the secure path of a science. Yet it must not be thought that it was as easy for it as for logic - in which reason has to do only itself - to find that royal path, or rather itself to open it up; rather, I believe mathematics was left groping about for a long time (chiefly among Egyptians), and that its transformation is to be ascribed to a rf"'VOIHcion, brought about by the happy inspiration of a single man in an at, Object '. , Objecte d Object , Objecte




Preface could no lon ger be to be take n tem pt from whi ch the a science was ente red on and premissed, and the secu re cou rse nite exte nt. The hist ory of this revscri bed for all tim e and to an infi whi ch was far mor e imp orta nt than olut ion in t,~e way of thin king of the lucky famous Cap e"the disc ove ry of the way aro und But the legus. for d has not bee n pres erve one who bro ugh t it nam es the who s gen es Lae rtiu end han ded dow n to us by Dio ons tradem l rica met elem ents of geo repu ted inve ntor of the sma lles t d in no stan t, men judg acc ord ing to com mon tions, even of thos e t by ugh wro on rati alte ory of the nee d of pro of - proves that the mem have t mus s step foot in its earl iest the discovery of this new path mat hem atic ians , and was ther eby to nt orta imp seem ed exceedingly who ligh t bro ke upo n the firs t pers on rend ered unf org etta ble. A new " ales "Th ed can was a he (wh ethe r dem ons trat ed the isosceles tria ngle was do to he t wha that som e oth er name).I2 For he fou nd or even trac e its mer e conor re, figu this in saw not to trac e wha t he BX11 the pro pert ies of the figure; but cep t, and read off, as it wer e, from wha t he him self thought to pro duc e the latt er from rath er that he acc ord ing to a pres ente d (thr oug h con stru ctio n) into the obje ct rely a priori secu ng kno w som ethi priori concepts, and that in ord er to owed necfoll t wha not hin g exc ept he had to ascribe to the thin g ce with its rdan acco put into it in essarily from wha t he him self had con cep t. er to find the highway of science; It too k natural science muc h long cen turi es since t.!}e suggestion of the for it is only abo ut one and a half asioned this discovery and partly furinge niou s Francis Bac on part ly occ ady on its tracks - whi ch discovery, ther stimulated it, since one was alre lained by a sud den revo luti on in the ther efor e, can just as muc h be exp r natu ral science only insofar as it way of thinking. Her e I will con side b es. is gro und ed on emp iric al principl wei ght cho sen by him self down an a of s I3 ball ed roll \Vh en Galileo ght that he 14 mad e the air bea r a wei inclined plane, or whe n Torricelli of water, mn colu wn kno al to that of a had previously tho ugh t to be equ then and caix" I into als cha nge d met or whe n in a late r tim e Stahl 5 and ng ethi som g ovin rem al by first cha nge d the latte r bac k into met BX111
ction in a letter to lateral); but on the basis of his corre Kam's text reads "gleichseitig" (equi (isosceles). nt mea have to ars appe Schiitz of 25 June 1787 (10466), he hronistic; prior to the this as "oxides," but that is anac , Kalk. Kem p Smit h translates eived to be what was conc was Lavoisier, the calx chemical revo lutio n of Pries tley and it discovered ilIat was later only off; n drive been had left of a metal after its phlo gisto n ation. this process was actually one of oxid


to the second edition <B>

then putting it back again,* a light dawned on all those ture. They compre hended that reason has insight It Itself produces accordi ng to its own design; that it must take the lead with piiriCiplesa for its judgme nts accordi ng to constan t laws and compel namre io answer its questions, rather than letting nature guide its movements by keeping reason, as it were, in leading-strings; for otherw ise accidental observations, made accordi ng to no previously designe d plan, can never connec t up into a necessary law, which is yet reason seeks and requires. Reason, in order to be taught nature, must approach nature with its principles b in one hand, accordi ng to alone the agreem ent among appearances can count as laws, in the other hand, the experiments thoogh t out in accordance these ciples c- yet in order to be instruc ted by nature not like a pupil, who has recited to him whatev er the teacher wants to say, but like an apJ)olnt l::d judge who compels witnesses to answer the questions he puts to them. Thus even physics owes the advantageous revolut ion in its ing to the inspira tion that what reason would not be able to of itself and has to learn from nature, it has to seek in the latter (though not merely ascribe to it) in accordance with what reason itself puts into nature. This is how natural science was first brough t to the secure course of a science after groping about for so many centuries. Metaph ysics - a wholly isolated speculative cognition of reason that elevates itself entirely above all instruc tion from experience, and that through mere concepts (not, like mathematics, throug h the application of concepts to intuitio n), where reason thus is supposed to be its own pupil- has up to now not been so favored by fate as to have been to enter upon the secure course of a science, even though it is than all other sciences, and would remain even if all the others were swallowed up by an all-consuming barbarism. For in it reason continu ously gets stuck, even when it claims a priori insight (as it pretend s) into those laws confirmed by the commo nest experience. In metaphysics we have to retrace our path countless times, because we find that it does not lead where we want to go, and it is so far from reachin g unanim ity in assertions of its adheren ts that it is rather a battlefield, and indeed one that appears to be especially determ ined for testing one's powers in mock combat; on this battlefield no comba tant has ever gained the least
* Here I am not following exactly the thread of the history of the experimental



method, whose first beginnin gs are also not precisely known.



Preface to base any lasting possession on bit gro und , nor has any bee n able that up to now the pro ced ure of his victory. Hen ce ther e is no ing, and wha t is the worst, a groping metaphysics has bee n a mer e grop amo ng mer e concepts. path of science stin could not be Now why is it that here the secure y then has natu re afflicted our reafound? Is it perh aps impossible? "Wh such a path , as if it wer e one of reason WiLh the restless striving for Still mor e, how little cause have we son's mos t imp orta nt occupations? of the mos t imp orta nt part s of our to place trus t in our reas on if in one ely forsake us but even entices us desire for knowledge it does not mer ays us: Or if the path has merely with delusions and in the end betr we use that mig ht lead us to hope elud ed us so far, wha t indications may be luckier than thos e who have gone that in rene wed atte mpt s we will befo re us? mat hem atic s and natu ral science, I sho uld thin k that the examples of are thro ugh a revo luti on brought whi ch have bec ome wha t they now BXV1 eno ugh that we mig ht reflect on the abo ut an at once, wer e rem arka ble so the ways of thin king that has been essential elem ent in the change in far inso it t as an experiment, imi tate advantageous to them , and, at leas as rational cognition, mig ht permit. as thei r analogy with metaphysics, all our cog niti on mus t conform to Up to now it has bee n assumed that out som ethi ng abo ut them a priori the objects; but all atte lnpt s to find nd our cog niti on have, on this prethro ugh concepts that would exte ce let us once try whe ther we do not supposition, com e to noth ing. Hen aphysics by assuming that the obget fart her with the prob lem s of met n, whi ch would agree bett er with the jectsa mus t con form to our cognitio ri cog niti on of them , whi ch is to estab requ este d possibility of an a prio d woul s Thi b re they are given to us. lish som ethi ng abo ut objects befo ernicus,'6 who, whe n he did not Cop of be just like the first thou ghts tion of the celestial mot ions if he asmak e good prog ress in the explana t revolves arou nd the observer, tried sum ed that the enti re celestial hos success if he mad e the observer reto see if he mig ht not have grea ter in metaphysics we can try in a simvolve and left the stars at rest. Now conform itio n of objects. If intu itio n has to Bxvii ilar way rega rdin g the intu know can we then I do not see how to the con stitu tion of the objects, C s) sense the of obje ct (as an obje ct any-thing of them a priori; but if the very can I faculty of intu itio n, then conforms to the con stitu tion of our elf. Yet because I can not stop \\~th mys well repr esen t this possibility to ome cognitions, but mus t refer them these intuitions, if they are to bec thei r obje ct and dete rmi ne this obas repr esen tatio ns to som ethi ng as

Objecte Objecte , Object



the second edition <B>

ject through them, I can assume either that the concep ts thl'Olllg- h I bring about this determ ination also conform to objects, and I am once again in the same difficulty about how I could know an'vttlin gabout them a priori, or else I assume that the objects, or what is the same thing, the experience in which alone they can be cognize d (as given jects) conform s to those concep ts, in which case I immed iately see an easier way out of the difficulty, since experience itself is a kind of cognition requiri ng the underst anding , whose rule I have to presupp ose in p:lyselfbefore any object is given to me, hence a priori, which rule is expressed in concep ts a priori, to which all objects of experience must therefore necessarily conform , ani'f with which they must agree. As objects insofar as they are though t merely throug h reason, and necessarily at that, but that (at least as reason thinks them) cannot be given in experience at all - the attemp t to think them (for they must be capable of being though t) will provide a splendi d touchst one we assume as the altered method of our way of thinkin g, namely that we can cognize of things a priori only what we ourselves have into them.* This experim ent succeeds as well as we could wish, and it promis es to metaphysics the secure course of a science in its first part, where it concerns itself with concep ts a priori to which the corresp onding objects appropriate to them can be given in experience. For after this alterati on in our way of thinkin g we can very well explain the possibility of a cognition a priori, and what is still more, we can provide satisfactory proofs of the laws that are the a priori grourtd of nature, as the sum total of objects of experience - which were both impossible accordi ng to the earlier way of proceeding. But from this deducti on of our faculty of cognizi ng a pri* This method, imitated from the method of those who studYl1ature, thus con-

B xvm


sists in this: to seek the elements of pure reason in that which admits of being confirmed or r.efuted through an experiment. Now the propositions of pure reason, especially when they venture beyond all boundaries of possible experience, admit of no test by experiment with their objects a (as in natural science): thus to experiment will be feasible only with concep ts and principles that we assume a priori by arranging the latter so that the same objects can be considered from two different sides, on the one side as obof the senses and the understanding for experience, and on the other side as objects that are merely thought at most for isolated reason striving beyond the bounds of experience. If we now find that there is agreement with the principleb of pure reason when things are considered from this twofold standpoint, but that an unavoidable conflict of reason with itself arises with a standpoint, then the experiment decides for the correctness of that distinction.



., Objecte


to the second edition <B> such procedu res speculative reason has at least made room an extension, even if it had to leave it empty; and we remain at llldeed we are called upon by reason to fill it if wc can data of reason. * Now the concer n of this critique of pure speculative reason consists in that attemp t to transfo rm the accepte d proced ure of metaph ysics, undertaking an entire revolut ion accordi ng to the exampl e of the geometers and natural scientists. It is a treatise on the method , not a system of the science itself; but it catalog s the entire outline of the science of metaphysics, both in respect of its bounda ries and in respect of its entire interna l structu re. Forrpu re speculative reason has about it, that it can and should measur e its own capacit y a accordi ng to the differen t ways for choosin g tl1e objects b of its tl1inking, and also completely enumer ate the manifo ld ways of putting problem scbefor e itself, so as to catalog the entire prelimi nary sketch of a whole system of metaphysics; because, regardi ng the first point, in a priori cogniti on nothing can be ascribe d to the objectsd except what the thinkin g subject takes out of itself, and regardi ng the second , pure speculative reason is, in respect of principlese of cogniti on, a tIDity entirely separat e and sisting for itself, in which, as in an organiz ed body, every exists the sake of all the others as all the others exist for its sake, and no ciplef can be taken with certain ty in one relation unless it h3S at tlle

B xxii


* In the same way, the central laws of the motion of the heavenly bodies estab- B xxii lished with certainty what Copernicus assumed at the beginning only as a hypothesis, and at the same time they proved the invisible force (of Newton ian attraction) that binds the universe,t which would have remained forever undiscovered if Copernicus had not venurred, in a manner contradictory to the senses yet true, to scck for the observed movements not in the objects of the heavens but in their observer. In this Preface I propose the transfor mation in our way of thinking presented in criticismh merely as a hypothe sis, analogous to that other hypothesis, only in order to draw our notice to the first attempts at such a transformation, which are always hypothetical, even though in the treatise itself it will be proved not hypothetically but rather apodictically from the constiurtion of our representations of space and and from the elementary conccpts of the understanding.


Wcltbml in der Kritik, which book as awhole.

could also be translated "in the

" referring to the present



the second edition <B>

such procedures speculative reason has at least room an extension, even if it had to leave it empty; and we at mdeed we are called upon hy reason to fill it if we can through data of reason. * Now the concern of d1is critique of pure speculative reason consists in that attempt to transform the accepted procedure undertaking an entire revolution according to example geometers and natural scientists. It is a treatise on the method, not a system the science itself; but it catalogs the entire oudine d1e science metaphysics, both in respect of its boundaries and in respect of its entire internal structure. For {lure speculative reason has this about it, that it can and should measure its own capacity a according to the different ways for choosing the objects b of its thinking, and also completely enumerate the manifold ways of putting problems" before itself, so as to catalog the entire preliminary sketch of a whole system of metaphysics; because, regarding the first point, in a priori cognition nothing can be ascribed to the objects d except what the thinking subject takes out of itself, and regarding the second, pure speculative reason is, in respect of principlese of cognition, a unity entirely separate and subsisting for itself, in which, as in an organized body, every part exists for the sake of all the others as all the others exist for its sake, and no ciplef can be taken with certainty in one relation unless it has at the
* In the same way, the central laws of the motion of the heavenly bodies estab-



B xxu

lished with certainty what Copernicus assumed at the beginning only as a hypothesis, and at the same time they proved the invisible force (ofNewtonian attraction) that binds the universel which would have remained forever undiscovered if Copernicus had not ventured, in a manner contradictory to the senses yet mIe, to seek for the observed movcmentsnot in the objects of the heavens but in their observer. In this Preface I propose the transformation in our way of thinking presented in criticismh merely as a hypothesis, analogous to that other hypothesis, only in order to draw our notice to the first attempts at such a transformation, which are always hypothetical, even though in the treatise itself it will be proved not hypothetically but rather apodictically from the constitution of our representations of space and and from the elementary concepts of the understanding.

Objecte , Principien

in del' Kritik, which could also be translated "in the Critique," referring to the present book as a whole.


Preface re tho rou ghg oin g relation to the enti same tim e bee n investigated in its , une fort d goo rare ysics also has a (for use of pur e reason. But then metaph cts obje with do nce that has to enjoyed by no othe r rational scie king in general), which is that if by thin of form the log ic deals only with onto the secure course of a science, this critique it has bee n bro ugh t it re field of cognitions belo ngin g to then it can fully embrace the enti ciprin a as ty teri lay it dow n for pos and thus can com plet e its wor k and BXX lV rged, since it has to do solely with b enla be er nev pal fram ewo rk that can r use, whi ch are dete rmi ned by the principlesc and the limitations on thei is fundamental science, metaphysics principles themselves. Hen ce as a of say to able be t eness, and we mus also bou nd to achieve this complet t agendum. d resse supe quid it: nil aetum reputans, si sure is it that we inte nd to leave trea of sort at But it will be asked: Wh ysics that has bee n purified through to posterity, in the form of a metaph into a changeless state?' On a cursory criticism but ther eby also bro ugh t eve that one perceives it to be only overview of this work, one mig ht beli on er to ven ture with speculative reaS of neg ativ e utility, teac hing us nev ulusef first its is nce; and in fact that bey ond the bou nda ries of experie e s pos itiv e whe n we bec ome awar ome bec n ness. But this utility soo its ond bey s ture ulative reas on ven that the principles with whi ch spec ext end ing our use of reason, but in lt resu bou nda ries do not in fact e closely, inevitably resu lt in narr owrather, if one considers them mor bou nda ries of sensibility, to which ing it by thre aten ing to extend the ond everything, and so even to disthes e principles really belong, bey BXX V on. Hen ce a critique that limits the lodg e the use of pur e (practical) reas besure, to that exte nt neg ativ e, but speculative use of reason is, to be atthre even or ts obstacle that limi cause it simultaneously removes an of fact in also is que reason, this criti ens to wipe out the practical use of oured vinc con e hav as soo n as we pos itiv e and very imp orta nt utility, essary practical use of pur e reason nec ly lute abso selves that t.~ere is an voidably extends itse lf beyond the (the mor al use), in which reas on una nee ding any assistance from speculabou nda ries of sensibility, with out be mad e secure against any countive reason, but in which it mus t also not to fall into con trad ictio n with tera ctio n from the latter, in ord er
" Objecte

tuhl, a loom or to be dravm from weaving (cf. Webs HauptITuhl; Kant's meta phor seems frame for weaving). c Principic1'l The corre ct quotation is: if some thing more is to be done ." d "Thi nkin g noth ing done t

insta eredens, cum quid supercslet "Caesar in omnia praeceps, nil anum some thing more ree whil done ing noth ving belie (Caesar, head long in everything, ' fiercely) (Lucan, De bello civili 2 :657) main ed to be done, pressed forward , bcbi7J7-1ichen Zustand



the second edition <B>

itself. deny this service of criticis mU is of any positiv e would be as much as to say that the police are of no positive cause their chief business is to put a stop to the violenc e that citizens have to fear from other citizens, so that each can carry on his own affairs in peace and safety. '7 In the analytical part of the it is proved that space and time are only forms of sensible intuitjoll1, therefore only conditi ons of the existen ce of the things as appeara nces, further that we have no concep ts of the unders tanding and hence no elements for the cogniti on of things except insofar as an intuitio n can be given corresp onding to these conrep ts, conseq uently that we can have cognition of no object as a thing in itself, but only insofar as it is an ject b of sensible intuitio n, i.e. as an appeara nce; from follows the limitation of all even possibl e speculative cogniti on of reason to mere objects of experie nce. Yet the reserva tion must also be noted, even if we cannot cogniz e these same objects as things in themse lves, we at least must be able to think them as things in themselves. * For otherwise there would follow the absurd propos ition that there is an pearance withou t anythin g that appears . Now if we were to assume the distinction betwee n things as objects of experie nce and the very same things as things in themselves, which our critique has made necessary, were not made at all, then the princip le of causality, hence the mechan ism of nature in determ ining causality, would be valid of all things in general as efficien t causes. I would not be able to say of one and the same thing, e.g., the human soul, that its will is free and yet it is simultaneously subject to natural necessity, i.e., that it is not free, without falling into an obviou s contrad iction; because in propos itions I would have taken the soul in just the same meanin g,c namely as a thing in general (as a thingd in itself), and withou t prior critique ,I
* To cognize an object, it is required that I be able to prove its possibility



(whether by the testimony of experience from its actuality or a through reason). But I can think whatever I like, as long as I do not contradict myself, i.e., as long as my concept is a possible thought, even if I cannot give any assurance whether or not there is a corresponding object'somewhere within the sum total of all possibilities. But in order to ascribe objective validity to such a concept (real possibility, for the first sort of possibility was merely logical) something more is required. This "more," however, need not be sought in theoretical sources of cognition; it may also lie in practical ones.


" der KYitik



"meaning" will translate this word for the remainde r of this paragrap h.

, Object


Preface not erre d in critique if not mea ning , take n in a teac hing in itsc1f;'8 if its ded ucti on of the pure nam ely as appearance or as of correct, and hen ce the principle concepts of the und erst and ing is far nam ely inso to thin gs take n in the first sense, causality applies nd meaning seco in gs thin experience, while are objects as the appearin of t ugh tho is same just are not subject to it; natu re and of the to ject ily sub ance (in visible actions) as necessar 3XXV lll thou ght of is it d han r othe on the to this exte nt not free , while yet not subject to that law, and hence as belo ngin g to a thin g in itself as by occ urri ng. Now alth oug h I cantree , with out any con trad ictio n here the latte r side, thro ugh any specnot cog nize my soul, considered , and hence I less thro ugh empirical observation) ulative reas on ch I ascribe whi to g bein pert y of any can not cog nize free dom as a pro nize such cog to e hav ld sense, because u'1en I wou effects in the (which is time in ned rmi not as dete an existence as dete rmi ned , and yet ), nevition intu any with t my con cep impossible, since I can not sup por t of it at n tatio esen repr myself, i.e., the ertheless, I can thin k free dom to ncdisti cal criti our as itself, so long least contains no con trad ictio n in intel and e sibl (sen s of repr esen ting tion prevails betw een the two way runde the of ts cep of the pur e con lectual), along with the limi tatio n that of the principles flowing from ce hen and stan ding arising from it, necessarily presupposes freedom (in them . Now suppose that mor ality our will, citing a priori as data for the strictest sense) as a pro pert y of n, tical principles lying in our reaso this freedom cert ain original prac of ion osit upp pres ble with out the whi ch would be absolutely impossi be not can dom free had prov ed that freedom, yet that speculative reas on B XXlX ion, namely the mor al one, would osit upp pres tho ugh t 3t all, then that r one , whose opposite cont3ins an necessarily have to yield to the othe y free dom and with it morality (for obvious contradiction; con sequ entl wer e not already con tain no con trad ictio n if freedom the latte r of natu re. But have to give way to the mec han iID presupposed) dom should free ing mor e than that then , since for mor ality I need noth it should that at least be thin kab le not con trad ict itself, that it sho uld the same in ure mec han ism of nat place no hind ranc e in the way of the me to for y ssar with out it bein g nece action (taken in ano ther relation), place its rts asse doc trin e of mor ality have any furt her insi ght into it: the ochave not ld however, wou doc trin e of natu re its own, which, and e ranc igno ht us of our unavoidable curr ed if criticism had not first taug we that ing es and limited everyth in resp ect of the thin gs in themselv earances. Jus t the same sort of exapp e can cog nize theo reti call y to mer can be of critical principles of pur e reason pos itio n of the positive



to the second edition <B>

given in respect to the concepts of God of the simple nature our soul, which, however, I forgo for the sake of brevity. I cannot even assume God, freedom and immortality for the sake of the necessary practical use of my reason unless I simultaneously deprive speculative reason of its pretension to extravagant insights; because in order to attain to such insights, speculative reason would have to itself to principles that in fact reach only to objects of possible experience, which, if they were to be applied to what cannot be an object of ence, then they would always actually transform it into an appearance, and thus declare all practical extension of pure reason to be impossible. Thus I had to deny ~owledge in order to make room for and the dogmatism of metaphysics, i.e., the prejudice criticism reason can make progress in metaphysics, is the true source of all unbelief conflicting with morality, which unbelief is always very dogmatic. - Thus even if it cannot be all that difficult to leave to posterity the legacy of a systematic metaphysics, constructed according to the crito tique of pure reason, this is still a gift deserving of no small see this, we need merely to compare the culture of reason is set on the course of a secure science with reason's unfounded groping and frivolous wandering about without critique, or to consider how much better young people hungry for knowledge might spend their time in the usual dogmatism that gives so early and so much encouragement to their complacent quibbling about things they do not understand, and things into which neither they nor anyone else in the world will ever have any insight, or even encourages them to launch on the In\TPntlcm of new thoughts and opinions, and thus to neglect to learn the wellgrounded sciences; but we see it above all when we take account of the way criticism puts an end for all future time to objections against morality and religion in a Socratic way, namely by the proof of the ignorance of the opponent. For there has always been some metaphysics or other to be met with in the world, and there always continue to be one, and with it a dialectic of pure reason, because dialectic is natural to reason. Hence it is the first and most important occupation of philosophy to deprive dialectic once and for all of all disadvantageous influence, by blocking off the source of the errors. with With this important alteration in the field of the sciences, the loss of its hitherto imagined possessions that speculative reason must suffer, everything yet remains in the same advantageous state as it was before concerning the universal human concern and the the world has so far drawn from the doctrines of pure reason, the loss touches only the monopoly of the schools and in no way the interest of human beings. I ask the most inflexible dogmatist whether the proof of the continuation of our soul after death drawn the simplicity of substance, or the proof of freedom of the against uni117




Preface the subtle though powerless distinctions versal mechan ism objective practical necessity, or the proof of the betwee n subjective existence of God drawn from the concep t of a most real being (or from , the conting ency of what is alterable and the necessity of a first mover) pubthe reach to able been , schools the in have ever, after origina ting lic or have the least influence over its convictions? If that has never hapthe pened, and if it can never be expected to happen, owing to subtle such for tanding unders unsuitability of the commo n human speculation; if rather the conviction that reaches the public, insofar ing as it rests on rational grounds, had to be effected by someth osipredisp able remark that on point, first else - namely, as regards the tion of our nature, noticeable to every human being, never to be capable of being satisfied by what is tempor al (since the tempor al is always the insufficient for the predispositions of our whole vocation) leading to exclear mere the point, second the hope of a future life; in respect of leadtions inclina the of claims an to ion positio n of our duties in opposit ing to the consciousness of freedom; and finally, touchin g on the third point, the splendid order, beauty, and provide nce shown forth everythe where in nature leading to the faith in a wise and great author of even it but rbed, undistu s remain world - then this possession not only to gains in respect 1:l~ough the fact that now the schools are instruc ted point any on insight hensive pretend to no higher or more compre touchin g the universal human concerns than the insight that is accessi, ble to the great multitu de (who are always most worthy of our respect) proof of s ground those of ion cultivat and to limit themselves to the alone that can be grasped universally and are sufficient from a moral of standpo int. The alterati on thus concerns only the arrogan t claims exsole the for taken be lves themse the schools, which would gladly let perts and guardians of such truths (as they can rightly be taken in many other parts of knowledge), sharing with the public only the use of such truths, while keeping the key to them for themselves (quod mecum nescit, the solus vult scire videri). a Yet care is taken for a more equitable claim on of trustee e exclusiv the s remain He part of the speculative philosopher. ge, knowled their t withou even a science that is useful to the public namely the critique of reason; for the latter can never become popular, to but also has no need of being so; for just as little as the people want litas just so truths, useful for nts fill their heads with fine-sp un argume tle do the equally subtle objections against these truths ever enter their as minds; on the contrary, because the school inevitably falls into both, reason of critique the tion, specula does everyone who raises himself to



The correct quobl"Vilhat he knows no more than T, he alone wants to seem to know." to me, that alone unknown is (What videri" scire volt solus 771CC1I771 lion is "Quod (Horace, Epistles 2.1.87). he wants to seem to


to the second edition <B>

is bound once and for to prevent, by a fundamental investigation to the rights of speculative reason, the scandal that sooner or be noticed even among the people in the disputes in which, in the absence of criticism, metaphysicians (and among these in the end even clerics) inevitably involve themselves, and in even falsify their own doctrines. Through criticism alone can we sever the very root of materialism, fatalism, atheism, of freethinking unbelief, of enthusiasm and superstition, which can become g-enerallv injurious, and finally also of idealism and skeptidsm, which are more If dangerous to the schools and CRn hardly be transmitted to the governments find it good to concern themselves the scholars, then it would accord better with their wise solicitude for the sciences and for humanity if they favored the freedom of such a critique, by which alone the treatments of reason can be put on a firm footing, instead of supporting the ridiculous despotism of schools, which raise a loud cry of public danger whenever someone tears apart their cobwebs, of which the public has never taken any notice, and hence the loss of which it can also never feel. Criticism is not opposed to the dogmatic procedure of reason in its pure cognition as science (for science must always be dogmatic, it must prove its conclusions strictly a priori from secure principles)O; rather, it is opposed only to dogmatism, i.e., to the presumption of getting on solely with pure cognition from (philosophical) concepts according to principles,b which reason has been using for a long time without first inquiring in what way and by what right it has obtained them. Dogmatism is therefore the dogmatic procedure of pure reason, without an antecedent critique of its own capadty.cThis OPPOSltJlon therefore must not be viewed as putting in a good word cious shallowness under the presumed name of popularIty, skepticism, which gives short shrift to metaphysics; is the preparatory activity necessary for the advancement of metaphysics as a well-grounded science, which must necessarily be dogmatic, carried out systematically in accordance with strictest requirement, hence according to scholastic rigor (and not in a popular way); for this requirement is one that it may not neglect, since it undertakes to carry out its business wholly a priori and thus to the full satisfaction of speculative reason. In someday carrying out the plan that prescribes, i.e., in the future system of metaphysics, we will have to follow the strict method of the famous Wolff, the greatest among all dogmatic philosophers, who gave us the first example ex-





Preface a spir it of well-groundedness became the which amp le the secure not extinguished) of the way in in Ger man y that is nme nt of rtai asce lar thro ugh the regu course of a science is to be taken, mpt at atte the ts, cep nati on of con the principles,a the clear dete rmi inferin s leap us acio aud the prev enti on of strictness in the proofs, as such nce scie a ing the skills for mov ences; for these reasons he preto him to d urre occ only it metaphysics into this con diti on, if ely pur e reason itself: nam n, orga the of ique crit pare the field for it by a ic way as not so muc h to him to the dog mat :xxxvii a lack that is to be charged his as of rs phe oso phil in his age; and for this the of thin king es. selv them h oac repr to for of all previous times have noth ing eproc the usly neo of teac hing and simulta Tho se who reject his pt exce d min in else ing can have noth dure of the crit ique of pur e reason k wor m sfor tran to altogether, and to thro w off t.J.,.e fetters of scie nce xy. odo phil phil oso phy into cert aint y into opinion, into I have wanted, as is only proper, , ion edit ond Con cer nin g this sec ove as far as possible thos e difficulnot to forgo the opp ortu nity to rem ermay have spru ng several misund ties and obscurities from which my on t faul e som aps not with out standings into which acute men, perh to ing noth d foun e this book. I hav part , have fallen in thei r judg men t f, proo of nds grou selves or in thei r alte r eith er in the prop osit ions them ibed ascr be to is hook's plan; this or in. the form and completeness which I subjected them prior to to tiny scru of part ly to the long peri od ly to the con stitu tion of the matter laying it before the public; and part whi ch contains natu re of a pur e speculative reason, itself, namely to g is an organ, thin bers in which each a trul y articulated stru ctur e of mem and each ber, mem the sake of each that is, in which eve ryth ing is for t frailty, leas the even so that of individual mem ber is for the sake nXXXV111 ay itself betr ly itab inev or a lack, mus t whe ther it be a mistake (an erro r) this unin lf itse n ntai hen cefo rth mai in its use. I hop e this system will rather but , this in t justifies my trus alterability. It is not self-conceit that result the that g win exp erim ent sho mer ely the evidence draw n from the to ents elem t lles sma cee d from the effected is the same whe ther we pro this (for t par y ever to from the whole the whole of pur e reason or retu rn the final inte ntio n of pur e reason in ugh thro lf itse whole too is given in alte r even the smallest part directly the practical); while the atte mpt to ely into the system, but into univerintr odu ces con trad ictio ns not mer tati on ther e is still muc h to do, and sal hum an reason. Yet in the pre sen make imp rove men ts in this edition, here is whe re I have atte mpt ed to , mis und erst and ing of the Aesthetic whi ch sho uld remove first, the the in ty curi obs the time; second, chiefly the one in the con cep t of Und erst and ing, nex t the supposed the of ts cep Ded ucti on of the Con


to second edition <B>

lack of sufficient evidence in proofs Understanding, and finally the misinte rpretati on the paralogisms vanced against rational psychology. revisions '9 of the mode of sentation* extend only to this point (namely, to the end of the chapter of the Transc endent al Dialectic) because time


* The only thing I can really call a supplement, and that

in the is what I have said at [B ] 2 73 in the form of a new refutation of pS]/ch!ol ()gical idealism, and a strict proof possible one, I believe) reality of outer intuition. No matter how innocen t idealism may be held to be as regards the essential ends of metaphysics (though in fact it is not so innocent), it always remains a scandal of philosophy and universal human reason that the existence of things outside us (from which we after all get the whole matter for our cognitions, even for our inner sense) should have to be assumed merely on faith, and that if it occurs to anyone to doubt it, we should be unable to answer him with a satisfactory proof. Because there are some obscurities in li~e expressions of this proof between the third and sixth lines, I ask leave to alter this passage as follows: "But this persisti ng elemen t cannot he an intuitio n in me. For all the detenn ining ground s of my existenc e that can be encoun tered in me are represe ntation s, and as such themse lves need someth ing persisti ng distinct from them, in relation to which their change, and thus my existenc e in the time in which they change, can be detenni ned." Against this proof one will perhaps say: I am immediately conscious to myself only of what is in me, i.e., of my represe ntation of external things; consequently it still remains undecided whether there is somethi ng outside me corresponding to it or not. Yet I am conscious through inner experience of my existen ce in time (and consequently also of its determi nability in time), and this is more than merely being conscious representation; yet it is identical with the empiric al conscio usness of existenc e, which is only determinable through a relation to somethi ng that, being bound up with my existence, is outside me. This consciousness of my existenc e in time is thus bound up identically with the consciousness of a relation to something outside me, and so it is experience and not fiction, sense and not imagination, that inseparably joins the outer with my inner sense; for outer sense is already in itself a relationa of intuition to somethi ng actnal outside me; and its reality, as distinct from imagination, rests only on the fact that it is in~;ep'ar: lbly bound up with inner experience itself, as the condition of its possibility, happens here. In could combine a determi nation of my existence through intellectual intuitio n simultaneously with the intellec tual conscio usness of my existence, in the representation I am, which accompanies all my judgmen ts and actions ofmy understanding, then no consciousness of a relation b to something outside me would necessarily belong to this. But now that intellect ual consciousness does to be sure precede, but the inner intuition, in which alone




Preface Bxl Bxli Bxlii the book no misund erwas too short, and also in respect of the rest ers has come my examin standin g on the part of expert and imparti al due to them; praise the with way, whom I have not been able to name to them evident be will ers remind but the attentio n I have paid to their up bound is r, howeve ement, in the approp riate passage s. This improv against d guarde be not could with a small loss for the reader, which that withou t making the book too volumi nous: namely , various things to had whole the of are not essentia lly require d for the comple teness that fact the despite , fashion be omitted or treated in an abbrevi ated still some readers may not like doing withou t them, since they could for room make I could way be useful in anothe r respect ; only in this enfundam which ation, what I hope is a more compre hensibl e present or even tally alters absolut ely nothing in regard to the propos itions s edipreviou the from far so their ground s of proof, but which departs d manage be not could it tion in the method of present ation that be can case any in which throug h interpo lations . This small loss, second compen sated for, if anyone likes, by compar ing the first and n of my existence can be determined, is sensible, and is bound to a conditio depends itself, ce experien inner hence time; however, this determination, and be outon something permanent, which is not in me, and consequently must ce in experien an for thus it; to relation" in myself r conside must I and me, side with up bound ily necessar is sense general to be possible, the reality of outer things are there that s consciou certainly as just am I u~at of inner sense, i.e., exist outside me to which my sensibility relates, as I am conscious that I myself outer to nd correspo determined in time. Now which given intuitions actually be asobjects, which therefore belong to outer sense, to which they are to particueach in decided be must that tion cribed rather than to u'l.e imagina inner lar case according to the rules through which experience in general (even re is procedu experience) is to be distinguished from imagination; which To ce. experien outer is actually grounded always on the proposition that there perng somethi of tation represen The added: be this the following remark can for that sisting in existence is not the same as a persisti ng represe ntation ; the can be quite variable and changeable, as all our representations are, even nt, permane ng somethi to representations of matter, while still being related exand tations represen my all from distinct thing a which must therefore be of nation detenni the in included ily necessar is which of e existenc ternal, the which ce, experien my own existence, which with it constitutes only a single outer. could not take place even as inner if it were not simultaneously (in part) how further explain can we than d explaine more no The "How?" of this can be what with eity simultan whose time, in abides what of all at we can think changes is what produces the concept of alteration.




to the second edition <B> editions, is, as I hope, more than compensated for greater comprehensibility. In various public writings (partly in reviews some books, partly in special treatises) I have perceived with and enjoyment that the spirit of well-groundedness has not died out in Germany, but has only been drowned out for a short time the ionable noise of a freedom of thought that fancies itself ingenious, I see that the thorny paths of criticism, leading to a science of pure reason that is scholastically rigorous but as such the lasting and hence the most necessary science, has not hindered courageous and comclear minds from masteringAhem. To these deserving men, bine well-groundedness of insight so fortunately with lucid presentation (something I am conscious of not myself), leave it to complete my treatment, which is perhaps defective here there in this latter regard. For in this case the danger is not that I be refuted, but that I will not be understood. For my own now on I cannot let myself become involved in controversies, altho,ug:h I shall attend carefully to all hints, whether they come from friends or from opponents, so that I may utilize them, in accordance with propaedeutic, in the future execution of the system. Since these labors I have come to be rather advanced in age (this I will attain my sixty-fourth year), I must proceed frugally with my time ifI am to carry out my plan of providing the metaphysics both of nature and both of of morals, as confirmation of the correctness of the theoretical and practical reason; and I must await the illumination of those obscurities that are hardly to be avoided at beginning of work, as well as the defense of the whole, from those deserving men who have made it their own. Any philosophical treatise may find itself under pressure in particular passages (for it cannot be as as a mathematical treatise), while the whole structure of the system, considered as a unity, proceeds without the least danger; when a system is new, few have the adroitness of mind a to gain an overview of and because all innovation is an inconvenience to them, still fewer have the desire to do so. Also, in any piece of writing apparent contradictions can be ferreted out if individual passages are tom out of their context and compared with each other, especially in a piece of informal discourse b that in the eyes of those who rely on the judgment of others cast a disadvantageous light on that piece of writing but that can be very easily resolved by someone who has mastered the idea the whole. Meanwhile, if a theory is really durable, then in time the effect

B xliii

" Geist b als freie Rede fortgrhC71dC71 Schrift


Preface grea t at first seem ed to thre aten it reac tion , of acti on imof men if to polish away its rou gh spots, danger, will serve this, do to s ines bus r mak e it thei true partiality, insight, eve n the requ ired elegance. uce prod then in a sho rt tim e

Kiinigsberg, in the month


Contents a


Introdu ction

page [I27]

I. Transce ndental Doctrin e of Elemen ts Part 1. Transc endenta l Aesthe tic Section I. On Space
Section no On Time Part no Transc endenta l Logic Divisio n I. Transc endenta l Analytic in two books and their various chapter s and section s

[15 1]
[153] [157]


Divisio n no Transc endenta l Dialect ic in two books and their various chapter s and section s


no Transcendental Doctrin e of Metho d

Chapte r Chapte r
10 20

[62 5]

The Discipl ine of Pure Reason The Canon of Pure Reason

[67 2 J

Chapte r 30 The Archite ctonic of Pure Reason Chapte r 40 The History of Pure Reason

Kant inclndes this table of contents only in the first edition.



1. The idea of transce ndenta l


Experience is withou t doubt the first produc t our understaw:lm.g brings forth as it works on the raw materia l of sensible sensations. I It is for this very reason the first teaching, and in its progres s it is so inexhaustible in new instruc tion that the chain of life in future generations will never have any lack of new inform ation that can be g-ath<:re d on this terrain. Nevert heless it is far from the only to our understanding can be restricted. It teUs us, to be sure, what is, never that it must necessarily be thus and not otherw ise For that very reason it gives us no true universality, and reason, which is so desirous of kind of cognitions, is more stimula ted than satisfied by it. Now universal cognitions, which at the same time have the charact er necessity, must be clear and certain for themselves, indepe ndently of experience; hence one calls them a priori cognitions: whereas that which is merely borrow ed from experience is, as it is put, cognize d only a posor empiric ally)

A 2

, We first present the introduc tion as it appeared in the first edition, followed the revised version that appeared in the second edition. Considerable changcs were made in the latter, including some deletions, major additions, and occasional alterations within the p~ssagcs that were repeated. We will use notes and references to the marginal pagination to show what changes were made from the first to the second editions. The following two paragraphs in the first edition were replaced with the first two numbere d sections of the second. b In his copy of the first edition, Kant made the following two notes: "r. On the possibility of a critique of pure reason. 2. On its necessity (not from other sciences). 3. On its division. 4. On its purpose, the science of all principles of pure reason. (Practical)" (E I, p. rz). ' "That reason has its boundari es with regard to its a priori principles concerning both degree and scope. Division of metaphysics into metaphysics of nature and of morals" (E n, p. rz). , The following note is added in Kant's copy of the first edition: "We cannot infer to any necessity a po.rtcriori if we do not already have a mle a p,-ior'i. E.g., 'If many cases are identical, there must be somethin g that makes this agreeme nt presupposes the a priori propositi on that everything continge nt has a cause that determines its concept a priori." (E Iv, p. 14)


Intr odu ctio n <A>

is that even amo ng our experiNow wha t is especially rem arka ble mus t have thei r orig in a priori and ences cog niti ons are mixed in that repr esen to establish con nec tion amo ng our that perh aps serve oves from our experiences everytatio ns of the senses. For if one rem ther e still rem ain cert ain original thin g that belo ngs to the senses, whi ch mus t have the judg men ts gen erat ed from con cep ts , bec ause they nce erie exp den tly of aris en enti rely a priori, inde pen to the senses ear app that the obje cts mak e one able to say mor e abo ut believe that one e mak t leas h, or at than mer e experience wou ld teac lity and ersa univ true tain con rtions one can say this, and mak e asse on can niti cog l irica emp mer ely stri ct necessity, the likes of nev er afford. cert ain cog niti ons even ahanBut wha t says still mor e is this, that B6 expand the dopossible experiences, and seem to don the field of A3 nce through erie exp hou nds mai n of our judg men ts bey ond be given in can all at ing obje ct con cep ts to whi ch no corr esp ond exp ene nce . ns, whi ch go bey ond the world And precisely in these latt er cognitio give neit her guidance nor correcof the senses, whe re experience can e reas on that we hold to be far mor tion, lie the investigations of our than aim l fina r thei and sublime in pree min ent in thei r imp orta nce B7 can lear n in the field of appearances, ing and erst und eve ryth ing that the ture everything, even at the risk of and on whi ch we wou ld rath er ven nt investigations because of any sort erri ng, than give up such imp orta a indifference. of rese rvat ion or from con tem pt and n as one has aba ndo ned the terNow it may seem natu ral that as soo imm edia tely erec t an edifice with rain of experience, one wou ld not out kno win g whence, and on the cog niti ons that one possesses with one does not know, with out having cred it of principles who se orig in s, tion thro ugh careful investigation first assured one self of its fou nda unthe how n stio que the e raised thus that one wou ld have long sinc cog niti ons a priori and what doe thes all to e com ders tand ing could more mig ht have. And in fact noth ing is A4 mai n, validity, and value they and erly prop ch whi that d this wor natu ral, if one und erst and s by h whic that it by s and erst und if one reasonably oug ht to hap pen ; but B8 not hin g is mor e natu ral and compreusually happens, then conversely n sho uld long have bee n neglected. hen sibl e than that this investigatio the mat hem atic al, has long been reFor one par t of these cognitions, rs a favorable expectation abo ut othe liable, and ther eby gives rise to Furre. natu t eren diff rely an enti as well, alth oug h thes e may be of

of pure reason. See sentences char acter izing the tasks Here the second editi on adds two B7 below.


Introduction <A> thermore, if one is beyond the circle of experience, one is sure not to be contradicted through experience. The charm in expanding one's cognitions is so grcat that one can be stopped in one's progress bumping into a clear contradiction. This, however, one can one makes his inventions carefully, even though are not thl:re:by inventions any the less. Mathematics gives us a splenlclld eX~lmiple how far we can go with a priori cognition independently of experience. Now it is occupied, to be sure, with objects and cognitions so far as these can be exhibited in intuition. This circumstance, IS easily overlooked, since the intuition in question can be a and thus can hardly be distinguished from a mere pure C011cc:pt. Encouraged by such a proof of the power of reason, the drive for expansion sees no bounds. The light dove, in free flight ('11J-1-1n,O' tllrough the air the resistance of which it feels, could get the ideaa do even better in airless space. Likewise, Plato abandoned the senses because it posed so many hindrances for the understanding, and dared to go beyond it on the wings of the ideas, in space of pure understanding. He did not notice that he made no he:ldvvay his efforts, for he had no resistance, no support, as it were, he could stiffen himself, and to which he could apply his powers in order to get his understanding off the ground. It is, however, a cw;tomalrv fate of human reason in specul::Jtion to finish its edifice as as possible and only then to investigate whether the ground has been But at that point all sorts of excuses will adequately prepared for be sought to assure us of its sturdiness or to refuse such a late dangerous examination. What keeps us free of all worry and cion during the construction, however, and flatters us with appa:reIlt thoroughness, is this. A great part, perhaps the greatest business of our reason consists in analyses of the concepts we already have of objects. This affords us a multitude of cognitions though they are nothing more than illuminations or of that which is already thought in our concepts (though still in a confused way), are, at least as far as their form is concerned, treasured as if they were new insights, though they do not extend the concepts that we have in either matter or content but only set them apart from each other. Now since this procedure does yield a real a priori cognition, which makes secure and useful progress, reason, without itself noticing it, under these pretenses surreptitiously makes assertions of quite another sort, in which it adds something entirely alien to given concepts a priori, without one knowing how it was -able to do this and without this question even being allowed to come to mind. I therefore deal


B 10


Introduction <Av with the distinct ion betwee n these two outset. a of cogniti on right at the

On the differe nce betwee n analyti c and synthe tic judgme nts. 4 judgme nts in which the relation of a subject to the predicate is In affirmative judgme nts, since the application though t (if! conside r to negative ones is easy), this relation is possible in two differe nt ways. is Either the predica te B belong s to the subject A as sOlTlething that the outside (covertly) contain ed in this concep t A; or B lies entirely sure it stands in connec tion with it. In the first concep t A, though to case I call the judgme nt analytic, in the second synthetic. Analytic of judgme nts (affirmative ones) are thus those in which the connec tion A7 conthe predica te is though t throug h identity, but those in which this nection is though t withou t identity are to be called synthet ic judgand ments. One could also call the former judgme nts of clarification E I I the te the latter judgme nts of amplification,b since throug h the predica only former do not add anythin g to the concep t of the subject, but break it up by means of analysis into its compo nent concepts, which the were already though t in it (though confusedly); while the latter, on not was that te predica a subject the of t concep the contrary, add to and could not have been extracted from it through though t in it at 1 say: "All bodies are extended," then this is an anif e.g., ; any analysis I alytic judgme nt. For I do not need to go outside the concept" that concombin e with the word "body" in order to find that extension is benected with it, but rather I need only to analyze that concep t, i.e., to come conscious of the manifo ld that 1 always think in it, in order t. judgmen encoun ter this predica te therein ; it is therefo re an analytic is te On the contrary, if I say: "All bodies are heavy," then the predica consometh ing entirely differe nt from that which I think in the mere cept of a body in general. The additio n of such a predica te thus yields a synthet ic judgme nt. dNow from this it is clear: I) that throug h analytic judgments our cogniti on is not amplified at all, but rather the concept, which I already
Kant's copy of the first edition has the following note: categorical ncg3tive "On synthetic hypothet ical and disjunctive judgmen ts as well as judgmen ts." (E V, p. 14) not and En,citcrlI71j4s1I77;i,ile. These terms are emphasized in the second but b in the first edition. t; 'A body ex, Kant's copy of the first edition here adds: " 'I exist' is an analytic judgmen ists' is a synthetic one." (E Vl, p. 14) single one in the second edition, the secd The next two paragrap hs are replaced with a ond of which incorpor ates part of the present one; see B I 1-12 below.



Introduction <A> have, is set out, and made intelligible to me; 2) in SV11lth(~tic ments I must have in additio n to the concep t of the subject sOlueth lngelse (X) on which the unders tanding depends in cognizi ng a predica te that does not lie in that concep t as nevertheless belong ing to it. a In the case of empirical judgme nts or judgme nts of experience is no difficulty here. b For this X is the comple te experience that I think throug h some concep t A, which constitu tes this experience. For althoug h C I do not at all include the of B 12 weight in the concep t of a body in general, the concep t nevertheless designates the comple te experience throug h a part of it, to I can therefore add still other parts of the very same experience as belong ing to the former. I can first cognize the concep t of through the marks of extension, of impenetrability, of shape, etc., are all though t in this concept. But now I amplify my cogniti on looking back to the experience from which I had extracted concep t of body, I find that weight is also always connec ted previou s marks. d Experie nce is therefo re that X that lies outside the concep tA and on which the possibility of the synthesis of the predica te of weight B with the concep t A is ground ed. But in synthet ic a priori judgme nts this means of help is entirely lackA9 ing) If I am to go outside the concep t A in order to cognize BB 13 as combined with it, what is it on which I depend and throug h which the synthesis becomes possible, since I here do not have the advanta ge oflooking around for it in the field of experience? Take the proposi tion: "Everything that happen s has its cause." In the concep t of someth ing that happens, I think, to be sure, of an existence which was precede d by a time, etc., and from that analytic judgme nts can be But the concept of a cause indicates someth ing differen t from concep t of something that happen s, and is not contain ed in the tion at all. How then do I come to say someth ing quite differen t that which happen s in general, and to cognize the concep t of cause as belonging to it even though not contain ed in it?e What is the Xhere on which the unders tanding depends when it believes itself to discover beyond the concep t of A a predica te that is foreign to it is
" Kant's copy of the first edition adds here: "Analytic judgmen ts could accordingly be called mere judgmen ts of clarification, synthetic judgmen ts, however, judgmen ts of amplification." (E VII, p. 15) In Kant's copy of the first edition, this was changed to: "In the case of empirical judgments or judgmen ts of experience there is no difficulty about how they are to be proved synthetically." (E VID, p. 15) From here the remainde r of the paragrap h is incorpor ated into the second edition. d The remainder of this paragrap h is changed in the second edition; see BI2. , Kant ends this and the next sentence with periods, for which we have substitut ed question marks.


Introdu ction <A>

it? It cannot be experience, for the princip le that has connec ted latter represe ntation s to the former not only been adduced adds experience can provide, but also with thc than ity with greater general entirely a priori and from mere concepts. hence ty, necessi expression of aim of our speculative a priori cogniti on rests on Now the entire A IQ to such synu1.etic, i.e., ampliative, principles; for the analytic ones are, disthat g attainin for only but ry, necessa and ant be sure, most import tinctness of concepts that is requisite for a secure and extended synthesis as a really new constructioll. here,* the elucidation of which bA certain mystery thus lies field of pure cogniti on of the ess boundl the in s alone can make progres to uncove r the ground of the , namely : reliable and unders tanding secure approp riate generality, to with nts judgme possibility of synthet ic kind of them possible, every make that ons conditi gain insight into the (which comprises its on cogniti entire this te and not merely to designa it completely and ine determ to but , outline own species) in a cursory with its primary nce accorda in system a in adequately for every use provisionally for much So ries. bounda and sources, divisions, nts. judgme the pecularities of synthetic of this there results the idea of a special science, which cNow from B 24 could serve for the critique of pure reason. Every cogniti on is called AI I pure, however, that is not mixed with anythin g foreign to it. But a cogor nition is called absolutely pure, in particular, in which no experience is reason Now priori. a fully thus is that and sensation at all is mixed d Hence priori. a on cogniti of les princip the s the faculty that provide pure reason is that which contains the principlese for cognizing something absolutely a priori. An organo n of pure reason would be a sum total of those principles!"in accordance with which all pure a priori cogB 25

this alone * If it had occurre d to one of the ancients even to raise this question , down reason pure of systems would have offered powerful resistance to all the were that s attempt vain many so us spared have to our own times, and would issue. at really was what of ge knowled without ken blindly underta

Anhau, changed to EnDerb (acquisition) in the second edition. second edition, and The following paragrap h, including the footnote, is omitted in the B25. through BI4 VI, and V Sections replaced with editions resumes; in the second edition, howe At this point the common text of the two vn and the ensuing heading. In addinumber section the inserted here ever, there is and there are minor tion, the second and third sentences of this paragrap h are omitted, B24 below. See . sentences fourth and opening the of wording changes in the d Pri71dpien
f Principien


Introduction <l\> nitions can be acquired and actually broug'ht about. plication of such an organon would create a system of pure reason. But whether such an since that requires a lot, and it is still an open cases it amplification of our cognition is possible at all and in would be possible, we can regard a science of the mere estimation pure reason, of its sources and boundaries, as the propaedeutic to system of pure reason. Such a thing would not be a doctrine, must be caned only a critique of pure reason, and its be only negative, serving not for the amplification but cation of our reason, and for keeping it free of errors, deal is already won. I call cognition transcendental is occupied not so much with objects but rather with our a priori concepts of objects in general. a ,6 A system of such concepts would be caned transcendental philosophy. But this is again too much for the beginning. For since a science would have to contain completely both analytic as as synthetic a priori cognition, it is, as far as our aim is concerned, too in scope, since we need to take the analysis only as far as is m<l1slpellsably necessary in order to provide insight into the of a priori synthesis in their entire scope, which is our only concern. This investigation, which we can properly call not doctrine transcendental critique, since it does not aim at the amplification cognitions . themselves but only at their correction, and is to supply the touchstone of the worth or worthlessness of all cognitions a priori, is that with which we are now concerned. Such a critique is accordingly a nrpnClr,,_ cion, if possible, for an organon, and, if this cannot be ac<:ornp!isihelj, then at least for a canon, in accordance with which the complete of the philosophy of pure reason, whether it is to consist in the fication or the mere limitation b of its cognition, can in any case at least some day be exhibited both analytically and For that should be possible, indeed that such a system should not be too great in scope for us to hope to be able entirely to complete can be assessed in advance from the fact that our object is not nature of which is inexhaustible, but the understanding, which judges nature of things, and this in turn only in regard to its a priori cognition, the supply of which, since we do not need to search for it ~~,.~~U'LUY, cannot remain hidden from us, and in all likelihood is small enough to be completely recorded, its worth or worthlessness assessed, jected to a correct appraisaJ.e
, In the second edition, "but ..." replaced with "but with our manner of cognition of objects insofar as this is to be possible a priori." See B25 below.




. Two sentences are added here in the second edition; see




Introduction <A> phy " Div isio n of Tra nsc end enta l Phi loso only an idea,b for whi ch the criTra nsce nde ntal phil oso phy is here enti re plan arch itec toni call y, i.e., tiqu e of pur e reason is to outl ine the ee for the completeness and cerfrom principles,' with a full gua rant d comprise this edifice. Tha t this critain ty of all the com pon ents that scen den tal phil oso phy rests solely tiqu e is not itself already called tran it wou ld also have to in ord er to be a com plet e system on the fact all of hum an cog niti on a priori. Now con tain an exhaustive analysis of of befo re us a com plet e enu mer atio n sure, our crit ique must, to ques in on niti cog e e pur comprise the all of the ancestral concepts that cone thes of ysis anal the exhaustive tion. Onl y it pro perl y refrains from com plet e review of all of those the from as l cepts themselves as wel this analysis would not be purposederived from them , part ly because d in the tain the difficulty that is enc oun tere :fIB 28 full since it does not con ken, erta und who le crit ique is actually synthesis on acco unt of which the on take to y to the unit y of the plan part ly because it would be con trar on, vati of such an analysis and deri responsibility for the completeness relieved given one's aim. Thi s com from which one could afte r all be the deri vati on from the a priori conpleteness of the analysis as well as to the future will nevertheless be easy cepts whi ch are to be provided in g as exhaustive principles of synthecom plet e as long as they are pres ent in rega rd to this essential aim. sis, and if noth ing is lacking in them e accordingly belo ngs everything To the critique of pur e reason ther osophy, and it is the complete idea that constitutes tran scen den tal phil it is not yet this science itself, since of tran scen den tal philosophy, but aestim e plet com is requisite for the goes only so far in the analysis as n. tion of synthetic a priori cognitio such a science is that absolutely no The chie f targ et in the division of tain any thin g empirical, or that the concepts mus t ente r into it that con Hen ce, alth oug h the supreme prina priori cog niti on be enti rely pure.

replaced by the in the second editi on, having been This num ber and title are omit ted B 24. num ber and title of Sect ion VII at idea of a sciced in the seco nd editi on with "the repla are idea" an only e "her s b The word ence"; see B 2 7 below.

of all the sente nce "It is the system Here the second editi on inser ts d here: "For adde had t Kan on, editi first the his copy of fPrin n;nir nl of pure reason." In entirely fore there and e, be with out any touc hston with out this the form er mus t also groundless." (E IX, p. IS)

f zcllcckm,iiJ!ig


Introduction <A> ciples of moralit y and the fundamental concepts of it are cognitions, they still do not belong in transce ndental since the concepts of pleasure and displeasure, of desires and inclinations, of choice, etc., which are all of empirical origin, must there be presupposed. a Hence transce ndental philoso phy is a philosophyb of pure, merely speculative reason. For everyth ing practical, insofar as it contains motives," is related to feelings, which belong among emlpllnC :ll sources of cogniti on. Now if one wants to set up the division of this science from the general viewpoint of a system in general, then the one that we now present must contain first a Doctri ne of Eleme nts and second a Doctri ne of Metho d of pure reason. Each of these main parts have its division, the ground s for which cannot yet be expounded here. that seems necessary for an introdu ction or a prelimi nary is that there are two stems of human cognition, which may perhaps arise a common but to us unknow n root, namely sensibi lity and unders tandin g, through the first of which objects are given to us, but throug h second of which they are though t. Now if sensibility were to contain a representations, which constit ute the conditions which objects are given to us, it would belong to transcendental philosophy. transcendental doctrin e of the senses will have to belong to the first part of the science of elements, since the conditions under which alone the objects of human cogniti on are given precede those under which those objects are though t.
, This sentence is revised in the second edition to reflect Kant's interveni ng argument, beginning with the Gr07lnd7!'O,.k ofthe ofMorals of 1785, that the principle of morality if not its application is indeed entirely a priori. See B28--9 below.


B 29

B 30


leave room the idea that although incentives based on feelings are not adequate for morality, there can be other, more purely rational motives for it (see Gr01md7!"ork, 4'4 2 7).

TVrln!'risbcit Bnt'c,r;u71f!:S)"!J7i;,!dr, replaced in the second edition with Tricbftder n (incentives) in order to



Introduction a


and emp iric al cog niti on. On the diff eren ce betw een pur e our cognition-heginswith_experiThe re is no dou bt wha teve r that into exerthe cognitive faculty be awakened .~e;, for how else sho uld stimulate our senses and in part cise if not thro ugh objects that in par t brin g the activity of our unthemselves prod uce representations, these, to con nec t or separate them, ders tand ing into mot ion to com pare a erial of sensible impressions into and thus to wor k up the raw mat conis e til!! as far As 7 experience? cog niti on of objects that is called edes experienC(.:,~l}(l}Yith.e:Xperi prec us in on niti ce~~QLtl1en, no cog el}Ee eYC:J:Ycg;!litiQI1Pegins. men ces wit h experience, yet it But alth oug h all our cog niti on com be from experience: FOflt couf;f well doe s not on th:lta~countall arise rewe ch whi t ftha iteo is a com pos on that even our ex~ien1iaJ(;()gI1iti faculty whi ch our own:CogI1iti~~ that and ions ress ceive thro ugh imp itself, of gut ides prov ) ress ions (merely pro mpt ed by sensible imp rial mate ntal ame fund ish from that whi ch add itio n we canIlOt distingu g tin ara c:p ._~ ntiv e to it and unti l long prac tice has. mad e us atte it out. requ irin g closer investigation, and It is ther efor e at least a que stio n ce, whe ther ther e is any such cogone not to be dismissed at first glan of the an~ e"e il ?fa llill1pressions niti on ind~!Lclent_ofalIexperience them es iisn ingl dist and l~piiQ'd,c senses. One cans such c()gnitiQnS~ in ely nam ri, erio post a thei r sources trclm .~llllIli:f]ic:ll ones, whi ch have 8 experience. d eless not yet sufficiently determiTh e former expression is neverth the que stio n before us. For it is cusnate to designate the whole sense of derived from experiential sources that tom ary to say of man y a cog niti on e it it a priori, because we do not deriv we are capable of it or part ake in

As in the second edition. the first edition first two paragraphs of Section I in Sections I and II (B 1-6) replace the (AI-Z). italics. emphasized by Kan t by the use of , Norm ally set in roma n type, here d Tha t is, "a priori."


Introduction <B> immediately from experience, but from a general we have nevertheless itself borrowed from experience. So one says of someone who undermined the foundation of his house that he could have known a priori that it would collapse, i.e., he need not have waited for the experience of it actually collapsing. Yet he could not have this entirely a priori. 9 For that bodies are heavy and hence fall is taken away must first have become known to through eXllerlence. not In the sequel therefore we will understand by a priori those that occur independently of this or that experience, but rather to those that occur absolute~y independently of all experience. them are empirical cognitions, or those that are possible a posteriori, i.e., through experience. Among a priori cognitions, however, are called pure witl1 which nothing empirical is intermixed. e.g., the proposition "Every alteration has its cause" is an a priori proposition, only not pure, since alteration is a concept that can be drawn from experience. 10


We are in possession of certain a priori cognitions, and even the common understanding is never without them. At issue here is a mark by means of which we can securely distinguish a pure cognition from an empirical one. I I Experience teaches us, to be sure, that something is constituted thus and so, but not that it could not be otherwise. First, then, if a proposition is along its necessity, it is an a priori judgment; if it is, moreover, also not derived from any proposition except one that in turn is valid as a necessary proposition, then it is absolutely a priori. Second: Exp.<:;.ri~nc:c;_never givesj1:~j~_clgill(;ms true or strict assumed comparative universality (through induction), so properly it must be said: as far as \~e11aveyet perceived, there is no exception to this or that Thus if B4 a judgment is thought in strict universality, i.e., in such a way no excepti(")~it:aIrisanowedto be possible, then it is not derived from experience, b~t is rather valid absolutely a priori. Empirical is only an arbitrary increase in validity from that most cases to that which holds in all, as in, e.g., the proposition bodies are heavy," whereas strict universality belongs to a essentially; this points to a special source of cognition for it, a ulty of a priori cognition. N~ity~J.:lcl strict are therefore secure indicationsa of an - and belong together




Introdu ction <B>

emseparably. But since in their use it is sometim es easier to show the piricall imitati on in judgme nts than the conting ency in them, or is often to more plausible to show the unrestr icted universality that we ascribe these ely a judgme nt than its necessity, it is advisable to employ separat two criteria, each of which is in itself infallible. 12 Now it is easy to show that in human cogniti on there actually are in the strictes t sense universal, thus pure a priori such necessa ry an example from the sciences, one need only wants judgme nts. If one of mathem atics; if one would have one itions propos the look at all the unders tanding , the propos ition that from the commo nest use B5 do; indeed in the latter the very every alterati on must have a cause ty of concep t of a cause so obviously contain s the concep t of a necessi would it that rule of ality univers strict a and effect an connec tion with t be entirely lost if one sought, as Hume did, to derive it from a frequen associa tion of that which happen s with that which precede s and a habit aris(thus a merely subjective necessity) of connec ting represe ntation s for es exampl such ng requiri t withou Even "3 tion. ing from that associa one on, cogniti our in les princip priori a pure of the proof of the reality ce could establis h their indispe nsabilit y for the possibility of experien get itself nce experie would where For priori. a it itself, thus establis h rules in accorda nce with which it proceed s were its certain ty if one themse lves in turn always empiric al, thus conting ent?; a hence can we here Yet les. princip first as count to these could hardly allow e cognitiv our of use pure the ed display having with es conten t ourselv b Not merely in judgments, faculty as a fact togethe r with its indicat ion. a however, but even in concep ts is an origin of some of them revealed body a of t concep ntial experie your from remove priori. Gradua lly everyth ing that is empiric al in it - the color, the hardne ss or softness, the weight , even the impene trabilit y - there still remain s the space that was occupie d by the body (which has now entirely disappeared), emand you cannot leave that out. Likewise, if you remove from your B6 all pirical concep t of every object,c whethe r corpore al or incorporeal, not still could you you, teaches nce experie which of those propert ies or take from it that by means of which you think of it as a substan ce more as depen dent on a substan ce (even though this concep t contains by determ ination than that of an object! in general). Thus, convinced the necessi ty with which this concep t presses itself on you, you must conced e that it has its seat in your faculty of cogniti on a priori.


Question mark not in original. Kennzeicbcn, i.e., sign. Objects Objects


Introduction <B>

Philoso phy needs a science that determ ines the possiibiljity, the princip les/ and the domain of cogniti ons a But what says still more than all the foregoing C is certain cognitions even abando n the field of all possible experiences, and seem A3 to expand the domain of our judgme nts beyond all of ence through concep ts to which no corresp onding object at all can be given in experience. And precisely in these latter cognitions, which go the world of the senses, where experience can give neither guidance nor correction, lie the investigations of our reason that we to be more preeminent in their import ance and sublime their final aim than l:l 7 everything that the unders tanding can learn in the field of appearances, in which we would rather venture everything, even at risk erring, than give up such import ant investigations because of any sort of reservation or from contem pt and indifference. dThese unavoidable problems of pure reason itself are God, freedo m and immor tality. But the science whose final aim in all its prepara tions is directe d properl y to the solution of these problem s is called metaph ysics, whose procedure is in the beginn ing dogma tic, i.e., it confidently takes on the execution of this task withou t an anteced ent examination of the capacity or incapacitye of reason for such a great undertaking. Now it may seem natural that as soon as one has abando ned the terrain of experience one would not immed iately erect an edifice cognitions that one possesses withou t knowin g whence, and on the credit of principles whose origin one does not know, withou t first assured oneself of its founda tion throug h careful investigations, that one would all the morefh ave long since raised the questio n how the understanding could come to all these cognitions a priori what IJlgLn, i1<llidity, and Y:.iI1J. they might have. And in fact nothing is more A4 natural, if one underst ands by the word natura l g that which nrr,npr l" and reasonably ought to happen; but if one underst ands by it that B8 usually happens, then conversely nothin g is more natural and compre , This scction number and title added in the second edition. The ensuing paragrap h commences the first part of the introduc tion common to both editions, extendin g from here to B14, though with one major interpolation in the next paragraph and another change

, "than all the foregoing" added in the second edition. The remainder of this paragraph added in the second edition. odeI' 1!""'o177Jiion" added in the second edition. g "dent Won natiirlich" substituted for "unter diesem I17one" in the second edition.


Introduction <B> long have bee n neglected. this investigation hensible long bee n relithe For one par t of these cognitions, rable expectation abo ut others as able, and ther eby gives rise to a favo therof an enti rely different natu re. Fur alth oug h these may of one is sure not of experience, mor e, if one is bey ond the circle b The cha rm in expanding one's cogbein g refu ted thro ugh experience. stop ped in one's prog ress only by niti ons is so grea t that one can be Thi s, however, one can avoid if one bum ping into a clea r contradiction. thou gh they are not ther eby invenmakes his inventions carefully, even far gives us a splendid example of how tions any the less. it Now nce. cog niti on inde pen den tly of experie a we can go these and cognitions only so far as is occupied, to be sure, with objects umstance, however, is easily s exhibited in intuitions. Thi circ i, que stio n can itse lf be given a prior Jej'rer'10,oked, since the intu itio n in ept. pur e conc be distinguished from a mer e can drive for exthe on, reas of er pow of the such a d~pli~;te(::lc A5 t dove, in free flight cutt ing through pan sion sees no bounds. The ligh d could it feels, could get the idea that it the air the resistance of vvise, Plat o aba ndo ned the world of do even bett er in airless space. Like B9 ow lim its' for the understanding, and the senses because it set such narr of the ideas, in the emp ty space of dare d to go bey ond it on the wings ce that he mad e no headway by his pur e understanding. He did not noti he no sup port , as it were, by which efforts, for he had no resistance, r orde in he could apply his powers could stiffen himself, and to whi ch fate ry ion. It is, however, a customa to put his und erst and ing into mot finish its ediflce as early as possible of hum an reas on in speculation to ly ther the gro und has bee n adequate and only then to investigate whe asto sort s of excuses will be sought prep ared for it. But at that poin t all bet ter! to refuse such a late and sure us of its sturdiness or also, even s us free of all wor ry and suspicion dan gero us examination. \Vh at keep , and flatters us witll apparent thordur ing the con stru ctio n, however aps the grea test part, of the business oughness, is this. A grea t part, perh the concepts that we already have of of our reason consists in analyses of de of cognitions that , although they objects. Thi s affords us a mul titu or clarifications of that which is alare not hin g mor e than illuminations are, at ts (tho ugh still in a confused way), A6 read y thou ght in our concep innew e ed, trea sure d as if they wer least as far as thei r form is con cern

instead The second editi on reads "lange" " inste ad of ",,,idcT'h,.richrl1 rlegt "wide reads on b The seco nd editi instead of "mi{p;cmuinte1'"t. " The second editi on reads

Vorstellung jltige Schm nkcn setzt " instead of "so vielj( ( The second editi on reads "so enge legt."

word s "auch" and "lieber gar." f The seco nd editi on inser ts the


Introduction <B> sights, though they do not extend the concepts we matter or conten t, but only set them apart from each since this proced ure does yield a real a priori cognition, which makes secure and useful progress, reason, withou t itself noticin g under these tenses surrept itiously makes assertions of sort, in reason adds someth ing entirely alien to given concepts and 'nnIAP,jU so a priori, withou t one knowin g how it was to and ""rh,,,,, . such ab question even being allowed to come to mind. I will therefo re deal with the distinction between these two sorts of right at the outset.

B 10

On the differe nce betwee n analyti c synthe tic judgme nts. 14

In all judgme nts in the relation of a subject to the predicate is thought (if I conside r only affirmative judgments, since the ap])li(:atjon to negative ones is easy) this relation is possible in two ways. Either the predica te B belongs to the subject A as someth ing is (covertly) contain ed in this concep t A; or B lies entirely outside the conceptA, though to be sure it stands in connec tion with it. In the first case I call the judgme nt analyti c, in the second synthe tic. ments (affirmative ones) are thus those in which the connec tion predicate is though t throug h identity, but those in which this connec tion is though t withou t identity are to be called judgments. One could also call the former judgm ents of clarific ation, the ter judgme nts of amplifi cation, d since through the predica te the former do not add anythin g to the concep t of the subject, but break it up by means of analysis into its compo nent were already thought in it (though confusedly); while the latter, on Imy, add to the concep t of the subject a predicate that was not th()U~;ht in it at all, and could not have been extracted from it any sis. E.g., if! say: "All bodies are extended," then this is an ~n~I\i'tlC ment. For I do not need to go beyond e the concep t that I combine the bodyfin order to find that extension is connec ted it, but I need only to analyze that concept, i.e., become conscious of the manifold that I always think in it, in order to encoun ter this predica te therein; it is therefo re an analytic judgme nt. On the contrary, if I say:
, The second edition adds the words "und zwar." second edition replaces "diese" with "eine solche." , Section number "N" added in the second edition. "F:ri'ii7ltrTlITlOT-" and" El71JCZ:tCrl'l77[truneilc. , The edition reads "uher" instead of "aus." f The second edition reads "dem Korper" instead of "dem U70rt Kilrper."
& The



Introduction <B> predica te is someth ing entirely differbodies are heavy," then mere concep t of a body in general. the in ent from that which I think yields a synthetic judgment. thus te predica a The additio n of such are an synthe tic. For it would such, as nce, experie of aJudgm ents on experience, since I do not nt judgme analytic an be absurd to ground to formul ate the judgorder in all at t concep my need to go nce for that. That experie from ny testimo no ment, and therefo re need a priori, and is not hed establis is that ition a body is extended is a propos I already have nce, experie to go I before For nce. a judgme nt of experie B T2 I merely which from t, concep the in nt judgme my all the conditions for iccontrad of le princip the with nce accorda in draw out the predica te necesthe of us conscio become time same the tion, and can thereby at the sity of the judgme nt, which experience could never teach me. On the in weight of te predica the include all at not AS contrary, althoug h I bdo oban tes designa eless neverth t concep the , concep t of a body in general still ject of experiencec throug h a part of it, to which I can therefo re add can I former. the with ing belong as nce other parts of the same experie exof marks the h throug ally analytic first cognize the concep t of body this in t though all are which etc., shape, tension, of impenetrability, of exconcept. But now I amplify my cogniti on and, looking back to the that find I body, of t concep this ed perienc e from which I had extract weight is also always connec ted with the previous marks, dand I therefore add this synthetically as predica te to that concept. It is thus expeof rience eon which the possibility of the synthesis of the predicate , concepts both since ed, ground is body weight with the concep t of tobelong eless neverth other, the in ed though the one is not contain gether, though only contingently, as parts of a whole, namely experience, which is itself a synthetic combin ation of intuitions. tEut in synthet ic a priori judgme nts tl"lis means of help is entirely A9 ang lacking. '5 If I am to go beyond the concep t A in order to cognize B 13 means by and depend I which on it is other B as combin ed with it, what the of which the synthesis becomes possible, since I here do not have the Take nce? experie of field the in it advantage of looking around for of propos ition: "Every thing that happen s has its cause." In the concept

in the first edition; The first part of the following paragrap h replaces two paragrap hs see A7-8 above. here. b The text common to the first edition resumes instead of the first editions der nd Gegensta "cinen has edition , The second " "die vollrriindigc sentence is added in the second edition. d The remainde r is modified and expanded in the second edition. C The remainde r of this sentence here. resumes text f The common in the first. g "iiber" substitut ed in the second edition for "ausscr':


Introduction <B> something that happens, I think, to be sure, of an existence was preceded by a time, etc., and from that analytic judgme nts can be drawn. But the concep t of a cause lies entirely outside that concep t, and a indicates someth ing differe nt than the concep t of what in general, and is therefo reb not contain ed in the latter represe ntation at all. How then do I come to say someth ing quite differen t which happens in general, and to cognize the concep t of cause as belonging to it, indeed necessarily,c even though not contain ed in it?d What is the unknow n =ex here on which the unders tanding de])erld s when it believes itself to discover beyond the concep t of A a predica te that is foreign to it yet which it nevertheless believes to be connec ted with it?!It cannot be experience, for the principle has been ad;::1u(~ed adds the latter represe ntation s to the former not only with greater generality than experience can provide, but also with the expression of necessity, hence entirely a priori and from mere concepts. Now entire final aim of our speculative a priori cogniti on rests on such i.e., ampliative principles; for the analytic ones are, to be sure, most important and necessary, but only for attainin g that distinctness of concepts which is requisite for a secure and extended synthesis as a newacquisition. g



Synthe tic a priori judgme nts are contain ed as princip lesi in all theoret ical science s of reason.

h. Mathem atical judgm ents are an synthe tic. 16 seems to have escaped the notice of the analysts of reason now, indeed to be diametrically oppose d to all of their conjectures, although it is incontr overtib ly certain and is very import4I1t in the For since one found that the inferences of the mathematicians ceed in accordance with the principle of contrad iction
, "liegt ganz auflerjene7JZ Begriffi, und" added in the second edition. b "ist also" in the second edition instead of "und ist" in the first. , "und so gar notwendig" added in the second edition.

Kant ends this and the next sentence with periods, for which we have substitut ed question marks. , "u17bckm771tf =" added in the second edition. i In the second edition, "welches er gleichwobl damit sein erachtet?" substituted for verknupj t sei." g In the second edition, "Erwcrb" replaces "Anbau." b At this point one paragrap h from the first edition is omitted and replaced with the folSections V and VI, BI4 through B2S.

: Princzj,?jcn Kant ~dapts the following five paragraphs from the

2 (4:268-9).


Introduction <B> quired by the nature of any apodictic certainty), one was persuaded that the principles could also be cognized from the principlea of contradiction, in which, however, theyb erred; for a synthetic proposition can of course be comprehended in accordance with the principle of contradiction, but only insofar as another synthetic proposition is presupposed from which it can be deduced, never in itself. It must first be remarked that properly mathematical propositions are always a priori judgments are never empirical, because they carry necessity with them, cannot be derived from experience. But if one does not want wen then, I will restrict my proposition to pure mathematics, the concept of which already implies that it does not contain empirical but merely pure a priori cognition. To be sure, one might initially think that the proposition "7 + 5 = 12" is a merely analytic proposition that follows from the concept of a sum of seven and five in accordance with the principle of contradiction. Yet if one considers it more closely, one finds that the concept of the sum of 7 and 5 contains nothing more than the unification of both numbers in a single one, through which it is not at all thought what this single number is which comprehends the two of them. The concept of twelve is by no means already thought merely by my thinking of that unification of seven and five, and no matter how long I a)lalyze my concept of such a possible sum I will still not find twelve in it.\Qne must go beyond these concepts, seeking assistance in the intuition that corresponds to one of the two, one's five fingers, say, or (as in Segner's arithmetic)!7 five points, and one after another add the units of the five given in the intuto the concept of seven. cFor I take first the number 7, and, as I take the fingers of my hand as an intuition for assistance with the concept of 5, to that image of mine I now add the units that I have previously taken together in order to constitute the number 5 one after another to the number 7, and thus see the number 12 arise. That 7 should be added to 5 I have, to be sure, thought in the concept of a sum = 7 + 5, but not that this sum is equal to the number 12. The arithmetical proposition is therefore always synthetic; one becomes all the more distinctly aware of that if one takes somewhat larger numbers, for it is then clear that, twist and turn our concepts as we will, without getting help from intuition we could never find the sum by means of the mere analysis of our concepts.




Kant switches number from "man" to "sie." , This and the following sentence are substituted here for the clause "Man envcitet also wirklich seinen Begriff dureh diesen Satz 7 + 5 = 12 und thut zu dem ersteren Begrijf eine" nmen hinzu, der in jenem gar nicht gedacht war" (One therefore really amplifies his concept through this proposition "7 + 5 = 12" and adds a new concept to the former, which was not thought in it) in the (4:269).


Introduction <B> Just as is any princip le pure geomet ry an;alvuc. straight betwee n two points is shortes t is a sVllttletlc nr",-,,,~i_ cion. For my concep t of the straigh t contain s nothin g of qU<Hltlty, only a quality. I8 The concep t of the shortes t is therefo re entin~lv cional to and cannot be extracte d out of the concep t of the stral~;ht line any analysis. Help must here be gotten from means of which alone the synthes is is possible. To be sure, a few princip les that the geomet ers presupp ose are actually analytic and rest on the princip le of contrad iction; but alsoa only serve, as identic al proposi tions, for the chain of and not as principles,b e.g., a = a, the whole is equal to itself, or (a + b) > a, i.e., whole is greater than its part. yet even these, althoug h are valid in accorda nce with mere concep ts, are admitte d in ln~lthen latics only because they can be exhibit ed in intuitio n. I9 ~at makes us believe here that the predica te of such apodict ic judgme nts <) I r,p<)rllu lies in om concep t, and that the judgme nt is therefo re analytic , is merely the ambigu ity of the expression. We should , namely, a certain predica te to a given concep t in though t, and this necessi ty already attaches to the concep ts. But the questio n is not what we should think in addition to the given, concep t, but what we actuall y in it, though only obscurely, )nd there it is manife st that predica te certainly adheres to those concep ts necessarily, though not as though t in the concep t itself, but by means of an intuitio n that must be added to the concept. / 2. Natura lscienc e (Physica) contain s within itself synthe tic Dn judgm ents as princip les. d I will adduce only a couple of propos itions as examples, such as the propos ition that in all alterati ons of the corporeal world the quantit y of matter remain s unalter ed, or in all communication of motion effect and counter -effect must always be equal. In both of these not only the necessity, thus their a priori origin, but also that they are synthet ic propos itions is clear. For in concep t of matter I do not think persiste nce, but only its presenc e in space through the filling of space. Thus I actually go beyond the concep t of matter in order to add someth ing to it a priori that I did not think in it. The proposi tion is thus not analytic, but synthet ic, and neverth eless thought a priori, and likewise with the other proposi tions of the pure part of natural science. 3. In metaph ysics, even if one regards it as a science has

B 17


" "auch" added to text from (4:269)b Prinripicn , "als im Begriffi selhst gedacht" substitut ed here for the word "lI777J1ittc lbmo, in the Prol(zo7llclW (4'269).


Introduction <B> is nevertheless indispensable because. of the namerely been sought cognit ions are supposed to tic a synthe ture of human reason, with analyzing concepts merely ed concern not is it and be ng them analytically, clarifyi thereby and a things of that we make end we must make this to priori; a on cogniti our amplify but we want to concepts that was given the to ing someth add that les use of such princip judgme nts go so priori a ic synthet h throug and not contain ed in them, far, e.g., in the that us follow cannot itself nce far beyond that experie and others being," beginn first a have must propos ition "The world ned, conconcer is end its as far as least at ysics, metaph sides, and thus 2o tions. proposi priori a ic sists of purely synthet

a The genera l proble m

pure reason. 2I

B 20

One has already gained a great deal if one can bring a multitude ofinnot vestigations under the formula of a single problem . For one thereby the also but y, precisel it ining determ by task, only lightens one's own satishave we r whethe e examin to wants who judgme nt of anyone else d fied our plan or not. The real problem of pure reason is now containe e? possibl a ents judgm tic in the question: How are synthe That metaphysics has until now remain ed in such a vacillating state of uncerta inty and contrad ictions is to be ascribed solely to the cause of that no one has previously though t of this problem and perhaps even sothe On nts. judgme tic synthe and c analyti the distinction betwee n lution of this problem , or on a satisfactory proof that the possibility that it demands to have explained does not in fact exist at all, metaphysics now stands or falls. David Hume, who among all philosophers came closest to this problem , still did not conceive of it anywhere near determinate ly enough and in its universality, but rather stopped with the synthetic propos ition of the connec tion of the effect with its cause believing himself to have brough t out that such (Principium inis an a priori propos ition entirely impossible, and according to his a to down come would ysics metaph call we ferences everyth ing that fact in has which that into reason of insight alleged mere delusion of an the merely been borrow ed from experience and from habit has taken on hy, philosop pure all of tive destruc n, assertio an ty; appearance of necessi its on which he would never have fallen if he had had our problem in that hended compre have would he then since generality before his eyes, according to his argume nt there could also be no pure mathematics, since this certain ly contains synthet ic a priori propositions, an assertion


Introduction <B> from which his sound unders tanding surely him. 22 In the solutio n of the above problem there is at the same time contained the possibility of the pure use of reason in the ground ing and execution of all sciences that contain a theoret ical a cogniti on of objects, i.e., the answer to the questions: How is pure mathem atics possibl e? How is pure natura l science possibl e? About these sciences, since they are actually given, it can appr,Cjp nately be asked how they are possible; for that they must be is proved through their actuality.* As far as metaph ysics is concern ed, however, its poor progres s up to now, and the fact no metaphysics thus far expoun ded can it even be said that, as far as its essentia l end is concerned, it even really exists, leaves everyon e with to doubt its possibility. But now this kind of cognit ion is in a certain sense also to be regarded as given, and metaphysics is actual, if not as a science as a natural predisp osition (71letaphysica naturalis). For human reason, without being moved by the mere vanity of knowin g it pushes on, driven by its own need to such questio ns that cannot be answ,ere :d by any experiential use of reason and of principlesa such a use; and thus a certain sort of metaph ysics has actually been present in all human beings as soon as reason has extende d itself to specula tion in them, and it will also always remain there. And now about this too the question is: How is metaph ysics as a natura l predisp osition possible? i.e., how do the questio ns that pure reason raises, and which it is driven by its own need to answer as well as it can, arise from the nature of universal human reason? But since unavoid able contrad ictions have always found in all previous attempt s to answer these natural questions, e.g., whethe r world has a beginn ing or exists from eternity, etc., one cannot leave it up to the mere natural predisp osition to metaphysics, i.e., to pure faculty of reason itself, from which, to be sure, some sort of metaphysics (whatever it might be) always grows, but it must be possible to bring it
* Some


B 22

may still doubt this last point in the case of pure natural science. Yet one need merely consider the various propositions that come forth at the outset of proper (empirical) physics, such as those of the persistence of the same quantity of matter, of inertia, of the equality of effect and counter-effect, etc., and one will quickly be convinced that they constitute a physica pura (or rat-umalt, r), which well deserves to be separately established, as a science of its own, in its whole domain, whether narrow or wide.



Introduction <B>
i.e., to certain ty regardi ng either the knowledge or ignoran ce of objects, about or ns questio the objects of its to come to a decision either the capacity and incapacity" of reason for judging someth ing about de~ them, thus either reliably to extend our pure reason or else to set from flows which n, questio termina te and secure limits for it. This last rightly be this: How is metaph ysics the general problem above, possib le as science ? The critique of reason thus finally leads necessarily to science; the ess dogmat ic use of it withou t critique, on the contrary, leads to groundl to thus ones, le plausib assertions, to which one can oppose equally B 23 skeptic ism. terribly extensive, for it does not deal Further , this science cannot b of reason, whose muItiplicityc is infinite , but merely with with objects that itself, with problem s that spring entirely from its own womb, and but it from distinct are the nature of things that are not set before it familiar tely comple throug h its own nature; so that, once it has become it with its own capacityd in regard to the objects that may come before and tely comple in experience, then it must become easy to determ ine, all securely, the domain and the bounds of its attemp ted use beyond bounds of experience. one can and must regard as undone all attemp ts made until now about a metaphysics dogma tically; for what is analytic in one to inor the other of them, namely the mere analysis of the concep ts that for tion prepara a only habit our reason a priori, is not the end at all, but syntheti on cogniti metaphysics proper, namely extendi ng its a priori conis what shows cally, and it is useless for this end, because it merely in tained in these concep ts, but not how we attain such concep ts a priori the to regard in use order thereaf ter to be able to determ ine their valid selfobjects of all cogniti on in general'!\It also require s only a little B 24 of ictions contrad the denial in order to give up all these claims, since dogin able reason, which cannot be denied and which are also unavoid prematic proced ure, have long since destroy ed the authori ty of every be to not order in vious metaphysics. Ivlore resolut ion will be necessary anusing from ce deterre d by interna l difficulty and external resistan proother approa ch,' entirely oppose d to the previous one, in order to indisis that science a mote the produc tive and fruitful growth of pensable for human reason, and from which one can chop down every stem that has shot up withou t ever being able to eradica te its root.



Un7xrmifr rcn


Introduction <B>

The idea and divisio n of a special science under the name of a critiqu e of pure reason. Now from all of this there results the idea of a special science, can be called the critiqu e of pure reason . b F or C reason is that provides the principlesd of cogniti on a priori. Hence pure reason is that which contains the principlese for cognizing someth ing ab:solut' c:ly a priori. An organo n of pure reason would be a sum total of all those principles!in accordance with which all pure a priori cognitions can acquired and actually brough t about. The exhaustive such an organo n would create a system of pure reason. But since that requires a lot, and it is still an open questio n whethe r such an arrlPl!tl cation of our knowledge is possible at all and in what cases it possible, we can regard a science of the mere estimat ion of pure reason, of its sources and boundaries, as the propae deutic to the system pure reason. Such a thing would not be a doctrin e, but must be only a critiqu e of pure reason, and its utility in regard to speculationg would really be only negative, serving not for the amplification but for the purification of our reason, and for keeping it free of errors, which a great deal is already won. LcalLall cog:nltl()fl-GanS,CeI1Q:11t:aU:hat is ()ccupied not so much with objects of cognition of objects insofar as this is to be possible apriori. b,2 3 A system of such concepts would be called transce ndenta l philoso phy. But this is again too much for the beginning. For since such a science have to contain comple tely both the analytic as well as the synthet ic a priori cognition, it is, so far i as our aim is concern ed, too broad in scope, since we need to take the analysis only as far as is indispensably necessary in order to provide insight into the principles of a priori synthesis in their entire scope, which is our only concern . This investigation, we
, The section number VII and the following title are inserted at this point in tile second edition, following which the text common to the two editions resumes, with minor alterations. b "die Kritik der reinen Vennmf t heiflen kann" substituted in the second edition for "die ZU7~ KYitik der rrillC7l diC77C7l k071l7C." The next two sentences in the first edition are omitted; see A I I above. , "Denn" suhstituted in the second edition for "Null."



B 26

ltUUIcHUW' ,J; der added in the second edition. "wmdcnz 77tit UllscrCT Erkc7177tnist771: von so fern diese a priori rniiglicb sein solI" substituted in the second edition for "sondcnz mit 1l71Sfnz Begrijftn a priOli von

"so weit" substituted for "i11mFr,.11" in the second edition.


Introduction <B> tran scen den tal critique, since it not doc trin e can at of cognitions themselves but only does not aim at the amplification thtouc hsto ne of the wor th or wor thei r correction, and is to supply the is that with which we are now conlessness of an cognitions a priori, an gly a prep arat ion, if possible, for cerned. Such a critique is accordin mplished, then at least for a canon, orga non , and, if this can not be acco plet e system of the philosophy of in accordance with whi ch the com in the amplification or mer e limipur e reason, whe ther it is to consist case at least some day be exhibited tatio na of its cog niti on, can in any For that this sho uld be possible, inbot h analytically and synthetically. be too grea t in scope for us to hope dee d that such a system sho uld not can be assessed in advance from the to be able enti rely to com plet e it, , re of things, which is inexhaustible fact that our obje ct is not the natu es abo ut the natu re of things, and this but the und erst and ing, whi ch judg AI3 cog niti on, the supply of which, since in turn only in rega rd to its a priori rnally, can not rem ain hidden from we do not need to search for it exte ugh to be com plet ely recorded, its us, and in all likelihood is small eno . and subjected to a corr ect appraisal wor th or worthlessness assessed, of ems syst and ks boo the crit ique of bEven less can one expect here a B 27 if pur e faculty of reas on itself. Only the of that er rath pur e reason, but g aisin appr for ne hsto e a secure touc this is one's gro und does one hav new works in this specialty; otherand old of tent con the philosophical judg e assesses the groundless asser wise the unqualified hist oria n and ss. ndle grou ally equ ch are tion s of othe rs thro ugh his own, whi the idea of a science,d for which here is phy oso phil 'Tra nsce nde ntal ine the enti re plan architectonically, the critique of pur e reas on is to outl rant ee for the completeness and ceri.e., from principles,' with a full gua m comprise this edifice. It is the syste tain ty of an the com pon ents that dy alrea lf itse not is g que criti Tha t this of all prin cipl es! of pur e reason. s solely on the fact that in order to rest phy oso phil tal den called tran scen have to con tain an exhaustive analy be a com plet e system it would also ri. Now our crit ique must, to be sure, sis of all of hum an cog niti on a prio n of all of the ancestral concepts! lay before us a com plet e enu mer atio in question. Onl y it prop erly refrains that comprise the pur e cog niti on e concepts themselves as well as from the exhaustive analysis of thes
second edition. next two sente nces are adde d in the edition al phil osop hy" pres ent in the first dent scen tran of " The title "n:. Divi sion omit ted in the second. for" bier nur eine Idee." aft" subs titute d in the seco nd editi on d "Die Idee einer FVissenscb , PTincipien
f !',-i:'?cif,ipn

nd edition. This sente nce inser ted in the seco Stammbcf!J"ifff


Introduction <B> from comple te review of of those derived cause this analysis would not be purposeful,a since it does not contain the difficulty encoun tered in the synthesis on accoun t of which the whole critique is actually underta ken, partly because it would be Contrary to the unity of the plan to take on responsibility for Dl.e completeness of such an analysis and derivation, which one be relieved given its aim. This completeness of the analysis as well as the derivation from the a priori concepts that are to be in the future will nevertheless be easy to comple te as long as as exhaustive principles b of syntllesis, and if nothing is in regard to this essential aim. To the critique of pure reason there accordingly belongs everything that constitutes transce ndental philosophy, and it is the idea of transcendental philosophy, but is not yet this science itself, since it goes only so far in the analysis as is requisite for the comple te estimation of synthetic a priori cognition. The chief target in the division of such a science is concept must enter into it that contains anythin g enlplflc:al, a cognition be entirely pure. Hence, althoug h suprem e ciples of morality and the fundamental concepts of it are a tions, they still do not belong in transce ndental while they do not, to be sure, take the concepts of pleasure and displeasure, of desires and inclinations, etc., which are all of empirical origin, as the ground of their precept s, they still must necessarily include them in the composition of tlle system of pure moralit y in the of as the hindrance that must be overcome or the attracti on not to be made into a motive. Hence transce ndental philoso phy is a philosophyI of pure, merely speculative reason. For everytlling practical, insofar as it contains incentives," is related to feelings, which belong among empirical sources of cogniti on. Now if one wants to set up the division of this science general viewpoint of a system in general, then what we now present must contain first a Doctri ne of Eleme nts and second a Doctri ne of Method of pure reason. Each of these main parts will have its subdivi sion, the grounds for which cannot yet be expounded. All that seems


The remainder of this sentence in the second edition is substituted for the following in the first: "since the concepts of pleasure and displeasure, of desires and inclinations, of choice, etc., which are all of empirical origin, must thereby be presuppo sed." d HHt7rclJhcit BCW,C(Tl{1:WS91'iil"ic in the first edition is replaced in the second with to leave room for idea that although incentives based on feelings are not adequate for morality, there can be other, more purely rational motives for it.


Introduction <B> necessary for an introduction or preliminary is that there are two stems of human cognition, which may perhaps arise from a common to us unknown root, namely sensibility and understanding, through the first of which objects are given to us, but through the second of which they are thought. Now if sensibility were to contain a representations, which constitute conditiona under which objects are given to us, it will belong to transcendental philosophy. The transcendental doctrine of the senses will have w to the first part of the science of elements, since the conditions which alone the objects of human cognition are given precede those under which those objects are thought.
"B!?di1C?f!!t:TJP;" in the second edition replaces "Br'dinp;7171p;o7" in the first.





Transcendental Doctrine

The Transcendental First The Transcendental A esthetic a , I


In whatever way and through whatever means a cognition may to objects, that through which it relates immediately to and at which all thought as a means is directed as an end, is mt:UlltJon. however, takes place only insofar as the object is given to us; rum, is possible only if it affects C the mind in a certain way. ity (receptivity) to acquire representations through the way in we are affected by objects is called sensibility. Objects are given to us by means of sensibility, and it alone affords us mt::urtlOns; but they are thought through the understanding, and from it arise concepts. But all thought, whether straightaway (directe) or a detour (indirecte), must ultimately be related to in our case, to sensibility, since there is no other way in which objects can be given to us. The effect of an object on the capacity for representation, insofar as we are affected by it, is sensation. d That intuition which is related to object through sensation is called empirical. The undel:enmilled of an empirical intuition is called appearance) I call that in t..~e appearance which corresponds to se]"lsation its matter, but that which allows the manifold of appearance to intuited as 01'a



The "Transcendental Aesthetic" underwent major changes between the two editions of the Critique, including but not limited to the separation of the "Metaphysical" and "Transcendental" expositions of space and time and the addition of three sections to the concluding "General Remarks." We therefore present both versions in their entirety, using the marginal pagination and notes to show where specific changes were made. The following version from the first edition also includes the notes Kant made in his 0"11 copy of that edition. [, The following note is inserted in Kant's copy: "[intuition] is opposed to the concept, which is merely the mark of intuition. "The universal must be given in the particular. Through that it has significance." (EX, p. 15; 23:21) Added in Kant's copy: "If the representation is not in itself the cause of the object [Objects]." (E XI, p. 15; 23:21) d Added in Kant's copy: "Intuition is related to the object [Object]' sensation merely to the subject." (E XII, p. 15; 23:21)


Doc trine of Elem ents Part I <A>

fan n of appearance. Since that within in certain orde red and placed in a certain fonn which the sensations can the mat ter of all appearance is only can not itself be in turn sensation, in the mind a but its fon n mus t all lie ready for it given to us a ation. sens all ly from can ther efor e be considered separate priori, in which e) sens tal the tran scen den I call all repr esen tatio ns pur e ngly the ordi Acc n. enc oun tere d that belongs to sensatio not hin g is to d in the tere oun general is to be enc pur e form of sensible intu itio ns in ited in intu is s nce ifold of appeara min d a priori, whe rein all of the man pure ed call also is lf of sensibility itse cert ain relations. Thi s pur e form h whic that a repr esen tatio n of intu itio n. So if I separate from the B 35 y, ibilit divis e, forc such as substance, the und erst and ing thinks abo ut it, sensation, such as impenetrability, to ngs etc., as well as that whi ch belo A2I this empirical intu itio n is still left hardness, color, etc., som ethi ng from , . The se belo ng to the pur e intuition for me, nam ely extension and form senor es sens an actual obje ct of the which occurs a priori, even with out in the mind. ty ibili sation, as a mer e form of sens b of a priori sensibility the transcenes cipl I call a science of all prin h ther efor e be such a science, whic den tal aesthetic.*A The re mus t in , ents elem scen den tal doc trin e of con stitu tes the first par t of the tran B 36 c principles of pur e thinking, and is the s con tras t to that whi ch contain nam ed tran scen den tal logic.
A 2 I / B 35

B 36

thetics" to who now emp loy the wor d "aes * The Ger man s are the only ones this is a for nd grou the criti que of taste. The desi gnat e that whic h othe rs call critical the ging brin of , rten anal yst Bau mga failed hope , held by the exce llent rules its ating elev and n,d reaso of es prin cipl estim atio n of L~e beau tiful und er are ria crite e. For the puta tive rule s or to a science. But this effo rt is futil never e efor ther can ces are conc erne d, and mer ely emp irica l as far as thei r sour t be diwhi ch our judg men t of taste mus to g rdin acco s rule serv e as a priori ctness corre the of ne hsto touc ine genu es the rect ed, rath er the latte r cons titut of this use the advisable agai n to desi st from of the former. For this reas on it is would one y ereb (wh whic h is true scie nce term and to save it for that doct rine the whom ng amo , ents anci the sens e of the com e clos er to the lang uage and n). know well very was TOl VO'Y] KOlL division of cogn ition into uLuth yTOl

things occu to denote the relation among several te the relation between objects and the deno to word u~e ving time, reser the only four times, to be noted below, in nitive subject (in which sense it is used almost thus will l plura its or tion" "rela "). Since section of the "Transcendental Aesthetic term this of e rrenc occu the plural, further notes of always be translating Verhalmis or its ted. omit be in the "Transcendental Aesthetic" will

ndental Verhiiltnis throughout the "Transce Verhiiltni'sen. Rant uses the term pying different positions in space or



Section 1. On Space <A> In the transcendental aesthetic we will therefore first isolate sensibility by separating off everything that the understanding thinks through its concepts, so that nothing but empirical intuition remains. Second, we will then detach from the latter everything that belongs to sensation, so that nothing remains except pure intuition and the mere form appearances, which is the only thing that sensibility can make a priori. In this investigation it will be found that there are two pure forms of sensible intuition as principlesa of a priori cognition, space and time, with the assessment of which we will now be concerned.

The Transcendental Aesthetic First Section On space. By means of outer sense (a property of our mind) we represent to ourselves objects as outside us, and all as in space. In space magnitude, and relation to one another is determined, or determinable. Inner sense, by means of which the mind intuits itself, or its inner state, gives, to be sure, no intuition of the soul itself, as an object;b yet it is still a determinate form, under which the intuition of its inner state is alone possible, so that everything that belongs to the inner determinations is represented in relations of time. Time can no more be intuited externally than space can be intuited as something in us. Now what are space and time? Are they actual entities?c Are only determinations or relations of things, yet ones would to them even if they were not intuited, or are they relations attach to the form of intuition alone, and thus to the tution of our mind, without which these predicates not be ascribed to any thing at all?5 In order to instruct ourselves about we will consider space first. 6 r) Space is not an empirical concept that has been from outer experiences. For in order for certain sensations to be relatedd to something outside me (i.e., to something in another place in space from that in which I find myself), thus in order for me to represent them as outside one another, thus not merely as different but as in different places, the representation of space must already be their ground.? Thus the representation of space cannot be obtained from the relations of outer





scendental Aesthetic <A> Doctrine of Elements. Part I. The Tran this oute r experience is itself first appearance thro ugh experience, atio n. a possible only thro ugh this repr esem n, a priori, which is the ground of 2)b Space is a necessary representatio A24 8 repr esen t that ther e is no space, alall oute r intuitions. One can never ther e are no objects to be encounthou gh one can very well thin k that rded as the condition of the possitere d in it.9 It is ther efor e to be rega B 39 is rmi nati on dep end ent on them, and bility appearances, not as a dete .' nces eara app r oute ssarily grou nds an a priori repr esen tatio n that nece igeo met rica l principles and the poss all of y aint cert 3) Thc apodictic . ssity nece a this in ed are gro und bility of thei r a priori con stru ctio n ; riOri poste a d uire acq t cep space were a con For if this repr esen tatio n es of oute r experience, the first principl eral gen of whi ch was drawn out y The . ons epti perc but ld be noth ing mathematical dete rmi nati on wou not ld wou it and on, epti enc y of perc wou ld ther efor e have all the con ting ight line lie betw een two points, but stra one only even be necessary that h that. \Yh at is bor row ed from exexperience would mer ely always teac intive universality, nam ely through perience always has only compara been has as far as that be able to say duction. One would ther efor e only n found that has mor e than three bee has e spac observed to date, no dimensions.d,IO said, general con cep t of relations 4) Space is not a discursive or, as is itio n. For, tIrst, one can only repreof thin gs in general, but a pur e intu A25 ks of man y spaces, one understands sen t a single space, and if one spea same uniq ue space. I ! And these parts by that only part s of one and the le all-encompassing space as its comcanTlot as it were prec ede the sing n wou ld be possible), but rather are pon ents (from whi ch its com pos itio single; the manifold in it, thus also only tho ugh t in it. It is essentially

: botto m of this page in Kant's copy The following note is adde d at the intuition. "[1.] Space is not a concept, but an can ... tion, for everything empirical one - not an emp irica l intui 2. 3. It is an a priori intui tion ... (E XIII, p. 16; 23:22 ) 4. Space is the subjective fonn ..." : copy 's b Adde d in Kant that which relations, as Leib niz supposed, but "Space is not a conc ept of external relations. grou nds the possibility of external is a proof propositions to some thing external "The necessity of the relation of our sm." (E ideali st again s; thing nal exter we stand with of the real conn ectio n' in whic h XIv, p. 16; 23: 22) , Inse rted in Kant's copy: outer ex experience, but a grou nd of possible "Space is not a conc ept derived from ) 23:22 16; p. XV; (E space if ..." perience. I must have a conc ept of and of. !Sii] the synthetic a priori proposition. of. "Pro of of the ideality of space from ) 23:22 16; p. , XVI This is no hypothesis ... [sic]" (E ced by 3, "The Transcendened in the second edition, and repla d This para grap h is delet ). 40-1 of the Con cept of Space" (B tal


Section I. On Space


the general concep t of spaces in general, rests merely on Imutatlolns. From this it follows that in respect to it an a priori intuitio n IS not empirical) ground s all concepts of them. Thus also all ge,orrletrica! principles, e.g., that in a triangle two sides togethe r are always greater than the third, are never derived from general concepts of line triangle, but rather are derived from intuitio n indeed a with apodictic certainty. 5) Space is represe nted as a given infinite magnitude. A general concept of space (which is commo n to a foot as well as an ell) can determ ine nothing in respect to magnitude. If there were not boundlessness in progress of intuitio n, no concep t of relations could bring it a ciplea of their infinity.b,I2

Conclu sions from the above concep ts.

A26/S4 2

a) Space represents no propert y at all of any things in themselves nor any relation of them to each other, i.e., no determ ination of them that attaches to objects themselves and that would remain even if one were to abstract from all subjective conditi ons of intuitio n. For neither absolute nor relative determinatiol1s can be intuited prior to the existen ce of the things to which they pertain, thus be intuited a priori. '3 b) Space is nothing other than merely the form of all appearances of outer sense, i.e., the subjective conditi on of sensibility, under which alone outer intuitio n is possible for us. Now since the receptivity of subject to be affected by objects necessarily precedes intuitio ns of these objects, it can be unders tood how the form of all appearances can be given in the mind prior to all actual perceptions, thus a priori, and how as a pure intuitio n, in which all objects must be dete.qnined, it can contain principlesc of their relations prior to all experience. d We can accordingly speak of space, extended beings, and so on, from the human standpo int. If we depart from L1.e subjective condition under which alone we can acquire outer intuitio n, namely that throug h which we may be affected by objects, then the represe ntation of space

This paragraph is changed in the second edition; see paragrap h 4, B39--40 below. in Kant's copy: "Space and time carry with them in their represen tation the concept of necessity. Now this is not the necessity of a concept. For we can prove that their non-existence is not contradictory. Necessit y also cannot lie in the empirical intuition. For tfus can, to be sure, carry with it the concept of existence, but not of necessary existence. Thus this necessity is not in the object [Object] -objective - at all; con~;cql1cntjy it is only a necessary condition of the subject for all perceptio ns of the senses. (E XVII, p. I7; 23:22-3)


scendental Aesthetic <A> Doctrine of Elements. Part L The Tran ibut ed to things only insoa Thi s pred icat e is attr signifies noth ing at cts of sensibility.b,I4 The constant far as they appear to us, i.e., are obje icall sensibility, is a necessary cond form of this receptivity, whi ch we ide can be intu ited as outs the relations with in which objects tion of h it e objects, is a pur e intuition, whic us, and, if one abstracts from thes of can not make the special conditions hears the nam e of space. Since we their of only but gs, thin ibility sensibility into con diti ons of the poss e com preh end s all things that spac that say l wel can appearances, we ther but not all thin gs in themselves, whe may appear to us externally, . ited intu be may d they ject sub wha teve r they be intu ited or not, or king bewhe ther the intu itio ns of othe r thin For we can not judge at and that n itio intu our ns ings are bou nd to the same conditio t to the men judg a of n' tatio the limi are universally valid for us. If we add . The valid y nall itio ond unc is men t con cep t of the subject, then the judg only d vali is e"! spac in ther one ano prop osit ion "All thin gs are next to ible sens our of cts obje as n gs be take und er the limitation that these thin s, thing "All say: and t cep on to the con intu itio n. If here I add the con diti is rule this then e," spac in ano ther as oute r intuitions, are nex t to one gly rdin acco ions osit exp and with out limitation. Our B44 valid universally rd to everything , objective validity) of space in rega A28 teac h the real ity (i.e. same timeg the the at but as an object, that can come before us externally whe n they are considered in themidea lity of space in rega rd to things taki ng acco unt of the constitution of selves thro ugh reason, i.e., with out the emp iric al real ity of space (vvith our sensibility. We ther efor e assert to be sure at the same possible oute r experience), thou gh resp ect to ing as soon as we noth is i.e., that it tim e its tran sce nde ntal idea lity , nce, and take it as erie exp all lity of leave out the con diti on of the possibi es. in themselv som ethi ng that grou nds the things
that we do not all created beings are boun d to it, Inse rted in Kant's copy: "Per haps most important The . form ble sensi ly mere a is it know. This muc h one can know, that r intuition we conc ept a priori, and thro ugh inne thing is that it yields a dete rmin ate science of objects no and tions senta repre rical empi would not have sensations, thus no 23=23) [Objecte] a priori." (EXVIII, p. 17; ictically assert, since he still ts: "as Men delss ohn could so apod b Here Kant's copy inser p. 17; 2344 ) gave space objective reality." (E XX, herable note also appears: At abou t this point, this partially decip "Fie ld of spac e and of time . s, thus not to God; J. d furth er than to objects of the sense "1. Both cann ot exten (E XIX, p. 17; 2p3) ..." of ts of things as objec Even amo ng these they are valid only ). 23:44 18; p. , XXI (E rr) ill171t (nur , Here Kant's copy inserts "ever" 23'44 )' es out "or not, or" (E XXII, p. 18; d In his copy Kan t cross XXIII, p- 18; 23=45). (E " ition cond iting "lim to , Kant's copy changes "limitation" to one another in space next are s thing prop ositi on to "All f In his copy Kan t changes this . ). 23:45 18; p. ; or they are some wher e" (E XXIV auch" (E XXV; p. 18; "also" (changing "ob zwar" to "aber g In his copy Kan t inser ts


Section 1. On Space <A> Besides space, however, there is no other subjective represe ntation relateda to someth ing external that could be called a bHence this subjective conditi on of all outer appearances cannot be compared with any other. The pleasan t taste of a wine does not to the objective determ ination s of the wine, thus of an object'" even considered as an appearance, rather to the constit ution of sense in the subject that enjoys it. Colors are not objective of the bodies to the intuitio n of which they are attached, but are modifications of the sense of sight, which is affected 111 a certain way. Space, on the contrary, as a conditi on of outer d necessarily belongs to their appearance or intuitio n. Taste colors are no means necessary conditions under which alone the objects can objects' of the senses for us. They are only combin ed with the appearance as conting ently added effects of the particu lar organization. Hence they are not a priori representations, are ground ed on sensation, pleasant taste is even ground ed on feeling (of pleasure and dlsple:asl1re as an effect of the sensation. And no one can have a priori the represe ntation either of a color or of any taste: but space concerns the pure form of intuition, thus it includes no sensation (nothin g empirical) in itself, and all kinds and determ ination s of space can and even must be able to be represe nted a priori if concepts of shapes as well as re!;ati(m s are to arise. Throug h space alone is it possible for things to be outer objects for us! The aim of this remark is only to preven t one from of illustrating the asserted ideality of space with completely inadequ ate examples, since things like colors, taste, etc., are correct ly considered not as qualities of things but as mere alterations of our subject, which can even be different in differen t people. For in this case that which is origina lly only appearance, e.g., a rose, counts in an sense as a thing in itself, which yet can appear different to every eye in regard to color. The transcendental concep t of appearances in space, on the contrary, is a critical remind er that absolutely nothing that is intuited in space is a thing in itself, and that space is not a form that is proper to


B 45


'The rem8indcr of this paragrap h is altered in the second edition: see

44-5 below.

c Objecte }Inserted in the margin of Kant's copy: "Pure idealism concerns the existence of things outside us. Critical idealism leaves that undecided, and asserts only that the form of their intuition is merely in us." (E XXVI,p. r8; 23'13) 0\ nlfthcr note adds: "An idealism, from which the possibility of an a priori cognition and of mathematics can be cognized." (E X)"'VII, p. 19; 23:23)


Tran scen dent al Aesthetic <A> Doc trine of Elem ents . Part I. The

are not known rath er that objects in themselves any thin g in itself, g othe r than hin are not and u~at wha t we call oute r objects to us at e, but whose spac is ty, whose form mer e repr esen tatio ns of our sensibili be cognized not can f, is not and true correlate, i.e., the thin g in itsel nce. d afte r in experie thro ugh them , but is also nev er aske The Tra nsc end enta l Aes thet ic Sec ond Sec tion On tim e. I5


is som eho w draw n from an exTim e is not an empirical con cep t that e ession would not themselves com perience. For sim ulta neit y or succ apri n of tim e did not gro und them into perc epti on if the repr esen tatio s one repr esen t that several thing ori. Onl y und er its pres upp osit ion can ultaneously) or in different times exist at one and the same time (sim (successively). s. In esen tatio n that grou nds all intuition 2) Tim e is a nec essa ry repr A 3I one one can not remove time, thou gh rega rd to appearances in general e s away from time. Tim e is therefor can very well take the appearance latactuality of appearances possible. The given apriori. In it alone is their of on diti con l ersa univ f, as the ter cou ld an disappear, but time itsel d. ove rem be not can possibility, und s the possibility of apodictic 3) Thi s a priori necessity also gro B47 has , or axioms of tim e in general. It principles of the rela tion s of time s are not simultaneous, but succesonly one dimension: diff eren t time . not successive, but simultaneous) sive (just as diff eren t spaces are d n from experience, for this woul The se principles cou ld not be draw dict ic certainty. We would only apo yield neit her stri ct universality nor perc epti on teaches, but not: This be able to say: Thi s is wha t com mon principles are valid as rules under is how mat ters mus t stand. The se an, and inst ruct us prio r to them, not whi ch experiences are possible at b thro ugh it. it, general concept, but a pure Tim e is no discursive or, as one cans nt times are only part s of one and the form of sensible intu itio n. Dif fere A 32 however, whi ch can only be given I6 same time. Tha t repr esen tatio n, itio n. Fur ther , the proposition that thro ugh a single object, is an intu neo us can not be derived from a gendifferent time s caIU10t be sim ulta

paragraph rather e the cent er of the first line of this The "I" is actually print ed at abov than at its begi nnin g. lbe." Earlier editors suguns vor dcrselhcn, und nicht dureh diese b The text reads "belchren preted t9 mean "diesclben" but if the sente nce is inter gested eme ndin g the last word to it can be read n," eptio perc mon com ugh not thro "instructs us prior to experiences, with out emen datio n.


Section H. On Time <A> eral concept. propos ition is synthetic, and cannot arise concepts alone. It is therefo re immed iately contain ed in intuitio n and representation of time. 5) The infinitu de of time signifies nothing more than that every determinate magnit ude of time is only possible throug h of a single time ground ing it. The original represe ntation , time, must therefore be given as unlimited. But where the parts themselves and magnitude of an object can be determinately represe nted limitation, there the entire represe ntation cannot be given thlou.g h concepts (for then the partial represe ntation s precede) but their immediate intuitio n must be the ground . '7 Conclu sions from these concep ts.
a) Time is not someth ing that would subsist for itself or attach to as an objective determ ination , and thus remain if one abstracted all subjective conditi ons of the intuitio n of them; for in first case it would be someth ing that was actual yet withou t an object. As as the second case is concerned, however, time could not precede the objects as a determ ination or order attachi ng to the things themselves as their conditi on and be cognized and intuited a priori through synthetic propositions. But the latter, on the contrary, can very well occur if time is nothing other than the subjective conditi on under which all intuitions can take place in us. For then this form of inner intuitio n can be represented prior to the objects, thus a priori. 18 b) Time is nothin g other than the form of inner sense, i.e., of intuition of our self and our inner state. '9 For time cannot be a determ ination of outer appearances; it belongs neither to a shape or a etc., but on the contrar y determ ines the relation of represe ntation s in our inner state. And just because this inner intuitio n yields no shape we also attempt to remedy this lack throug h analogies, and represe nt temporal sequence throug h a line progres sing to infinity, in the manifold constitutes a series that is of only one dimension, and infer from the propert ies of this line to all the propert ies of with the sole difference that the parts of the former arc simultaneous but those of the latter always exist successively. From this it is also appare nt that the representation of time is itself an intuitio n, since its relations can be expressed in an outer intuitio n. c) Time is the a priori formal conditi on of all appearances in general . Space, as the pure form of all outer intuitions, is limited as an a priori condition merely to outer intuitions. But since, on the contrary, all representations, whethe r or not they have outer things as their object, nevertheless as determ ination s of the mind themselves belong to the inner state, while this inner state belongs under the formal conditi on of inner


B 49




Doctrine of Elements. Part I. The Transcendental Aesthetic <A> time, so time is an a priori conditi on of all apintmtl.On, pearanc e in general, and indeed the immed iate conditi on of the inner intuitio n (of our souls), and thereby also the mediat e conditi on of outer and appearances. If I can say a priori: all outer appearances are in space B 5I the from so space, of s accordi ng to the relation determ ined a in nces appeara a all ly: princip le of inner sense I can say entirely general stand rily necessa objects of the senses, are in time, and general, i.e., in relation s of time. by If we abstrac t from our way of interna lly intuitin g ourselves and the in ns intuitio outer all means of this intuitio n also dealing with take objects as they may he in thempower of represe ntation , and of objective validity in regard to selves, then time is nothing . It is of appearances, because these are already things that we take as objects senthe from ts abstrac it is no longer objective if one A35 our senses ; pesibility of our intuitio n, thus from that kind of represe ntation that is merely re therefo is things in genera l. Time culiar to us, and speaks our (human) intuitio n (which is always sensia subjective conditi on the ble, i.e., insofar as we are affected by objects), and in itself, outside to regard in e subject, is nothing . Noneth eless it is necessarily objectiv us before come things that can all appearances, thus also in regard to are in time, because with the things say cannot We nce. in experie inconcep t of things in general abstrac tion is made from every kind of B 52 belongs time which tuition of them, but this is the real conditi on under the to the represe ntation of objects. Now if the conditi on is added to of (objects nces concep t, and the princip le says that all things as appeara e objectiv sound its sensible intuitio n) are in time, then the princip le has correct ness and a priori universality. Our assertions accordi ngly teach the empiri cal reality of time, i.e., our objective validity in regard to all objects that may ever be given to be ever can object no senses. And since our intuitio n is always sensible, of n conditio the given to us in experience that would not belong under reabsolute to time claim of time. But, on the contrary, we dispute n ality, namely where it would attach to things absolutely as a conditio . intuition sensible our or propert y even withou t regard to the form of A 36 be never ean lves, Such propert ies, which pertain to things in themse ngiven to us throug h the senses. In this therefo re consists the transce abone if all at nothing dental idealit y of time, accordi ng to which it is stracts from the subjective conditi ons of sensible intuitio n, and cannot es be counte d as either subsisting or inherin g in the objects in themselv combe to is ideality (withou t their relation to our intuitio n). Yet this with the subrept ions of sensati on just as little as that of space is, pared 53 B because in that case one presupp oses that the appeara nce itself, in which


Section n. On Time <A> these predicates inhere, has objective reality, is here entueilV sent except insofar as it is merely empirical, i.e., object itself is regarded merely as appearance: concer ning which the above remark in the previous section is to be consulted.a,b Elucid ation. Against this theory, which concedes empirical reality to time but putes its absolute and transce ndental reality, insightful men so unanimously propos ed one objecti on that I conclud e that it must naturally occur to every reader who is not accustomed to these COJ[1sid eJrations. 20 It goes thus: Alterations are real (this is proved our own represe ntation s, even if one would deny all outer appearances together with their alterations). Now alterations are possible in time, therefo re time is someth ing real. There is no difficulty in answer ing. I admit the entire argume nt. Time is certain ly someth ing namely the real form of inner intuitio n. It therefo re has subJc:ctlve ity in regard to inner experience, i.e., I really have the represe ntation of time and of my determ ination s in it. It is therefo re to be regarde d really not as object:" but as the way of represe nting myself as object. e But if! or another being could intuit myself withou t this conditi on sensibility, then these very determ ination s, which we now represe nt to ourselves as alterations, would yield us a cogniti on in which the represe ntation of time and thus also of alterati on would not occur at all. Its empirical reality therefo re remains as a conditi on of all our experiences. Only absolute reality cannot be granted to it accordi ng to has been adduced above. It is nothing except the form of our inner intuitio n. * If
* I can, to be sure, say: my representations succeed one


B 54

but that means that we are conscious of them as in a temporal sequence, i.e., ing to the form of inner sense. Time is not on that account something in itself, nor any determination objectively adhering to things.

, This refers to A28-30/ B44-5 in 3. b Inserted in Kant's copy, before the next section: "Space and time are not merely logical fonns of our sensibility, i.e., they do not consist in the fact that we represen t actual relations to ourselves confusedly; for then how could we derive from them a priori synthetic and true propositi ons? We do not intuit space, but in a confused manner; rather it is the form of our intuition . Sensibility is not confusio n of represen tations, but the condition of consciousness." (E x'XVIII, p. 20; 23:23) , Kant's copy adds: "So is space. This proves that here a reality (consequ ently also individual inhrition) is given, which yet always grOlmds the reality as a thing. Space and time do not belong to the reality of things, but only to our represen tations." (E XXIX, p. 20; 23'14) d Object , Objects


Doctrine of Elements. Part 1. The Transcendental Aesthetic <I\> then the one removes the special conditi on of our sensibility from concep t of time also disappears, and it does not adhere to the objects themselves, rather merely to the subject that intuits them. ! A 38 The cause, however, on accoun t of which this objection is so unanithose who nevertheless know of nothing mously made, and indeed ' is convincing to object against the doctrin e of the ideality of space/ B 55 this. They did not expect to be able to demon strate the absolute reality idealism, accordof space apodictically, since they were confron ted any strict proof; of capable not is objects outer of reality the ing to which on the contrary, the reality of the object of our inner sense (of myself and my state) is immed iately clear throug h consciousness. The former could have been a mere illusion, but the latter, according to their opinion, is undeni ably someth ing reaL But they did not consider that both, bewithou t their reality as represe ntation s being disputed, nevertheless oblong only to appearance, which always has two sides, one where the be to is it which in way the to regard t (withou itself in ject" is considered alreason very that for must r howeve which of ution intuited , the constit n of ways remain problematic), the other where the form of the intuitio itin object the in sought be not must which red, this object is conside really eless neverth which but , appears it which to self but in the subject and necessarily pertains to the represe ntation of this object. Time and space are accordingly two sources of cogniti on, from which ly differen t synthetic cognitions can be drawn a priori, of which especial A 39 s relation its and space of ons cogniti the to regard in atics pure mathem 23 the provides a splendi d example. Both taken togethe r are, namely, B 56 c syntheti e possibl make thereby and n, intuitio pure forms of all sensible ne determi on cogniti of sources priori a these But a priori propositions. ns their own boundaries by that very fact (that they are merely conditio are they as far so only objects to apply they that namely of sensibility), conside red as appearances, but do not present things in themselves. obThose alone are the field of their validity, beyond which no further further, time, and space of reality This place. takes jective use of them just leaves the certain ty of experiential cogniti on untouc hed: for we are things the to adhere rily necessa forms these r whethe as certain of that , in themselves or only to our intuitio n of these things. Those, however who assert the absolute reality of space and time, whethe r they assume it to be subsisting or only inherin g, must themselves come into conflict with the principles b of experience. For if they decide in favor of the first na(which is generally the positio n of the mathem atical investigators of sisting self-sub infinite and eternal two assume must ture)/4 then they non-en tities (space and time), which exist (yet withou t there being any2


b Pri77Cil~icn


Section H. On Time <A>

thing real) only in order to compre hend everyth ing real selves. If they adopt the second positio n (as do some metaphysicians nature), and hold space and time to be relations of appearances (next to or successive to one another ) that are abstracted from experie nce though confusedly represe nted in this abstraction, they must dispute the validity or at least the apodictic certain ty of a priori malthc~m atical doctrines in regard to real things (e.g., in space), since this certaijrr ty does not and on this view apriori concepts of space and creatures of the imagination, the origin of which must really in experience, out of whose abstracted relations imagsometh ing that, to be sure, contains is general in cannot occur withou t the restrictions that nature attached to 25 The first succeed in openin g the field of appeara nces for mathematical assertions; however, they themselves become confused throug h precisely these conditions if the unders tanding go beyond this field. The second succecd, to be sure, with respect to the latter, in that the represe ntation s of space and time do not stand in way if they would judge of objects not as appearances in relation to the underst anding ; but they can neither offer any ground for the possibility of a priori mathem atical cognitions (since lack a true and objectively valid a priori intuitio n), nor can they bring propos itions of experience into necessary accord with those assertions. On our theory of the true constit ution of these two original forms of sensibil ity both difficulties are remedied. a Finally, that the transce ndental aesthetic cannot contain more than these two elements, namely space and time, is dear from the fact all other concepts belong ing to sensibility, even that of motion , unites both elements, presuppose someth ing empirical. 26 For this presupposes the percep tion of someth ing movable. In space considered in itself there is nothing movable; hence the movable must be someth ing that is found in space only throug h experie nce, thus an empiric al datum. In the same way the transce ndental aesthetic cannot count concept of alterati on among its a priori data; for time itself does not alter, but only someth ing that is within time. For this there is reC!Uil"e d the perception of some existence and the succession of its determ inations, thus experience. b


B 57

A41 B


" Inserted in Kant's copy: "Leibniz's system of space and time was to transform both into intellectual but confused concepts. But from this the possibility ofa priori cognition cannot be understo od, for in that case both must precede." (E XXX, p. 20; 23:24) b Inserted in Kant's copy: "Conclus ion: That space and time of course have objective re ality, but not for what pertains to things outside of their relation [Relation] to our faculty of cognition, but rather only in relation to it, and thus to the form of sensibility, hence solely as appearances." (E XXXI, p. 2 I; 23:24)


Doctrine of Elements. Part I. The Transcendental Aesthetic <Pc>



General remarks on the transcendental aesthetic. It will first be necessary to explain as distinctly as possible our opinion in regard to the fundamental constitution of sensible cognition in general, in order to preclude all misinterpretation of it. We have therefore wanted to say that our intuition is nothing but the representation of appearance; that the things that we intuit are not in themselves what we intuit them to be, nor are their relations so constituted in themselves as they appear to us; and that if we remove our senses in own subject or even only the subjective constitution general, then the constitution, an relations of objectsa in space and time, indeed space and time themselves would disappear, and as appearances they cannot exist in themselves, but only in us. What may be the case with objects in themselves and abstracted from all this receptivity of our sensibility remains entirely unknown to us. We are acquainted with nothing except our way of perceiving them, which is peculiar to us, and which therefore does not necessarily pertain to every being, though to be sure it pertains to every human being. We are concerned solely with this. Space and time are its pure forms, sensation in general its matter. We can cognize only the former a priori, i.e., prior to all actual perception, and they are therefore called pure intuition; the latter, however, is that in our cognition that is responsible for it being called a posterim'i cognition, i.e., empirical intuition. The former adheres to our sensibility absolutely necessarily, whatever sort of sensations we may have; the latter can be very different. Even if we could bring this intuition of ours to the highest degree of distinctness we would not thereby come any closer to the constitution of objects in themselves. For in any case we would still completely cognizc only our own way of intuiting, i.e., our sensibility, and this always only under the conditions originally depending on the subject, space and time; what the objects may be in themselves would still never be known through the most enlightened cognition of their appearance, which is alone given to us. That onr entire sensibility is nothing but the confused representation of things, which contains solely that which pertains to them in themselves but only under a heap of marks and partial representations that we can never consciously separate from one another, is therefore a falsification of the concept of sensibility and of appearance that renders the entire theory of them useless and empty. The difference between an indistinct and a distinct representation is merely logical, and does not concern the content. Without doubt the concept of right that is used


B 60


B 6I



Section II. On Time <A> the healthy understanding contains very same pracmost subtle speculation can evolve out of it, only in common tical use one is not conscious of these manifold representations in these thoughts. Thus one cannot say that the common concept is sensible and contains a mere appearance, for right cannot appear at rather its concept lies in the understanding and represents a constitution in thicmlseJlves. moral constitution) of actions that pertains to representation of a body in intuition, on the contrary, contains nothIng at all that could pertain to an object in itself, but merely the apllealraJ1Ce of something and the way in which we are affected by this receptivity of our cognitive capacity is called sensibility remains even if one worlds apart from the cognition of the object in see through to the very bottom of it (the appearance). The Leibnizian-Wolffian philosophy has therefore all investigations of the nature and origin of our cognitions to an entirely point of view in considering the distinction between sensibility and intellectual as merely logical, since it is obviously transcendental, and does not concern merely the form of distinctness or indistinctness, its origin and content, so that through sensibility we do not cognize the rather not at constitution of things in themselves merely indistinctly, all, and, as soon as we take away our subjective constitution, the represented object" with the properties that sensible intuition attributes to it is nowhere to be encountered, nor can it be encountered, for it is just this subjective constitution that determines its form as appearance. ? We ordinarily distinguish quite well between that which is essentially attached to the intuition of appearances, and is valid for every human sense in general, and that which pertains to them only contingently cause it is not valid for the relationb to sensibility in general but for a particular situation or organization of this or tllat s~nse. And thus one calls the first cognition one that represents tlle objed in itself, but second one only its appearance. This distinction, however, is empirical. If one stands by it (as commonly happens) and does not that empirical intuition as in turn mere appearance (as ought to pen), so that there is nothing to be encountered in it that pertains to any thing in itself, then our transcendental distinction is lost, and we believe ourselves to cognize things in themselves, altllOugh we nothing to do with anything except appearances anywhere (in the of sense), even in the deepest research into its objects. Thus, we would {'p"rl-",n 1,7


B 62


B 63



Object Here is where Kant switches from Vel~hdltnis to Bczieh7i77g as his topic switches from the relation of objects in space or time to each other to the relation of space and time to us. WIth one exception to be noted, therefore, for the remainder of the section "relation" translates BeziebrmJ!:.


Doctrine of Elements. Part 1. The Transcendental Aesthetic <A> this call a rainbow a mere appearance in a sun-shower, but would the tand unders we as this is correct , as long rain the thing in itself, exal univers in which latter concep t in a merely physical sense, as that dealways is senses the to perienc e and all differen t positions relative this termine d thus and not otherw ise in intuitio n. But if we conside r with ent agreem its to A46 empirical object in general and, withou t turning apevery human sense, ask whethe r it (not the raindrops, since these, as itself, in object an nts pearances, are already empirical objects)a represe is then the questio n of the relation of the represe ntation to the object but nces, appeara mere are these drops transcendental, and not fall even their round form, indeed even the space throug h which they b ions foundat or ations mere modific are nothing in themselves, but of our sensible intuitio n; the transce ndental object, ' however, remains unknow n to us. The second import ant concer n of our transce ndental aesthetic is that it be as as a plausible hypothesis, but it not n1.erely earn some is to that theory a of certain and indubitable as can ever be demand ed we ing convinc fully ty serve as an organo n. In order to make this certain . obvious will choose a case in which its validity can become B 64 Thus, if it were to be supposed that space and time are in themselves objective and conditions of the possibility of things in themselves, then it would be shown, first, that there is a large numbe r of a priori apodictic and synthetic propos itions about both, but especially about space, which we will therefo re here investigate as our primary example. Since and the propositions of geomet ry are cognized synthetically a priori tions, proposi such A47 with apodictic certainty, I ask: vVhence do you take ly and on what does our unders tanding rely in attainin g to such absolute than way other no is necessary and universally valid truths?d There are throug h concepts or throug h intuitions, both of which, however, empirinamely latter, given, as such, either a priori or a posteriori. The l cal concepts, togethe r with that on which they are ground ed, empirica also is that one intuitio n, cannot yield any synthetic propos ition except merely empirical, i.e., a propos ition of experience; thus it can never contain necessity and absolute universality of the sort that is neverthe first the ning Concer geometry. less characteristic of all propositions and only means for attainin g to such cognitions, however, namely throug h mere concepts or a priori intuitions, it is clear that from mere can concepts no synthetic cogniti on but only merely analytic cognition space no lines t B6S be attained. Take the propos ition that with two straigh


, Object d The question mark replaces a period in the text.


Section n. On Time <A> at all can be enclosed, thus no figure is possible, and try to it the concept of straigh t lines and the numbe r two; or take the n"t.n"~l _ cion that a figure is possible with three straigh t lines, and in way try to derive it from these copFepts. All of your effort is in you see yoursel f forced to take ?d'uge in intuitio n, as Q'e()metrv always does. You thus give yoursel f an object in intuition; kind is this, is it a pure a priori intuitio n or an empirical one? If it were latter, then no universally valid, let alone apodictic propos ition could ever come from it: for experience can never provide of this sort. You must therefo re give your object a priori in and ground your synthet ic propos ition on this. If there did not lie in you a faculty for intuitin g a priori; if this subjective conditi on regardi ng form were not at the same time the universal a priori conditi on under which alone the object" of this (outer) intuitio n is itself possible; if the object (the triangle) were someth ing in itself withou t relation to your subject: then how could you say that what necessarily lies in your subjective conditions for constru cting a triangle must also necessarily pertain to triangle in itse1f?b for you could not add to your concep t (of three lines) something new (the figure) that must thereby necessarily be encoun - n66 tered in the object, since this is given prior to your cogniti on not through it. If, therefo re, space (and time as well) were not a mere of your intuitio n that contains a priori conditions under which alone things could be outer objects for you, which are nothing in themselves \\~thout these subjective conditi ons, then you could make out absolutely nothing synthetic and a priori about outer objects. c,28 It is therefo re indubitably certain, and not merely possible or even probable, that space and time, as the necessary conditions of all (outer and inner) experience, A49 are merely subjective conditi ons of all our intuitio n, in to which therefore all objects are mere appearances and given for themselves in this way; about these appearances, further, much may be said a priori that concern s their form, but nothing whatsoever the things in themselves that may ground them.'
" Object b Question mark added. Objccte Verbiilmis t Kant adds Lhree paragrap hs and a conclusio n following this point in the second edition (n In his copy of the first edition, he here inserted the following note, which to some extent outlines the additions to be made in the second: "On the necessity of space and time as a priori conditions belongin g to the existence of things - On the effort nevertheless to remove both from a being that is no object of the senses, God - Mendels sohn. "On the theory of nature: how it is to be seen from that that bodies are mere phe"(EXXX II, p. 21: 23:24)


afElements rm'1SCiene,tentat Aesthetic a



r means a cog niti on may relate to imm edia tely to them , and at which objects, that thro ugh whi ch it relates , an end, is intu itio n. Thi s, however all tho ugh t as a means is directed as <at , but this in turn insofar as tJ.~e obje ct is given to us; takes place a certain min d the cts affe it if least for us humans,> is possible only acquire repr esen tatio ns through the way. The capacity (receptivity) to are objects is called sens ibil ity. Objects way in which we are affected insensibility, and it alone affords us ther efor e giv en to us by means of ugh the underst:mding, and from it tuit ion s; but they are tho ugh t thro whe ther straightaway (directe) or arise con cep ts. But all thou ght, <by mea ns of cert ain marks,> ultithro ugh a deto ur (indirecte), must, , in our case, to sensibility, since mat ely be related to intuitions, thus cts can be given to us. ther e is no othe r way in which obje acity for repr esen tatio n, insofar as The effect of an obje ct on the cap E 34 Tha t intu itio n whi ch is related to the we are affected by it, is sen sati on. A 20 t emp iric al. The und eter min ed objec obje ct thro ugh sens atio n is caned app eara nce . of an empirical intu itio n is called corresponds to sensation its matI call that in the appearance whi ch n ifold of appearance to be ordered'i ter, but that whi ch allows the man

In wha teve r way and thro ugh whateve

" that Kant on of the "Tra nsce nden tal Aesthetic We here pres ent the revised versi majo r changes the to tion addi in e Sinc pare d for the seco nd editi on of the r changes that it d, Kan t also mad e num erou s mino he made, all of whic h will be note ges Kant made chan the all se enclo will we idually, wou ld be cum bers ome to note indiv . Edjtori~! noted e rwis othe , whe ther or not they are in B with in angled brackets ... ated. repe be A will not note s on passages unch ange d from Doc trine of Elements" from Kan t divided the "Tra nsce nden tal b In the seco nd editi on, the end of the "Transcenugh thro " hetic tal Aest the begi nnin g of the "Tra nsce nden ty-seven Ollm epts of the Und ersta ndin g" into twen dent al Ded uctio n of the Pure Conc d for m8teriai adde also were titles new ons, secti bere d sections. h the case of some on. editi r chan ge from the first othe rwis e taken over with out ott'le ..." ited as orde red in certain relations "intu reads this on editi first , In the


The Transce ndental Aestheti c <B> certain relation sa I call the fonn of appeara nce. Since the sensatio ns can alone be ordered and placed in a certain cannot itself be in turn sensatio n, the matter of all appeara nce is only given to us a posteriori, but its form must all lie ready for it in the mind a priori, and can therefo re be conside red separat ely from all sensatio n. I call all represe ntation s pure (in the transce ndental sense) in nothing is to be encoun tered that belongs to sensatio n. Accord ingly pure form of sensibl e intuitio ns in general is to be encoun tered in the mind a priori, wherei n all of the manifo ld of appeara nces is intuited in certain relation s. This pure form of sensibi lity itself is also pure intuitio n. So ifI separat e from the represe ntation of a body which the underst anding thinks about it, such as substan ce, force, di"isibi lity, etc., as well as that which belongs to sensatio n, such as impene trability , hardnes s, color, etc., someth ing from this empiric al intuitio n is still left for me, namely extensi on and form. These belong to the pure intuitio n, which occurs a priori, even withou t an actual object of the senses or sensation, as a mere form of sensibi lity in the mind. I call a science of all princip lesb of a priori sensibil ity the transce ndental aesthet ic.* There must therefo re be such a science , which constitutes the first part of the transce ndental doctrin e of elemen ts, in oppositi on to that which contain s the princip les c of pure thinkin g, and which is named transce ndental logic.
* The Germans are the only ones who now employ the word "aesthetics" to des-


A2 I



ignate that which others call the critique of taste. The ground for this is a failed hope, held by the excellent analyst Baumgarten, of bringing the critical estimation of the beautiful under principles of reason,d and elevating its mles to a science. But this effort is futile. For the putative mles or criteria are merely empirical as far as their <most promine nt> sources are concerned, and can therefor e never serve as <determinate> a priori mles accordin g to which our judgmen t of taste must be directed; rather the latter constitutes the genuine touchsto ne of the correctness of the former. For this reason it is advisable <either> again to desist from the use of this term and preserve it for that doctrine which is true science (whereby one would come closer to the language and the sense of the ancients, among whom the division of cognition into aLuEl'lTCX KCXL VOTjTCX was very well <or else to share the term with speculative philosophy and L1ke aesthetics partly in a transcen dental meaning, partly in a psychological meaning >. already noted at p. 156, note a, with the exception of four cases in its final section, throughout the "Transcendental Aesthetic" Kant characteristically uses the term Ve,.hiill77is, connoting a relation among objects, rather than Bezichung, connoting a relation between subject and object; thus, unless otherwise noted, "relation" or its plural translates Verha/tIlis or its derivatives.

A 2 I / II



a As



scendental Aesthetic <B> Doctrine of Elements. Part I. The Tran ther efor e first isol ate sensiIn the tran scen den tal aesthetic we the und erst and ing thinks by sepa rati ng off eve ryth ing that g but em.pirical intu itio n remains. thro ugh its concepts, so that not hin to the latte r everything that belongs Second, we will then deta ch from mere the and n itio except pur e intu sensation, so that noth ing remains thin g that sensibility can make the is ch form of appearances, whi n it will be found that ther e are two available a priori. In this investigatio , as principles" of a priori cognition pur e forms of sensible intu itio n be now win we ch assessment of whi nam ely space and time, with the con cern ed. The Tra nsc end enta l Aes thet ic Firs t Sec tion On spac e.


con cep t.> Met aph ysic al exp osit ion of this d) we represent to oute r sense (a pro pert y of our min By means space thei r shape, In e. spac all as in ourselves objects as outside us, and or determinable. , ned rmi ther is dete mag nitu de, and rela tion to one ano its inner state, or f, itsel its intu min d Inn er sense, by mea ns of which the is the soul itself, as an object;h yet it gives, to be sure, no intu itio n of ch the intu itio n of its inne r state is still a dete rmi nate form, und er whi A23 r determinaeverything that belongs to the inne alone possible, so e be intuited mor no can e time. Tim tion s is repr esen ted in relations of what are Now us. in ng ethi as som externally than space can be intu ited ations rmin dete only they Are ies?C space and time? Are they actual entit wou ld pert ain to them even if they or relations of things, yet ones that of tions that only atta ch to the form wer e not intu ited , or are they rela , subjective con stitu tion of our mind intu itio n alone, and thus to the B 38 ld not be ascribed to any thing at allt with out whi ch thes e predicates cou ut this, we will <expound the concept In ord er to inst ruct ourselves abo d exp osit ion (expositio) the distinct of space> first. <I und erst and by n of that whi ch belongs to a con(even if not complete) repr esen tatio ysic al whe n it contains that which cept; but the exposition is met aph ri.> exhibits the con cep t as give n a prio t that has bee n draw n from outet cep con l irica emp an not is I)

Object , wirkliche TVesen t consider space." d In the first editi on: "firs


Section I. On Space <B>

e~.~.i~I!<;~S. For order for sensations to to something-Blltside me (i.e., to someth ing in anothe r place in space from in which I find myself), th-qs in order for 1Ue to represe nt them as outside <and lle.x:t te one thus as in diff~!elltplaces, the represe ntation ground. Thus the represe ntation 'C,~"",", "_.~ ",,,~ Qbt:uned relation sof ()uter appearance experience, rience'lsli~lLfu~K!20ssibl~Q!lIymE()llgh mis reI)reSerltal:101l 2) Sp~ce i~:l_~ece~~:l!I.I"epEesentation, a priori, that is of all Ql:l!(;I"Jf.l.~t:!Qns. One canl1ev~Erepresent that there is no space, though one can verywe ll think that there are no objects to encoun tered in It~Ii:'lstherefore to be regarde d as the conditi on bilitY:~f:1EQe~r:~!lces, not as a determ ination depend ent on and is an a priQ!iEeJlI~en1:atiQn that necessarily grounds outer appearances. b <3- is not a discursive or, as is said, general concep t of relations of things in general, but a pure intuition. For, first, one can represent a single space, and if one speaks of many spaces, one only parts of one and the same unique space. these parts cannot as it were precede the single all~encompassingspace as its c,QIDponents (from which its compo sition would be possible), buuath er areQcl ythoug ht in it. Itis essentially single; the manifold in it, also the general concep t of spaces in general, rests merely on limitations. From this it follows that in respect to it intuitio n (which is not empirical) grounds all concepts of it. Thus also all geometrical principles, e.g., that in a triangle two sides togethe r are always greater than the third, are never derived from general concepts of line and triangle, but rather are derived from intuitio n and indeed derived a with apodictic certainty. <d4 ) SJ:l.,!c~is represe nted as an infinite givenm agl1itu de. Now one must, to be sure, think of every concep t as a represe ntation that is contained in an i~finite set of different possible representations (as common mark), which thus contains these under itself; no concept, as such, can be though t as if it contained an infinite set of representations within itself. Nevertheless space is so though t (for all parts of space, even to infinity, are simultaneous). Theref ore the nalrep[(;s(;l1t;ltion of sp:lc<::e is an intuitio n, not a concep t.>
.. .. "'C.





In first edition there follows a paragrap h (3) (at A24 above) that is replaced by the "Transcendental Expositio n of the Concept of Space" in the second (see B 40-1 below); the following paragraphs, (3) and (4), were thus originally numbere d (4) and (5); the content of the original paragrap h (5), now renumbe red (4), is also changed. , In the first edition: "of them," i.e., the limitations of space. d As previously mentione d, the content of this paragrap h is changed from the first edition.


Tran scen dent al Aes thet ic <B> Doc trine of Elem ents . Part I. The

con cep t of spac e. Tra nsc end enta l exp osit ion of the explanation of a a tran sce nde nta l exp osit ion the LlJn ders tand other DJ3~~f hcp insightintot from con cep t as a r) ired requ is it aim this C()gJ.1il:jgl.1~Call!:Jegain ed. For syn thet ic that 2) and t, cep con given cognitions actually flow from the i.hit pres upp osit ion of a given the er und e sibl pos these cognitions are only way of explaining this concept. nes the pro pert ies of space synGeo met ry is a science that dete rmi mus t the repr esen tatio n of space thetically and yet a priori. Vilhat then possible? It mus t originally be intube for such a cog niti on of it to be beprop osit ions can be drawn that go ition; for from a mer e con cep t no B 41 n ctio odu (Intr ry met , hap pen s in geo yon d the concept, which, however all to r prio i.e., ri, prio oun tere d in us a V). But this intu itio n mus t be enc t be pure , not empirical intuition. mus it thus perc epti on of an object, apodictic, i.e., com bine d with conFor geometrical prop osit ions are space has only thre e dimensions; but sciousness of thei r necessity, e.g., l or judg men ts of experience, nor such prop osit ions can not be empirica II). infe rred from them (Int rod ucti on inha bit the min d that precedes the n itio intu Now how can an oute r the con cep t of the latt er can be deobjects b themselves, and in which rwis e than insofar as it has its seat term ined a priori? Obviously not othe con stitu tion for bein g affected by obmer ely in the subject, as its formal ate rep rese nta tion , i.e., intuition, jectsC and ther eby acquiring imm edi oute r sen se in general. of them, thus only as the form of the pos sibi lity of geometry as a Thu s our explanation alone makes ensi ble. Any kind of explanation syn thet ic a priori cog niti on com preh n if it appears to have some similarthat does not acco mpl ish this, eve ingu ishe d from it by means of this ity with it, can mos t surely be dist characteristic.>29
A26 /B4 2

< 3

cep ts. Con clus ion s from the abo ve con all of any thin gs in themselves a) Space repr esen ts no pro pert y at r, i.e., no dete rmi nati on of them any rela tion of them to each othe that would rem ain even if one attaches to objects themselves and diti ons of intu itio n. For neither a to abst ract from all subjective con can be intu ited prio r to the existen solute nor relative dete rmi nati ons thus be intu ited a priori. of the thin gs to whi ch they pertain,

Objecten , Objecten


Section I. On Space <B>

b) Space is nothin g other than merely the outer sense, i.e., the subjective conditi on of "Ll1""UUJ LCy, alone outer intuitio n is possible for us. Now since the of the subject to be affected by objects necessarily precedes all intuitio ns of these objects, it can be unders tood how the form of all appearances can be given in the mind prior to actual perceptions, a and how as a pure intuitio n, in which objects must contain principles a of their relations prior to all experience. We can accordingly speak of space, extended beings, and so on, from the human standpo int. If we depart from the subjective conditi on under which alone we can acquire outer intuitio n, nClm,'lv which we may be affected by objects, then the represe ntation of space signifies nothing at all. This predica te is attribut ed to things insofar as they appear to us, i.e., are objects of sensibility. The constan t of this receptivity, which we caU sensibility, is a necessary conditi on of all the relations within which objects can be intuited as outside us, if one abstracts from these objects, it is a pure bears the name of space. Since we cannot make the special conditi ons of sensibi lity into conditions of the possibility of things, but only of their appearances, we can well say that space compre hends all things that may appear to us externally, but not all things in themselves, wr,pH1P r be intuited or not, or by whatever subject they may be intuited . For we cannot judge at all whethe r the intuitio ns of other thinkin g beings are bound to the same conditions that limit our intuitio n and that are univcrsallyvalid for us. If we add the limitati on of a judgme nt to concept of the subject, then the judgme nt is uncond itionall y The proposition: "All things are next to one anothe r in space," is valid the limitation that these things be taken as objects of our sensible intuition. If here I add the condition to the concep t and say "All things, as outer intuitions, are next to one anothe r in space," then this rule is universally and withou t limitation. Qurexp osition s .accordingly teach thu~a1ity (i.e., objectiYe validity) of space regard to everyth ing come before lJ.sext<.;wally as an object, the same time ideality. spacei n regard to things when they are considered in selves through reason, i.e., withou t taking accoun t of the constit ution of oul'-sellsioility.We therefo re assert empiri cal reality of space respect to all possible outer experience), though to be sure its transcendentaIl.deaHty, i.e., that it is nothing as soon as we leave aside condition of the possibility of all experience, it as SOllletl11ng that grounds the things in themselves. Besides space, however, there is no other subjective represe ntation


Doctrine of Elements. Part I. The Transcendental Aesthetic <B> called a priori objective. could re1ateda to someth ing itions from any such propos priori a ic b<For one cannot derive synthet ( 3). Strictly speakspace in n intuitio represe ntation , as one can from h they coincide althoug them, to not does ing, therefo re, ideality the subjective to only ing belong in space of with the represe ntation , and feeling, hearing sight, of e.g., sense, of constit ution of the kind however, which, , warmth , sounds throug h the sensatio ns of colors, in themnot do ns, intuitio not and ns since they are merely sensatio C priori.> a all of least d, cognize be selves allow any object to The aim of this remark is only to preven t one from thinkin g of illusB45 examtrating the asserted ideality space with comple tely inadequ ate not as red conside ly correct are etc., taste, ples, since things like colors, even can which , subject our of ons alterati qualities of things but as mere ly original is which that case this in For people. be differen t in differen t a as sense al empiric an in counts rose, a itself only appeara nce, e.g., to regard in eye every to t differen appear can thing in itself, which A 30 concolor. The transce ndental concep t of appeara nces in space, on the in intuited is that nothing ely absolut trary, is a critical remind er that to proper is that form a not is space space is a thing in itself, and that known not are lves themse in objects anythin g in itself, but rather that that what we call outer objects are nothing other than to us at all, mere represe ntation s of our sensibility, whose form is space, but whose true correla te, i.e., the thing in itself, is not and cannot be cognized throug h them, but is also never asked after in experie nce. B46 The Transc endent al Aesthe tic Second Section time.

< 4
Metaph ysical exposit ion of the concep t of time.> not an empiric al concep t that is someho w drawn from an Time is come experience. For simulta neity or success ion would not themselves a P11'them percep tion if the represe ntation of time did not ground things several that Only under its presupp osition can one represe nt times exist at one and the same time (simult aneousl y) or in differen t


A 3I

(successively). s all intuitions. In 2) Time is a necessa ry represe ntation that ground one regard to appeara nces in general one cannot remove time, though


above, first edition, the remainder of this paragrap h reads differently; see A28-9

, Object


Section n. On Time


can very well take the appearances away from time. Time is therefo re given a priori. In it alone is all actuality of appearances possible. ter could disappear, but time itself (as the universal conditi on of their possibilityY cannot be removed. 3) This a priori necessity also ground s the possibility of B47 principles of relations of time, or axioms of time in general. It has one dimension: differen t times are not simultaneous, successive Gust as differen t spaces are not successive, but simultaneous). These principles could not be drawn from experience, for this would neither strict universality nor apodictic certainty. We would able to say: This is what commo n percept ion teaches, but not: This is ters must stand. These principles are valid as rules experiences are possible at and instruc t us prior to through it. b 4) T@~isl}Q_discursiveor,asone calls it, 5""""<" conce]pt,brlt fOillLofsensLbleintuition. Di:[fel:enttime',sare same_time. Thatr~jJresentation, however, A 32 through a single object; is-an il.ltrlition. diffe.renttimescannot be simultaneous ef3Lconcept. The propos itionjss yntheti c, and cannot arise concepts.alone. It is therefo re contain ed in the intuitio n and repr..esentaJion of 5) The infinitude of time signifies nothing more than that every determinate magnit ude of time is only possible throug h limitations of a B48 single time ground ing it. The original represe ntation time must fore be given as unlimited. But where the parts themselves every magnitude of an object can be determ inately represe nted only through limitation, there the entire represe ntation cannot be given throug h concepts, (<for they contain only partial represe ntation s," but immediate intuition must grOlmd them. d < 5 Transc endent al exposit ion of the concep t of time. I can appeal to No. 3 where, in order to be brief, I have placed that which is properly transce ndental under the heading of the metaphysical exposition. Here I add further that the concep t of alterati on and,
" These parentheses added in B. b The text reads "bclcbren uns vor dcnelbm, und nicbt durcb dieselbe." Earlier editors suggested emending the last word to "dicselben"; but if the sentence is interpret ed to mean "instructs us prior to experiences, not through common perceptio n," it can be read without emendation. In the first edition: "for there the partial represen tations precede. " d Bhas ibncn instead of ibre here.


Tran scen dent al Aes theti c <B> Doc trine of Elem ents . Part 1. The

possible is mot ion (as alte rati on of place), it, the con cep t if this representation of tim e thro ugh and in the repr esen tatio n wha teve r it might (inner) intu itio n, then no concept, wer e not a a possibility of an alteration, i.e., of be, cou ld make com preh ensi ble the g bein osed predicates (e.g., a thing's com bina tion of con trad icto rily opp very same thin g in the same place) in a place and the not- bein g of the a in tim e can both con trad icto rily opin one and the same object. Onl y g be enc oun tere d, nam ely succespos ed dete rmi nati ons in one thin B 49 e explains the possibility of as much sively. Our con cep t of time ther efor ente d by the general theo ry of mosynthetic a priori cog niti on as is pres tion, which is no less fruitfu1.>30
< 6>

Con clus ions from thes e con cep ts. A32 s ld subsist for itself or atta ch to thing a) Tim e is not som ethi ng that wou from thus remain if one abstracted as an objective dete rmi nati on, and itio n of them; for in the first case it all subjective con diti ons of the intu far al yet with out an actual object. As would be som ethi ng that was actu , time could not precede the second case is con cern ed, however as A 33 s er atta chin g to the thin gs themselve objects as a dete rmi nati on or ord d and intu ited a priori thro ugh synas thei r con diti on and be cognize r on the contrary, can very well occu thet ic propositions. But the latter, all h subjective con diti on und er whic if tim e is noth ing oth er than the then this form of inne r intuition can intu itio ns can take place us. For thus a priori. be repr esen ted prio r to the objects, the informofinI1e~sense,i:e., of b) Tim e is not hin g othe r than the e. FQr time can not beadeterminatuition-O[c;1iLseTfarid our inne r stat ngs neit her to a shape or a position, tion of oute r appearances; it belo B 50 rela tion of representations in but on the con trar y dete rmi nes the ds no shape we just because this inne r intu itio n yiel analogies, and represent the atte mpt to rem edy this lack thro ugh prog ress ing to infinity, in which the tem por al sequence thro ugh a line is of only one dimension, and infer manifold constitutes a series that all the prop erti es of time, with the from the prop erti es of this line to form er are simultaneous but those sole difference that the parts of the . Fro m this it is also apparent that of the latte r always exist successively since-an its relations can an the repr esen tatio n of time is itse lf be expressed in an alIter lnt1l11101:l. )n of all appe:-ll:'~I1C:E:~jI1g~eral, c)T iI11ejs the a priori formal conditi( A 34 r intuitions, is limited as an a Space, as the pur e form of all oute



Section n. On Time <B>

determ ination s of the mind th(~m:5e1'ves ""'''-U''"'~ .. f!''H'~' this innerst ;Ite belongs IllmitiDD, and tIms of time, so pearance in general, and indeed immed iate COndltl()~.()t...~~e._ mtmtion (ofoU:i'souls), and thereby also the mediate cOlad.luOill appearaiices. If I can say a priori: all outer appearances are space and B 5I -aetermined a priori according to the relations of space, so from the principle of inner sense I can say entirely generally: all appearances in general, i.e., all objects of senses, are in time, necessarily in relations of time. If we abstract from our way of interna lly intuitin g ourselves and means of this intuitio n also dealing with all outer intuitio ns in the power of represe ntation , and thus take objects as they may be in selves, then time is nothing . It is only of objective validity in regard to appearances, because these are already things that we take as objects of our senses; but it is no longer objective if one abstracts from the senA35 sibility of our intuitio n, thus from that kind of represe ntation that is peculiar to us, and speaks of things in genera l. Time is therefo re merely a subjective conditi on of our (human) intuitio n (which is sensible, i.e., insofar as we are affected by objects), and in itself, outside the subject, is nothing . Noneth eless it is necessarily objective in regard to all appearances, thus also in regard to an things that can come before us in experience. We cannot say things are in time, because concept of things in general abstrac tion is made from every of in- B 52 tuition of them, but this is the real conditi on under which time belongs to the representation of objects. Now if the conditi on is added to the concept, and the principle says that all things as appearallces (objects of sensible intuition) are in time, then the principle has its sound objectiv e correctness and a priori universality. Our assertions accordingly teach the empiri cal reality of time, i.e., objective validity in regard to all objects that may ever be given to our senses. And since our intuitio n is always sensible, no object can ever given to us in experience that would not belong under the conditi on of time. But, on the contrary, we dispute all claim of time to absolute reality, namely where it would attach to things absolutely as a conditi on or property even withou t regard to the form of our sensible A36 Such properties, which pertain to things in themselves, can never ~ven to us through the senses. In this therefo re consists the transce ndental ideality of time, according to which it is nothing at if one state.




scendental Aesthetic Doctrine of Elements. Part L The Tran cannot sensible subjective stracts cts in themselves subsisting or inhe ring in the obje be cou nted as is to be comlity idea this Yet to our {"rit- hICW lt- thei r rela tion space is, of sensation just as little as that of pare d with the eara nce itself, in which case one presupposes that the app because in is here entirelyabe reality, thes e pred icat es inhe re, has objectiv obje ct itse lf is rethe i.e., l, empirica sen t except insofar as it is mer ely remark in the ch whi cern ing gard ed mer ely as appearance: con a sulted. the previous sections is to be con

to time but diswhi ch concedes emp irica l reality Aga inst this ul men have so ghtf insi ity, tran scen den tal real pute s its absolute it mus t natuthat e clud con I n that una nim ous ly pro pos ed one obje ctio accu stom ed to thes e consideraocc ur to every read er who is not the change of real (this is prov ed tion sY It goes thus: Alterations are appearances r oute all one wou ld our own repr esen tatio ns, even if A 37 ible only in poss are ns Now alteratio toge ther with thei r alterations). in answerty icul diff no is real. The re time, ther efor e time is som ethi ng ng real, ethi som y ainl cert t. Tim e is ing. I adm it the enti re argu men e realectiv subj has e efor ther itio n. It nam ely the real form of inne r intu n of tatio esen repr the e hav ly i.e., I real ity in rega rd to inne r experience, rerded rega be to e efor ther it. It is d But tim e and <my>b dete rmi nati ons in B 54 ct. obje as elf mys ting of repr esen ally not as object" but as the way mys elf with out this con diti on of senit intu ld cou g bein if I or ano ther ons , whi ch we now repr esen t to oursibility, then these very dete rmi nati us a cog niti on in whi ch the repre selves as alterations, wou ld yield Its alte rati on wou ld not occ ur at all. sen tatio n of time and thus also of as a con diti on of all our experiences. emp irica l reality ther efor e remains ted to it acco rdin g to wha t has been Onl y absolute reality can not be gran the form of our inne r intuition! If add uce d above. It is noth ing exc ept
only tations succeed one another; but that * I can, to be sure, say: my represen d accor i.e., , ence sequ l pora as in a tem means that we are conscious of them not on that account something in it

is ing to the form of inner sense. Time adhering to things. ly ctive obje n atio rmin dete any nor self,
This refers to A 28-3 o!B4 4-5 In the first edition: "of my."

in 3.


Object Object


Section n. On Time <B> one removes special conditi on of our sensibility concept of time also disappears, and it does not adhere to themselves, rather merely to the subject that intuits them. The cause, however, on accoun t of which this objecti on is so unanimously made, and indeed by those who nevertheless of notluug convincing to object against the doctrin e of of space,3 is this. They did not expect to able to demon strate the ab~;o1111te of space apodictically, since they were confron ted according to which the reality of outer objects is not capable of any strict proof: on the contrary, the reality of the object of our inner sense myself and my state) is immed iately clear throug h consciousness. former could have been a mere illusion, but the according to their opinion, is undeni ably someth ing real. But they did not conside r that both, withou t their reality as represe ntation s being nevertheless belong only to appearance, which always has two sides, one where the object O is considered in itself (without regard to way in which it is to be intuited , the constit ution of which howeve r must that very reason always remain problematic), the other where the of the intuition of this object is considered, which must not be in the object in itself but in the subject to it appears, nevertheless really and necessarily pertain s to the represe ntation object. ~~,.~.,~"......",..~. space. are accordi ngly two sources of cogniti on, whLc:.llAiffere.l1t syn.thetic cognitions can be drawn a priori, pkCially purema themat regard of space relatillnsprovides a splendid example. togethe r are, namely, the pure forms of all sensible and thereby make si:resynthetic a priori propos itions. But these a priori sources of cog~itiol1determinc their oWn bounda ries by that very are m~rely conditions of sensibility), 0111ys()far as they are conside red as appearances, t4il1gsil1 themselves. Those alone are the field of their vaJlld.ltv, UC\'UllU which no further objective use of them takes place. reality of space and time, further, leaves the certain ty of experiential cogniti on untouched: for we are just as certain of that whethe r these forms necessarily adhere to the things in themselves or only to our intuitio n of these things. Those, however, who assert the absolute reality of space and time, whethe r they assume it to be subsisting or must themselves come into conflict with the principles b of experience.







Doctrin e of Elemen ts. Part 1. The Transce ndental Aesthet ic <B>

is generally the position of the first For if they decide in ,B of the mathem atical investigators of nature) then they must assume two eternal and infinite self-subsisting non-en tities (space and time), to which exist (yet withou t there being anythin g real) only in order secthe adopt they If lves. themse compre hend everyth ing real within positio n (as do some metaphysicians of nature) , and hold space A40 one and time to be relation s of appearances (next to or successive to repredly confuse though anothe r) that are abstrac ted from experience B 57 sented in this abstrac tion, then they must dispute the validity or at least to the apodictic certain ty of a priori mathem atical doctrin es in regard posteria occur not does ty certain real things (e.g., in space), since this ori, and on this view the a priori concep ts of space and time are only creatur es of the imagin ation, the origin of which must really be sought in experience, out of whose abstrac ted relation s imagin ation has made someth ing that, to be sure, contains what is general in them but that to cannot occur withou t the restrict ions that nature has attached mathfor nces appeara of them. 34 The first succeed in openin g the field ematical assertions. a However, they themselves become very confused bethroug h precisely these conditi ons if the unders tanding would go latthe to respect with sure, yond this field. The second succeed, to be ter, in that the represe ntation s of space and time do not stand in their reway if they would judge of objects not as appeara nces but merely in for ground any offer neither lation to the unders tanding ; but they can a lack they (since ons cogniti the possibility of a priori mathem atical the bring they can nor n), true and objectively valid a priori intuitio s. propos itions of experience into necessary accord with those assertion A41 of forms l origina two these of On our theory of the true constit ution B 58 sensibility both difficulties are remedi ed. Finally, that the transce ndental aesthetic cannot contain more than these two elements, namely space and time, is clear from the fact that all other concepts belong ing to sensibility, even that of motion, which unites both elements, presuppose someth ing empirical. For this prein supposes the percep tion of someth ing movable. In space considered g somethin be must e movabl the itself there is nothing movable; hence empirical an thus nce, experie h that is found in space only throug datum. In the same way the transce ndental aesthetic cannot count the concep t of alterati on among its a priori data; for time itself does not alter, but only someth ing that is within time. For this there is required the percep tion of some existence and the succession of its determinations, thus experience.

A colon in the first edition is replaced with a period i.n the second.


Section n. On Time <B>

< 8>
(;enera lreruar ks on the transce ndenta l aesthet ic


d.>o It will first be necessary to explain as distinct ly as possible our opinion in regard to tlle fundam ental constit ution of sensible A42 in general, in order to preclud e misinte rpretati on it. We have therefo re wanted to say that allQtLLjJltuitiQI1-l~;.Jlot1lill.g therepr esentat ieRea PI*aF3 nce; that the things that we intuit ::lre not in themselves what we intuit them to be, nor are their so constituted in themselves as they appear to us; and that if we remove our own subject or even only the subjective constit ution of senses in general, then all constitu tion, all relation s of objects b in space and indeed space and time themselves would disappear, and as appearances they cannot exist in themselves, but only in us. What may be case with objects in themselves and abstrac ted from all this recepti vity sensibility remains entirely unknow n to us. We are nothing except our way of perceiving pe,cll1iar'tg which therefore does not necessarily pertain to every be sure it pertain s to every human being. this. Space and time are its pllre forms, sensati on in general its B 60 We can cognize only the former a priori, i.e., prior to all actual perception, and they are therefo re called pure intuitio n; the latter, howeve r, is that in our cogniti on that is responsible for it being called a fJos,tcr;;o ri cognition, i.e., empirical intuitio n. The former adheres to our sensibil ity absolutely necessarily, whatev er sort of sensations we may have; the latter can be very different. Even if we could bring this intuitio n of ours A43 to the highest degree of distinctness we would not thereby come any closer to the constit ution ef objects in themselves. For in any case we would still comple tely cognize only our own way of intuitin g, i.e., our sensibility, and this always only under the conditi ons originally aelperlQ ing on the subject, space and time; wha.t_th~()Qjects ---".J_.."'_" _.~ ..~_' selves would still never beknow l1 throug h the mest enlight ened cognition of their appearance, which alone is given to us. That our entire sensibility is nothin g but the confused represe ntatien of things, which contain s solely that which pertains to them in themselves but only under a heap of marks and partial represe ntation s we can never conscieusly separate from One another, is therefo re a sification of the concep t of sensibility and of appearance that renders
" "I." is added in the second edition because of the addition of the further numbere d paragraphs (II through N) added at B 66-73. ; Objecte


Doctrin e of Elemen ts. Part 1. The Transce ndental Aesthet ic <B>

empty. The difference betwee n an the entire theory of them useless not indistin ct and a distinct represe ntation is merely logical, and does B61 used is that the concep t of right concer n the content . VVithout the healthy unders tanding contain s the very same things that the in commo n and pracmost subtle specula tion can evolve out of tical use one is not conscious of these manifo ld represe ntation s in these and though ts. Thus one cannot say that the commo n concep t is sensible its rather all; at A44 contain s a mere appeara nce, for right cannot appear (the ution constit a concep t lies in the unders tanding and represe nts pertain s to them in themselves. The moral constitu tion) of actions in intuitio n, on the contrary, contain s nothing represe ntation of a merely the appearance could pertain to an object in itself, at all reof someth ing and the way in which we are affected by it; and this remains and lity ceptivity of our cognitive capacity is called sensibi cogniti on of the object in itself even if one might worlds apart from see throug h to the very bottom of it (the appearance). The Leibniz ian-Wo lffian philoso phy has therefo re directe d all investigation s of the nature and origin of our cogniti ons to an entirely unjust the point of view in conside ring the distinct ion betwee n sensibility and and , ndental intellec tual as merely logical, since it is obviously transce but does not concer n merely the form of distinctness or indistinctness, B 62 the cognize its origin and content , so that throug h sensibility we do not not rather but ctly, constit ution of things in themselves merely indistin and, as soon as we take away our subjective constitu tion, the repat the propert ies that sensible intuitio n attributes to resente d object" just it is nowher e to be encoun tered, nor can it be encoun tered, for it is nce)5 appeara this subjective constit ution that determ ines its form as We ordinar ily distinguish quite well betwee n that which is essentially A45 attache d to the intuitio n of appearances, and is valid for every human besense in general, and that which pertain s to them only contingently b of sensibi lity in general but only for cause it is not valid for the relation one a particu lar situatio n or organiz ation of this or that sense. And thus the but itself, in calls the first cogniti on one that represe nts the object onlyern is r, second one only its appearance. This distinction, howeve regard not does and s) pirical. If one stands by it (as commo nly happen that empirical intuitio n as in turn mere appeara nce (as ought to hapto pen), so that there is nothing to be encoun tered in it that pertains bewe and lost, is ion anythin g in itself, then our transce ndental distinct

Object from VcrhdltJ1i, to As noted in the first-edit ion version above, here Rant switches time to each other or space in objects of relation the from Bez,ich1lng as his topic switches therefore, for to the relation of space and time to us. With one exception to be noted, paragraphs new the In . Verhiiltni, translates " "relation (I) the remainde r of this section . II through N added below, however, Kant again reverts to VerhiiltniI


Section n. On Time <B>

lieve ourselves to cognize things in themselves, though we have ing to do with anythin g except appearances anywhere (in sense), even in the deepest researc h into its objects. Thus, we would n63 certainly call a rainbow a mere appearance in a sun-shower, but would call this rain the thing in itself, and this is correct, as long as we understand the latter concep t in a merely physical sense, as that which in universal experience and all differen t positions relative to the senses is always determ ined thus and not otherw ise in intuitio n. But if we consider this empirical object in general and, withou t turning to its agreement with every human sense, ask whethe r it (not the raindrops, since these, as appearances, are already empirical objects)a represents an ject in itself, then the question of the relation of the represe ntation to the object is transcendental, and not only these drops are mere appearances, but even their round form, indeed even the space throug h which they fall are nothing in themselves, but only mere modifications or foundations b of our sensible intuition; the transcendental object," however, remains unknow n to us. The second import ant concern of our transcendental aesthetic is that it not merely earn some favor as a plausible hypothesis, but that it be as certain and indubit able as can ever be demand ed of a theory that is to serve as an organo n. In order to make this certain ty fully convincing we will choose a case in which its validity can become obvious <and that B 64 can serve to make that which has been adduced in 3 even more clear>. Thus, if it were to be supposed that space and time are in themselves objective and conditi ons of the possibility of things in themselves, it would be shown, first, that there is a large numbe r of a priori apodictic and synthetic propos itions about both, but especially about space, which we will therefo re here investigate as our primar y example. Since the propositions of geomet ry are cognized synthetically a priori and with apodictic certainty, I ask: Whenc e do you take such propositions, A47 and on what does our unders tanding rely in attainin g to such absolut ely necessary and universally valid truths? There is no other way than through concepts or throug h intuitio ns, both of which, however, are given, as such, either a priori or a postcr-iori. The latter, namely empirical concepts, togethe r with that on which they are ground ed, empiric al intuition, cannot yield any synthetic propos ition except one that is also merely empirical, i.e., a propos ition of experience; thus it can never contain necessity and absolute universality of the sort that is neverth eless characteristic of all propos itions of geometry. Concer ning the first and only means for attainin g to such cognitions, however, namely
, Objecte Object


scendental Aesthetic <B> Doctrine of Elements. Part I. The Tran e intuitions, itis- clea r-th at-f rom mer thro ugh Inere con cep ts or a can only mer ely analytic cQgniti~l1_ oon cep tsno syn thet ic cog niti on but s no space osit ion that with two stra ight line B6S ~b@-auained. Take the prop re is possible, and try to derive it from at all can be enclosed, thus no figu num ber two; or take the proposithe con cep t of stra ight lines and the thre e stra ight lines, and in the same tion that a figure is possible with cepts. All ofy our effo rt isilLvain, and way try to derive it from thes e con as inde ed geometry forced to take refuge in intu itio n, see

mLJ:)ljqclil one? Ii~ere the pro pos itio n could let alone of this can never provide ever com e from it: for. experience and a priori in mus t ther efor e give you r sort. a you If theLeclidQQ1JjJ~_in .greJlmd you r form if faculty which the e tim e sam the at were not is itself possible; ifth egb ject (outer) ,alo ne the ect: in itse lf with ollt relation to you r subj (the triangle) Were connecessarily lies in you r subjec:tive then how could you say that wha t mus t also necessarily pert ain to the ditions for con stru ctin g a triangle ) add to you r con cep t (of thre e lines tria ngle in itself? for you cou ld not unbe enco re) that mus t ther eby necessarily B66 som ethi ng new (the figu on and not niti cog r you to r prio n tere d in the object, since this is give tim e as wen) were not a mere form thro ugh it. If, ther efor e, space (and alone priori conditions of you r intu itio n that con tain s s e!ve , whi ch are noth ing il1iliems thin gs could be oute r objects for you y lutel s, then yOll could make.o1,l~ ?:pso with out these subjectivec:ondition b ut oute r objects. It is ther efor e indu noth ing synthetic and a priori abo and e ible or even prob able that spac bitably cert ain and not mer ely poss rience, are diti ons of all (ou ter and inne r) expe A49 time, as the necessary con in rela tion to which mer ely subjective earances and not things given for ther efor e all objects are mer e app e appearances, further, muc h maybe themselves in this way; abo ut thes but noth ing whatsoever about the said a priori iliat con cern s thei r form und them . things in themselves that may gro as theo ry of the ideality of oute r as well c<II. For con firm atio n of this this ces, d the senses, as mer e appearan inne r sense, thus of all objects of eve ryth ing in our cog niti on that becom men t is especially useful: that n, ther efor e, of the feeling of plealongs to intu itio n (with the exceptio



Object Objectc nd edition. ndental Aesthetic" added in the seco From here to the end of the "Transce Objecte


Section H. On Time <B>

sure and displeasure and the which are not CO)g'l1l!tlCtilS tains nothing but mere relations,a of places in one intuitio n (extellsion), alteration of places (motion), and laws in accordance with which this alteration is determ ined (moving forces). But what is presen t in the or what it produc es in the things themselves besides alterati on place, is not given throug h these relations.

B 67

niUQiI \ginits elf iscognizecl;

to us tbroug h outer sense except in its rel)n~sentation f)t'lJy relation of an to subject, and not that to is the same the case of inner sense. It is not merely that the represe ntation s of outer sense make up material with which we occupy our mind, bULalsQ we place these representations, which itself precedes the consciousness of them in experience and ground s the way in we place them in mind as a formal condition, ah"~:ldy cont:alrtS n~l:.ltl(l11s simultaneity, and of that whicb is simultaneous with succession which persists). Now that which, as represe ntation , can precede any act of thinking someth ing is intuitio n and, if it contains nothing relations, it is the form of intuitio n, which, since it does not represe nt anything except insofar as someth ing is posited in mind, can be nothin g other than the way in which the mind is 3ffected by its own aCltlvlttv , namely this positin g of its represe ntation , thus the way it is through itseIf, i.e., it is an inner sense as far as regards its form. Everything that is represe nted throug h a sense is to that extent appear3nce, and an inner sense must therefo re either not be 3t all or else the subject, which is the object of this sense, can be represented by its means as appearance, not as it would judge of itself if its intuition were mere self-activity, i.e., intellectual. Anydif ficulty in this depends merely on the question how 3 subject can intern3lly intuit itself; yet this difficulty is commo n to every theory. Consciousness of itself (3pperception) is the simple represe nt3tion of the I, and if 3ll of the manifold in the subject were given self-ac tively throug h that alone, then the inner intuitio n would be intellectual. In hum3n beings this consciousness require s inner percep tion of the manifold that is 3ntecedently given in the subject, and the manne r in which this is given in the mind withou t spontan eity must be called sensibility on accoun t this difference. If the faculty for becomi ng conscious of oneself is to seek out (apprehend) that which lies in the mind, it must affect the latQt:t~fit:iQn, outer sense can alSQ
, Here Kant reverts to the use of Verhiilt1lis for the remainde r of the "Transce ndental Aesthetic," and it is thus this word that is translated by "relation " here and for the remainder of the section unless otherwise noted.




Doctrine of Elements. Part 1. The Transcendental Aesthetic <B> itself in such a way, whose an it can ter, it in the mind, determi nes s ground ently anteced form, howeve r, the luind in the represe nin r togethe is ld manifo the the way in which as it would immedi ately not itself intuits Lhen it there tation of time; B 69 the way in which it nce accorda in but itself, nt self-act ively represe conseq uently as it appears to itself, not as it is. is affected from outer objectsa time intuitio n represe nts Ill. If! say: in space i.e., as well as the self-int uition of the mind as each affects our senses, illumere a be would objects these that say to not is as it appear s, that d that sion. b,37 For in the appeara nce the objects ,' indeed even propert ies regarde d as someth ing really given, are we attribut e to s only on the kind of intuitio n of the depend ty only insofar as this proper object to it then this object as apgiven the of e subject in the relation itself as object! in itself. Thus I from uished disting be pearan ce is to g soul do not say that bodies merely seem to exist outside me or that my - in time and h space of quality the that assert if! only seems to be given of both posit I ce, existen their of on conditi as which, accorda nce themin objects! these in not and n intuitio of these -lies in my as selves. It would be my own fault if I made that which I should count to ng accordi happen not does this But * appeara nce into mere B 70



in rela*The predicates of appearance can be attributed to the object! in itself, illusion the but rose; the to e fragranc or calor red the e.g., tion to our sense, would can never be attributed to the object as predicate, precisely because that k for itself what pertains to it only in relation to be to attribute to the object origithe senses or in general to the subject, e.g., the two handles that were l in itobject the in ered encount be to not is \Vhat Saturn. to d nally attribute is and subject the self at all, but is always to be encountered in its relation to the thus and nce, appeara is object, inseparable from. the representation of the of the predicates of space and of time are rightly attributed to the objects attribute I if , contrary the On this. in illusion no is there and senses as such, outer the redness to the rose in itself, the handles to Saturn or extension to an obthese of relation nate determi a objects in themsel ves, without looking to arises first illusion then this, to nt judgme my jects to the subject and limiting

Objecte Scbein , Objecte this paragraph. d Bescha,ffc1J'heiten, here and in the remainde r of

, Relation f Object g scheincn h scheint



Objecten Objecte Object Objecte


Section II. On Time <B> our princip le of ideality of of our sensible if one ascribes objecti ve reality to those forms of then one cannot avoid thereby transfo rming everyth ing into mere illusion . For if one regards space and time as propert ies that, as far as their possibility is concerned, must be encoun tered in things in themselves, reflects on the absurdities in which one then becomes entangled, cause two infinite things that are neither substances nor anvtlun!l inhering in substances must nevertheless be someth ing existing, the necessary conditi on of the existence of all things, which also remain even if all existing things are removed; then one cannot well good Berkeley if he demote s bodies to mere illusion;3 8 indeed even our own existence, which would be made depend ent in a way on the self-subsisting reality of a non-en tity such as time, be transformed along with this into mere illusion; an absurdity no one has yet allowed himsel f to be guilty. Iv. In natural theology, where one conceives of an object is not only not an object of intuitio n for us but cannot even be an object of sensible intuitio n for itself, one is careful to remove conditions of time and space from all of its intuitio n (for all of its cogniti on must intuition and not thinking, which is always proof of limitations). with what right can one do this if one has anteced ently these into forms of things in themselves, and indeed ones as a priori conditions of the existence of things, would remain even if one removed the things themselves? - for as conditions of all existence in general they would also have to be conditi ons of the existence God. If one will not make them into objective forms of all things, no alternative remains but to make them into subjective forms of our of outer as well as inner intuitio n, which is called sensible because it is not original, i.e., one throug h which the existence of the object b of intuition is itself given (and tl1at, so far as we can have insight, can only pertain to the original being); ratl1er it is depend ent on the existence of the object," thus it is possible only insofar as the represe ntation al capacity of the subject is affected throug h that.3 9 It is also not necessary for us to limit the kind of intuitio n in space and time to the sensibility of human beings; it may well be that all finite thinking beings must necessarily agree with human beings in this regard (though we cannot decide this), yet even given such universal validity this kind of intuitio n would not cease to be sensibility, for the very reason that it is derived (intuitus derivativus),d not original (intuitius orig" hincip b O/dects Objects

B 71


derivative intuition


Doctrin e of Elemen ts. Part I. The Transce ndental Aesthetic <B>

intuitio n, which for the ground already inarius),a thus not to the original being, never to one that pertain adduce d seems to existence and its intuitio n (which deits regards as is depend ent b to given objects);C althoug h the last retermine s its existence in relation and mark must be counte d only as an illustra tion of our aesthetic theory not as a ground of its proof.

Conclu sion of the Transc endent al Aesthe tic. Here we now have one of the require d pieces for the solutio n of the transce ndental philoso phy - how are synthet ic a general problem priori propos itions possibl e? - namely pure a priori intuitions, space a and time, in which, if we want to go beyond the given concep t in an and priori judgme nt, we encoun ter that which is to be discovered a priori it, not in the concep t but in the intuition synthetically connec ted this ground such a judgme nt never exon but it; to onds that corresp d of tends beyond the objects of the senses and can hold only for objects possible experience.>

original intuition



The Transcendental Doctrine Second The Transcendental Logic

A5 olB ;

Introdu ction The Idea of a Transc endent al Logic

On logic in genera l. Our cognition arises from two fundamental sources fiEstgf which is the recepti on of represe ntation s (the re(:eptlVlty pressions), the second the faculty for cognizing an object these represe mation s (spontaneity of concepts); throug h ~bject is given to us, through latter it is though t in relation representation (as a mere determ ination of the mind). and c()l1c:epts therefo re constit ute the elemen ts of all our cognition, so that neither concepts withou t intuitio n corresp onding to them in some way nor intuition withou t concepts cana yield a cogniti on. ~_'..~~ ~. pnre or empirical. E.l;ilpiricaI, if sensation (which presupposes actual presence of the object) is contain ed therein ; pure if no sensation is mixed into the represe ntation . One can call latter the matter of sensible cognition. Thus pure intuitio n contains form nuder which someth ing is intuited , and pure concep t thinking of an object in general. Only pure intuitio ns or concepts are possible tiphori , empirical ones only a posteriori. If we will call the recepti vity of our mind to receive represe ntation s insofar as it is affected in some way sensibi lity, then on the contrar y faculty for bringin g ford1 represe ntation s itself, or the sponta neity of cognition, is the unders tandin g. It comes along our nature that intuition can never be other than sensibl e, i.e., it contains only the way in which we are affected by objects. The faculty for thinkill lg of objects of sensible intuitio n, on the contrary, is the unders tandin g. qNeither of these propert ies is to be to the other. sen~ibi1ity no objeCt would be given to us, unders tanding none would be though t. Thpug hts withou t conten t are empty, intuitio ns .. _.
---~ ~""~ .. ,.-, ..



" The second edition has the plural verb


the first had the singular kann.


Doctrine of Elements. Pt. H. Transcendental Logic It is thus just as necessary to make an object to them in intuition) as mind's concepts sensible (i.e., to them under (i.e., to andable underst it is to make its intuitio ns exchange~ cannot ies capacit or s facultie two concepts). Further , these g, anythin g intuitin of capable not is tanding their functions. The unders their from Only g. anythin g thinkin of and the senses are not capable mix unifica tion can cogniti on arise. But on this accoun t one must not B 76 carefully them e separat to cause up their roles, rather one has great A 52 from each other and distinguish them. Hence we distinguish the science the rules of sensibility in general, i.e., aesthetic, from the science of the rules of unders tanding in general, i.e., logic. Now logic in turn can be underta ken with two different aims, either as of the understanding. The r theJogi c Qf~tg thinking, withou t which of former contain s re concerns these therefo it and place, no use of the underst anding takes which it may be to objects the of ce rules withou t regard to the differen anding contains 2 underst the of use lar directed. The logic of the particu The forobjects. of kind certain a the rules for correctly thinkin g about of organon the r, howeve latter, the mer can be called elemen tary logic, scithe before stuck often is latter this or that science. In the schools the ences as their propaedeutic, though in the course of human reason they are certainly the latest to be reached, once the science is already long complete, and requires only the final touch for its improv ement and perwill fection. For one must already know the objects rather well if one about. t brough be to is them of offer the rules for how a science B 77 ;Now general logic is either pure or applied logic. In the former we anding A53 abstrac t from all empirical conditi ons under which our underst play of the from senses, the of ce is exercised, e.g., from the influen etc., ion, inclinat habit, of power the y, imagination,a the laws of memor all from general in indeed ce, prejudi hence also from the sources of arise, to ed suppos be may or arise ons causes from which certain cogniti because these merely concern the unders tanding under certain circumstances of its application, and experience is require d in order to know a prido with these. A genera l but pure logic

~~;1~~~~~1F~~~~~i~1~~may and is a ~~~I~ ori principles,b to what is j

only in regard (empirical or transcendental). A genera l logic, however, is then called applied if it is directe d to the rules of the use of the understanding It under the subjective empirical conditi ons that psychology teaches us. ingeneral sure be to is it h althoug es," therefo re has empirical principl


Introduction sofar as it concern s the use of the unders tanding regard to difference of objects. On this accoun t it is also neither a canon the understandi_~gX~Lgeneralnor an organo n of sciences, but B 78 merely a@~ili~IfisJof the commo n UIlderstanding. In general logic the part that is to constit ute the pure doctrin e of reason must therefo re be entirely separat ed from that constitutes applied (though stillgeneral) logic. The former alone is SClA54 ence,althmigh- brIef diy~-as The scholastically correct pnosentaltic,n of a doctrine of the elements of the unders tanding this therefore logicians must always have two rules in view. I) As general logic it abstracts from all content s of cogmtlOn the underst anding and of the difference of its objects, and has to do nothing but the mere form of thinking. 2) As pure logic it has no empirical principles,a thus it draws nothIng from psychology (as one has occasionally been persuaded), fore has no influence at all on the canon of the understanding. It is a proven doctrin e, and everyth ing in it must be comple tely a priori. What I call applied logic (in opposit ion to the commo n signification of this word, according to which it ought to contain certain exercises to which pure logic gives the rule) is thus a represe ntation of the standing and the rules of its necessary use in concreto, namely the contingent conditions of the subject, which can hinder or promot e this B 79 use, and which can all be given only empirically. It deals with attentio n, its hindrance and consequences, the cause of error, the condition of doubt, of reservation, of conviction, etc., and general and pure logic is related to it as pure morality, which contains merely the necessary ASS laws of a free will in general, is related to the doctrin e of virtue which assesses these laws under the hindrances of the feelings, m(~llrla lions, and passions to which human beings are more oi-Iess subject, which can never yield a true and proven science, since it requires ical and psychological principles b just as much as that applied logic does.


On transce ndenta l logic. General logic abstracts, as we have shown, from all conten t of cognition, i.e. from any relation of it to the object,d and considers the

The contrast between this term and the following use of Verhaltnis 196, note a) that Kant continue s to use the former to connote a relation between subject and object and the latter among objects, though in this case objects of thought rather than sensibility. Further, unnoted instances of "relation " translate Heziehl''77;:;.



Doc trine of Elem ents . Pt.

n. Transcendental Log ic

form cog niti ons to one ano ther , i.e., the iriemp as as e pur are ther e of thin king in general. bca ed), prov c heti aest cal intu itio ns (as the tran scen den tal found. of objects twe en pur e and empirical thin king B 80 not abst ract from be a logic in whi ch one this case ther e mer ely the rules ed tain cop c that con tent of cog niti on; for that logi ld exclude all thos e cog niti ons that of the pur e thin king of an obje ct wou of our ther efor e con cern the wer e of empirical con tent . It cts; obje to not can cts insofar as A 56(;()g~i~tions of obje in y, has noth ing to do with this orig whi le general logic, on the contrar are of cognition, but rath er considers empirically, mer ely in res or originally given a priori in ourselve erst and ing brin gs them und ch the spe ct of the laws acco rdin g to whi ther efor e it deals only and ks, b it thin into re1ation to one ano ther given to the reprebe can whi ch with the form of the und erst and ing, ed. e orig inat sent atio ns whe reve r they may hav ort of which extends to all of the imp the ark rem a e And here I mak we mus t keep well in view, namely following considerations, and that t be called tran scen den tal, but only that not every a priori cog niti on mus ns that and how cert ain representatio that by means ofw hich we cognize (i.e., ible poss are or ri, enti rely a prio (int uitio ns or con cep ts) are applied a priori). Hen ce neit her space nor use its or on niti cog the possibility of B 81 eit a priori is a tran scen den tal repr any geo met rica l dete rmi nati on of of not are ns tatio esen repr e that thes sent atio n, but only the cog niti on be sibility that they cane nevertheless empirical orig in at all and the pos tal. den scen tran ed call be erience can rela ted a priori to objects of exp be all objects in gen eral wou ld also ut abo e spac of Likewise the use then es, sens the of cts obje to solely tran scen den tal; but if it is rest rict ed betw een the tran scen den tal and ce eren diff The L irica it is called emp A 57 to the crit ique of cognitions and the empirical ther efor e belongs only thei r object. does not con cern thei r rela tion to ther e can perh aps be concepts that In the expectation, ther efor e, that t not as pur e or sensible intuigQnsbu may be related to objects a priori, , that are thus con cep ts but of nei~ rath er mer ely as acts of pur e thin king we provisionally form ulat e the idea ther empirical nor aest heti c origin, and of the pur e cog niti on of reason, of a science of pur e und erst and ing com plet ely a priori. Suc h a science, by mea ns of whi ch we thin k objects ~ the dom ain, and the objective valid whi ch wou ld dete rmi ne the origin, , e to be called tran sce nde ntal logic ity of such cognitions, wou ld hav n, laws of the und erst and ing and reaso since it has to do mer ely with the


Verhiiltnisse Verhiiltnis cn instead of kiinnc. Following Erdm ann, reading kOn!7


Introduction but solely insofar as they are related to objects not, as in case of general logic, to empirical as as pure cognitions reason without distinction.
13 82

On divisio n of genera l logic into analyti c and dialecti c. The old and famous question with which the logicians were to be driven into a corner and brough t to such a pass that they must either faU into a miserable cirdea or else confess their ignoran ce, hence vanity of their entire art, is this: What is tmth? The of truth, namely that it is the agreem ent of cogniti on with its object, is here granted and presupposed; but one demands to know is the general and certain criterio n of the truth of any cognition. It is already a great and necessary proof of cleverness or insight to know what one should reasonably ask. For if the questio n is in itself and demands unnecessary answers, then, besides the embarrassment of the one who proposes it, it also has the disadvantage of misleading the incautious listener into absurd answers, and present ing ridiculous sight (as the ancients said) of one person milking a billy-go at while the other holds a sie~e undern eath) If truth consists in the agreem ent of a cogniti on with its object, then this object must thereby be distinguished from others; for a COls11lu on is false if it does not agree with the object to which it is related even if it contains someth ing that could well be valid of other objects. Now a general criterio n of truth would be that which was valid of all cognitions without any distinct ion among their objects. But it is dear that since with such a criterio n one abstracts from all conten t of cognition tion to its object), b yet truth concern s precisely this content , it completely impossible and absurd to ask for a mark of the this content of cogniti on, and thus it is dear that a sufficient and yet at same time general sign of truth cannot possibly be provided. Since above we have called the conten t of a cogniti on its matter, one must therefore say that no general sign of the truth of the matter of cognition can be demand ed, because it is self-contradictory. But concern ing the mere form of cogniti on (setting aside all content), it is equally dear that a logic, so far as it expounds the general and necessary rules of underst anding , must present criteria of truth in these very rules. For that which contrad icts these is false, since the understanding thereby contradicts its general rules of thinkin g and thus con" In the second edition, Dialexis; in the first, Dialele, i.e. reasoning in a circle.


13 83


13 84



Doctrine of Elements. Pt. n. Transcendental Logic i.e., of tradicts itself. But these criteria concer n only the form of sufnot but correct thinkin g in general, and are to that extent entirely logiwith accord te comple in ficient. For althoug h a cogniti on may be the cal form, i.e., not contrad ict itself, yet it can still always contrad ict of ent agreem the namely object. The merely logical criterio n of reaand tanding unders of a cogniti on with the general and formal laws thus the negative son, is therefo re certainly the conditio sine qua non and the error go, cannot conditi on of all truth; further, however, logic A60 any touchby red discove be that concern s not form but conten t cannot 4 stone of logic. andGenera l logic analyzes the entire formal business of the underst a all of les princip as these s ing and reason into its elements, and present b of our cogniti on. This part of logic;.C::.llJth~J:eforebe logical assessment called an analytic, and is on that very accoun t. at lc;3:.sLtlle negative else examin.e a-!1.cfeva1li:touchst one of truth, since one must before by means ufthes e rules the form of all cogniti on before investigatits conten t in order to find out whethe r with regard to the object it B 85 r .cC)lll:anls positive truth. But since the mere form of cogniti on, howeve the ute constit to g well it may agree with logical laws, is far from sufficin of materia l (objective) truth of the cognition, nobody can dare to judge without logic with objects and to assert anythin g about them merely having drawn on anteced ently well-founded inform ation about them use from outside oflogic , in order subsequently merely to investigate its betor, laws, logical to ng and connec tion in a cohere nt whole accordi ter, solely to examine them according to such laws. Nevert heless there is someth ing so seductive in the possession of an appare nt art for giving reall of our cognitions the form of underst anding , even though with genthis that poor, and gard to their conten t one may yet be very empty A61 if it erallog ic, which is merely a canon for judging,c has been used as of nce sembla the least at of were an organo n for the actual produc tion Now d. !J1isuse been objective assertions, and thus in fact it has thereby general logic, as a putative organo n, is called dialect ic. As different as the significance of the employ ment of this designation of a science or art among the ancients may have been, one can still infer other than B86 from thcir actual use of it that among them it was nothing ce, inignoran its to giving the logic of illusio n - a sophistical art for g the imitatin by truth, of deed even to its intentio nal tricks, the air using and , general in es method of thoroug hness, which logic prescrib one Now ion. pretens its topics for the embell ishmen t of every empty considlogic, general can take it as a certain and useful warnin g that


Introduction ered as an organo n, is always a logic of illusion, i.e., is dI~l1e(:ti(:al. since it teaches us nothing at about conten t only the formal conditions of agreem ent with the understandiIlg, are entirely indiffe rent with regard to the objects, the effroJntt~ry using it as a tool (organon) for an expansion and extension of its mation,a or at least the pretens ion of so doing, comes to nothIng but idle chatter, asserting or impeac hing whatever one wants plausibility. Such instruc tion by no means befits the dignity of For this reason it would be better to take this designation of" dialectic" as a critiqu e of dialect ical illusion , which is counted as logic, in such a way we would here have it be underst ood.


On the division of transce ndenta l logic into the transce ndenta l analyti c and dialecti c. In a transcendental logic we isolate the unders tanding (as we with sensibility in the transcendental aesthetic), and elevate from our cognition merely the part of our though t that has its origin in understanding. The use of this pure cognition, however, on this as its condition: that objects are given to us in to which it can be applied. For withou t intuitio n all of our would lack objects/ and therefo re remain comple tely empty. of transcenc-dental1ogic, therefo re, that-ex pounds the of the cQgJ;lition themld erstand ing and the no Qh~e_CLcan beJ:ho ughraL all, is the. transce ndentaL analyti c, and at the same time a logic of truth. For no cogniti on can contrad ict it at the same time losing all content , i.e., all relation to ariJobj ect,d hence all truth. But because it is very enticin g and seductive to make use these pure cognitions of the unders tanding and principles selves, and even beyond all bounds of experience, which however itself alone can give us the matter (objects)" to which those pure of the underst anding can be applied, the unders tanding falls into the ger of making a material use of the merely formal princip les! of pure understanding throug h empty sophistries, and of judging distinction about objects that are uot given to us, which perhap s



Objecten ,. Pri77ripic77 d Object , Objecte f Pri77ripicn



Doctrine of Elements. Pt. H. Transcendental Logic properl y be only a could not be given to us in any way. Since it it canon for the assessment of empirical use, it is misused if one lets to dares count as the organo n of a general and unrestr icted use, and the synthetically judge, assert, and decide about objects in general with in would pure unders tanding alone. The use of the pure unders tanding ndental this case therefo re be dialectical. The second part of the transce logic must therefo re be a critique of this dialectical illusion, and is called iltransce ndental dialectic, not as an art of dogmatically arousin g such works ld lusion (an unfortu nately highly prevale nt art among the manifo of metaphysical jugglery), but rather as a critique of the understanding the and reason in regard to their hyperphysical use, in order to uncove r claims false illusion of their groundless pretens ions and to reduce their A 64 to inventi on and amplification, putatively to be attained throug h transcende ntal principles, to the mere assessment and evaluation of the pure underst anding , guardin g it against sophistical tricks.


Transcendental Logic First Division

The Transcendental Analytic

This Analytic is the analysis" of the entirety of our a priori cOI~mtl On iIltothe elemen ts oft:b~j)Jlre the understandiI1g. It cerned with the following empiricalconcepts. "2; That tl:HwbeJleulg':no1tt(}illruliti,on aUla to bility..-butmther to 'thiI1lci11K;lI1d uI1clcrstanding'3) mentary concep ts,andc learIy .disting uished. from derived or compo sedfrol ll theIll.'Lj:': That the table t:h,el11!tJ(~ c()mpl(~ te, and that they entirely exhaus t the entire field of Now this comple teness of a science cannot reJia[}ly rough calculation of an aggregate put togethe r hence it is possible only by means of an idea of the of the a ori cognition of the underst anding , and through C the division of concepts that such an idea determ ines and that constitu tes it, thus through their( )unect ionin a system . The pure unders tanding separates itself comple tely not only from everyth ing even from all sensibility. It is therefo re a unity that subsists on its own, is sufficient by itself, and which is not to be supplem ented any external additions. Hence the sum total of its cogniti on constit ute a system that is to be grasped and determ ined under one idea, completeness and articula tion of which system can at the same time yield a touchstone of the correct ness and genuine ness of the pieces of cognition fitting into it. This whole part of the transcendel1tal1ogic, however, consists of two books, the first of which contains the concep ts of pure understanding, the second its princip les.
'The "1." is missing in the second edition. , Added in the second edition.




Transcendental An aly tic Book The An aly tic Concepts. a


l cepts not thei r analysis, or the usua I und erst and by an analytic of con ent cont the ations, that of analyzing.b pro ced ure of philosophical investig and brin ging them to distinctness, es of concepts that pres ent themselv c atte mpt ed anal.ysis of the faculty but rath er the muc h less frequently i to research the possibility of a prior of und erst and ing itself, in ord er e plac birth the und erst and ing as thei r concepts by seeking them only in ; for this is the pro per business of and analyzing its pur e use in general rest is the logical trea tme nt of cona transcendental philosophy; the

on: poin t in Kant's copy of the first editi The following note s appe ar at this etic propositions, and how synth of ists cons e rienc expe that "\\le rema rked above as a question reare possible is not to be rega rded synth etic a posteriori prop ositi ons quiri ng a solution, since it is a fact. is possible. "No w it is to be asked how this fact empirical but it is to be asked whet her these s, ment judg of "Exp erien ce consists The analysis ts. men judg ) (pure i prior a e ppos judg men ts do not in the end presu insofar as judgments sis of experience contains, first, its analy epts. conc i prior a also epts i01-i conc are in it; second, beyo nd the a postcr ding do 1. Wha t does the understan ble? possi e rienc expe is "The prob lem is: How 3. In ems? ment judg rical empi in do s t do the sense in judg men ts in general? 2. "\Tha tions of the rstan ding , applied to the representa pirical cognition, wha t does the unde cte]? [Obje ts objec of ition a cogn senses, do in orde r to brin g forth a p1-iori propois only possible thro ugh synth ctic "On e sees at first that experience in accordance with use; t: anen imm 1. are sitions. Hen ce a priori principles ther they are also trans cend ent. 2. it is to be asked, whe a fact, is as it were experisome thing is also eJ'1lerience, i.e., ther whe for test "The empirical judgositi ons unde r whic h the parti cular men tatio n with the universal prop ing, if no concept judg for rule ersal univ a r unde stand men t belongs. If the latte r cann ot [vicious fallaCJ!J. Why in superstia vitium can be mad e out of that, then it is 4-5) 23:2 ; 21-2 tion and credulity." (E XXXIIL pp.


Div. 1. Transce ndental '\nalytic

cepts in philoso phy in general. We therefo re pursue cepts into their first seeds and predispositions in the standing, where they lie ready, Imtil with the opportu:nl1:y they are finally developed and exhibited in their the understanding, liberated from empirical conditions them.



The the Pure

Concepts First cIJav:ter Discovery []nderstanding

If one sets a faculty of cogniti on into play, then on various occasions difwill make this faculty become promin ent ferent concepts exhaustive treatise deless or more a in d known and that can be collecte a longer time or with for d observe been pendin g on whethe r t..1}ey have comple ted can never be will ation investig greater acuteness. Where this it were mechanical as this of means be determ ined with certain ty only as the opporred discove are that ts proced ure. Further , the concep unity, but will atic system and order A67 tunity arises will not reveal any and placed in ties similari to ng accordi rather be ordered in pairs only B92 from the simt, conten their of ude magnit the series only in accord with systematic means no by are series which ite, ple to the more compos ed. produc ically even if to some extent method Transc endenta l philoso phy has the advantage but also the obligation to seek its concep ts in accordance with a principle,a since they spring pure and unmixed from the underst anding , as absolute unity, and must therefo re be connec ted among themselves in accordance with a concept or idea. Such a connec tion, however, provides a rule by means of which ethe place of each pure concep t of the unders tanding and the complet would which p7-iori, a ined determ be can r ness of all of them togethe otherwise depend upon whim or chance.
On the Transc endent al Clue for the Discov ery of all Pure Concep ts of the Unders tandin g First Section On the logical use of the unders tandin g in generaL The unders tanding has been explained above only negatively, as a nonsensible faculty of cognition. Now we cannot partake of intuition inde-



Section I. On the logical use of the understanding in general pendently of sensibility. The unders tanding is therefo re not a intuition. besides intuitio n there is no other of cogniti on than through concepts. Thus at least UllderuncleIstanC! the unity of 93

under a commo nas one. Concep ts ,:'>.eJ1Si,Q le Il1JpreS~;lons. Now the understajtldilllg can make no other use of concepts than by represe ntation pertain s to Immediately except intuitio n alone, a,.~:Qniceptls .WLlU"H"V'~l reIated toanob ject, is always related to some relm~seI1tation of it (whether that be an intuitio n or itself already a cOIlcept). Jwtgln ellt i$ thetefore the mediat e cogIlition object, hence the represe ntation of a represe ntation of it. In every judgme nt there is a concep t holds of many, and that among this many also compre hends a given representation, which is then related immed iately to the object. 5 So in the judgment, e.g., "All bodies are divisible,"c the concep t of the divisible is related to various other concepts; among these, however, it is here particularly related to the concep t of body, and in turn is related to certain appearances d that come before us. These objccts are A69 therefore mediately represe nted by the concep t of divisibility. All judgB 94 metlts are accordingly functions of unity among our representation s, since instead of an immed iate represe ntation a higher one, comprehends this and other represe ntation s under itself, is used cognition of the object, and many possible cognitions are thereby drawn together into one. We can, however, trace all actions of tlle understanding back to judgme nts, so t~at the unders tandin g in general can be represented as a faculty for judgin g. For according to what has been said above it is a faculty for thinking. Ihinki~Kis cogniti on thIough cQn<::eJlts. Concep ts, however, as predicates of possible judgments, related to some represe ntation of a still undete rmined The concept of body thus signifies someth ing, e.g., metal, can be cognized through that concept. It is therefo re a concep t because other representations are contain ed under it by means of it can be reXXIV, p. 23; 23'45) b Kant's copy of the first edition replaces this parenthetical aside with the following words, without parentheses: "which itself contains intuition only mediatel y or immedi(E xxxv, p. 23; 23:45) rather than veriindcrlicb, following the fourth edition. d Kmt's copy of the first edition changes "appeara nces" to "intuitions" (E XXXVI, p. 23; 23'45)
" In his copy of the first edition, Kant inserts here the word "other"




Doctrine of Elements. Pt.

n. Div. 1. Bk 1. Ch. I

lated to objects. It is therefore the predicate for a possible judgmen~ e.g., "Every metal is a body." The functions of the understanding can therefore all be found together if one can exhaustively exhibit the fimctions of unity in judgments. The following section will make it evident that this can readily be accomplished. Clue to the Discovery of Pure Concepts of the Understanding Second Section
< 9.> a logical function the understanding in judgments. If we abstract from all content of a judgment in general, and attend only to the mere form of the understanding in it, we find that the function of thinking in that can be brought under four titles, each of which contains under itself three moments. They can suitably be represented in the following table. 6

Quantity ofJudgments
Universal Particular Singular

Affirmative Negative Infinite

3 Relation' Categorical Hypothetical Disjunctive

4 Modality
Problematic Assertoric Apodictic B96
A 7I

Since this division seems to depart in several points, although not essential ones, from the customary technique of the logicians, the following protests against a worrisome misunderstanding are not unnecessary.
Here Kant resumes the numbering of paragraphs beg>ID in the "Transcendental Aesthetic" in the second edition. This will continue through the end of the "Transcendental Deduction." Here Kant uses the latinate word Relation instead of either or Verbiiltnis.


Section H. On the logical function in judgments

I. The logicians rightly say in use of judgme nts in syllog:is1l1S singular judgme nts can treated like universal ones. For because they have no domain at their predica te is not related to some of what is contain ed under the concep t of the while eluded from anothe r part of it. The predicate therefo re concept withou t exception, just as if the latter were a F;CUCLdI IY concept with a domain with the predica te applying to is signified. b on the contrary, we compar e a singular jlH1gJille:nt generally valid one, merely as cognition, with respect to quan1tit)T,c the former d relates to the latter as relates to and is therefore in itself essentially different from latter. Theref ore, a singular judgme nt (judicium singulare) not only respect to its internal validity, but also, as cogniti on in general, with respect to tlle quantitye it has in compar ison with other cognitions, then it is different from generally valid judgme nts and serves a special place in a comple te table of the momen ts of thlnkJillg general (though obviously not in that logic that is limited use of judgments with respect to each other). 2. Likewise, in a transce ndental logic infinite judgru ents must be distinguished from affinna tive ones, even though in general logic they are rightly include d with the latter and do not constit ute a special member of th~cbssification.Generallogic abstracts from all conten t of the predicate (eV:~n ifiti~ negati~~), and considers only whethe r it is attributed to the subject or opposed to it. Transc endenta l logic, howeve r, also considers the value or conten t of the logical made in a judgment by means of a lllerely llega~ve predicate, and sort of gain this yields for the whok-o{~~gnition~1flh~d said of the soul that it is not mortal, then I would at least have avoided an error means of anegative judgme nt. Now by means of the propos ition "'rhe soul is not mortal" I have certainly made an actual affirmation as far as logical is concerned, for I have placed the soul within the unlimit ed dom~Lm undying beings. Now since that which is mortal contains one part whole domain of possible beings, but that which is undyin g! the other,

B 97

A 72

, CrOfle

gemci71A:iiltlgCJ: lNhile this would normally be translate d "commo nly valid," in this context it refers to the universal (a!!gc711cin) judgmen t; we have used "generall y" to preserve this reference while still marking the difference from fIllCT711em . b VOll dcsscn ganzer here Kant uses Bedeutung, as Frege was later to use it, to mean the reference or denotatio n of a concept; more typically, he uses it to mean something closer to what Frege called Sinn or sense, that is, the connotat ion.

, CrOfle

The text has sie rather than es, but in spite of the shift in gender there is nothing for the pronoun to refer to except "a singular judgmen t."

f In the second edition, Nichtstcrbendc; in the first, Nichtsterb!iche, or "immorta l."


Doctrin e of Elemen ts. Pt. II. Div. 1. Bk I. Ch. I

the soul is one the infinite my propos ition nothing is said everyth ing that is mortal. away take multitu de things that remain to the exlimited thereby is e possibl But the infinite sphere of the is placed soul the and it, from ed separat is mortal tent that that which is a on this excepti this with even But . domain its the remain ing space of B 98 it from away taken be could parts more , still remains infinite affirmabeing and least the in g growin the concep t of the soul A 73 tively determ ined. In regard to logical domain, therefore, this infinite in judgme nt is merely limitin g with regard to the conten t of cognition transcen the from omitted be not must it general, and to this extent function the since nts, judgme in g thinkin of ts momen dental table in of unders tanding that is hereby exercised may perhap s be import ant 7 on. cogniti the field of its pure a prim-i 3. All relations b of thinkin g in judgme nts are those a) of the predicate the to the subject, b) of the ground to the consequence, and c) between . division the of rs membe the of all and" cogniti on that is to be divided in be to red conside are ts concep two In the first kind of judgme nt only relation to each other, in the second, two judgments, and in the third, jusseveral judgments. The hypothetical propos ition "If there is perfect of relation the s contain really d" tice, then obstinate evil will be punishe punis evil ate "Obstin and justice" two propositions, "There is a perfect reished." Vi7hether both of these propositions in themselves are true means by t though is that tion implica the mains unsettl ed here. It is only of this judgnlent. Finally, the disjunctive judgme nt contains the relations B 99 of of two or more propos itions to one another , though not the relation of sphere the as insofar ion, opposit sequence, but rather that of logical one judgme nt excludes that of the other, yet at the same time the relation of community, insofar as the judgme nts togethe r exhaust the sphere of cogniti on proper; it is therefo re a relation of the parts of the sphere A 74 of a cogniti on where the sphere of each part is the comple ment of that of the others in the sum total of the divided cognition, e.g., "The world or exists either throug h blind chance, or throug h inner necessity, one s occupie itions propos these of throug h an external cause." Each a part of the sphere of the possible cogniti on about the existence of remove To sphere. entire the occupy they r world in general, and togethe the the cognition from one of these spheres means to place it in one of

ibres Following the first edition, Rnum ihres Umfangs, rather than the second, Raums. the functions of judgmen t the table had b Verhaltnisse; although he is now speaking of , and in the relisted under the latinate heading Relation, Kant now reverts to Verhaltnis reversion to Kant's ." "relation by d translate is , Verhaltn; h paragrap this of mainder of the talking is he since e, elsewher term this of Verhiiltn;, here is consistent with his use us. to than rather other each to ts judgmen of parts of relation , p. 23; 23:45). , Kant's copy of the first edition replaces "and" with "of" (E XXXVII


Section n. On the logical function in judgments others, and to place it in one sphere, on the contrary, means to remove it from others. In a disjunctive judgme nt there is therefo re a certain community of cognitions, consisting in the fact that they exclude each other, yet thereby determ ine the true cogniti on in its entirety, since taken togethe r they constit ute the entire conten t a particular given cognition. 8 And this is also all I it necessary to remark upon for the sake of follows. a 4. The modali ty of judgme nts is a quite special function of which is distinctive in that it contrib utes nothing to the conten t of the judgment (for besides quantity, quality, and relation b there is nothIng more that constitutes the conten t of a judgment), concerns only the value of the copula in relation to thinkin g in genera1 9 . Problem atic judgme nts are those in which one regards the assertion or denial as merely possib le (arbitrary). Assert oric judgme nts are those in which it is considered actual (true). Apodic tic judgme nts are those in which it is seen as necessa ry. * Thus the two judgme nts whose reJatlon constitutes the hypothetical judgme nt (antecedens and consequens), as as those in whose reciprocal relatione the disjunctive judgme nt consists (the members of the division), are all merely problematic. In the above example the propos ition "There is a perfect justice" is not said asserto rically, but is only though t of as an arbitrar y judgme nt that it is possible that someone might assume, and only the implication is assertoric. Thus such judgments can be obviously false and yet, if taken pnobJlenlaticaHy , conditions of the cogniti on of truth. Thus the judgme nt "The exists throug h blind chance " is of only problem atic significance in the disjunctive judgme nt, that is, someon e might momen tarily assume this proposition, and yet it serves (like the designation of the false among the numbe r of all of those one can take) to find the true one. The problematic propos ition is therefo re that which only expresses logical possibility (which is not objective), i.e., a free choice to such a proposition to count as valid, a merely arbitrar y assumption of it in the understanding. The assertoric propos ition speaks of logical or truth, as say in a hypothetical syllogism the anteced ent in the premise is problematic, but that in the minor premise assertoric, and in* It is just as if in the first case thought were a function of the understanding, in the second of the power of judgment, and in the third of reason.







aremark the elucidation of which can be expected only in the sequel.

This is

, The following note occurs in Kant's copy of the first edition: "Judgme nts and propositions are different. That the latter are verbis cxprma [explicit words], since they are asscrtoric" (E XXXVIII, p. 23; 23: 2 5).
b f~Th:jftniJ


Doctrin e of Elemen ts.

Pt. H. Div. 1. Bk 1. Ch. I

acdicates that the propos ition is already bound to the underst anding one ic cording to its laws; the apodictic propos ition thinks of the assertor as as determ ined throug h these laws of the unders tanding itself, and Now y. necessit thus asserting a priori, and in this way expresses logical since everything here is gradually incorpo rated into the understanding, asso that one first judges someth ing problematically; then assumes it ed connect asserts it to be inseparably sertorically as true, and these ic, with the underst anding , i.e., asserts it as necessary and apodict of three functions of modali ty can also be called so many moments thinkin g in general.
B 102


Clue to the Discov ery of all Pure Conce pts of the Unders tandin g Section
< IO.>

On the pure concep ts of the unders tandin g or categor ies. Ailias-alr.eadybeen frequen tly said, genera llogic abstractsJwlll-all con~g~ell to it '-tenLo Lcogni tion, and expects that represe ntation s xm .!hem tral1sfQ to it for fromel sewber e, wherev er this may Y, has a .co1J1L3l the ,Qll allogic .LnJg.<::o!lc.:ePts'111alytically. Transc endent enJ:ransc hichthe priori,w a it before s manifold of sensibility thatJie pts J::Qnce thepure provide to order in it, has offered to A 77 withQe RoulQ__ ilie y wl:lich itJloult. otJ:heJID.QerstandingwitJ:l_a.Jnatiter,w a gut any contelJ J,thnsc 01llple telyem pty. Now space and time contain the among eless neverth belong but n, manifo ld of pure a priori intuitio reconditi ons of the receptivity of our mind, under which alone it can affect also always must they thus and , ceive represe ntation s of objects rethe concep t of these objects. Only the spontan eity of our though t d mmbine IIp,ancL h,mken thlollg quires that this manifold first beggne this call ofit.! out e bemad ionto for a cogllit ina certain way -action synthesis. a B.J72xnthesis i!UheJ llQst generalsens-,however, Lunder stand the B I03 ch other and .a~~()fp~~gclif"fel:t::!l.t represe ntation s togethe r withe:l synthesis is Such on. cogniti one in .C::~~Eehending theirlllanifolciness s"I[tha t in priofi(a a but iricallY p~t::~ if theJ:na nifoldi sgiyelL llotemp these must s ntation represe our of analysis sp.acea nd time). Prior to all the conas far as ally analytic arise can ts first be given, and no concep

ts come to be, stand by synthesi s, however, the action through which synthetic judgmen the words adds also Kant 23:452. 23; p. in the general sense, . . ." (E XXXIX, 24)' p. XL, (E nexus" and ion, composit ation, "Combin

this point to "I underIn his copy of the first edition, Rant cltlnges this sentence to


Section HI. On the pure concepts of the understanding a tent is concerned. The synthesis of a manifold, however, it be given empirically or a priori) first brings forth a cognition, to be sure may initially still be raw and confused, and thus in need of analysis; yet the synthesis alone is that which properly collects the elements for cognitions and unifies them into a certain content; it is therefore the first thing to which we have to attend if we to the first origin of our cognition. mere effect Synthesis in general is, as we shall subsequently see, of the imagination, of a blind though indispensable function of the soul,h wiUlOut which we would have no cognition at all, but of which we are seldom even conscious. Yet to bring this synthesis to concepts is a function that pertains to the understanding, and by means it first provides cognition in the proper sense Now pure synthesis, generally represented, yields the pure concept of the understanding. By this synthesis, however, I understand which rests on a ground of synthetic unity a priori; thus our counting (as is especially noticeable in the case of larger numbers) is a synthesis in accordance with concepts, since it takes place in accordance with a common ground of unity (e.g., the decad). Under this concept, therefore, the synthesis of the manifold becomes necessary. Different representations are brought under one concept cally (a business treated by general logic). Transcendental logic, ever, teaches how to bring under concepts not the representations but the pure synthesis of representations. The first thing that must be given to us a priori for the cognition of all objects is the manifold of pure intuition; the synthesis of this manifold by means of the imagination is the second thing, but it still does not yield cognition. The concepts that give this pure synthesis unity, and that consist solely in the representation of this necessary synthetic unity, are third thing necessary for cognition of an object tl1at comes before us, and they depend on the understanding. The-samelunction_thaLgiYesunity to in .~j~~gJIl~~taI2()gi~e~_unit:y totl1eJllere sYlltQesis of4ifferent repres.~ntati()nsina11,j1!tuition, which, expressed generally, iScCalled_theiJure co'lli.;~PtQf~Pc~~rsti1nding. I I The same understanding, therefore, and indeed by means of the very same actions through which it brings the logical form of a judgment into concepts by means of the analytical unity, also brings a transcendental content into its representations by means of
,>n,>ni'Tl_ IQ





" In the first edition, the right-hand running head is "Section Ill. On the pure concepts of understanding or categories" b In his copy of the first edition Kant replaces this clause with "of a function of the understanding" (E XLI, p. 24; 23:45) , in Bedeutung


Doctrin e of Elemen ts. Pt. H. Div. 1. Bk 1. Ch. I


t of the synthetic unity of the manifold in intuitio n in general, on accoun to pertain that tanding unders the of ts concep which they are called pure logic. a general by lished accomp be never can objects a priori; this In such a way there arise exactly as many pure concep ts of the unas derstanding, which apply to objects of intuitio n in general a priori, previous the in nts judgme e possibl of ns functio there were logica~ b table: for the unders tanding is comple tely exhausted and its capacity call will we le Aristot ing Follow s.c function entirely measur ed by these althese concepts catego ries, for our aim is basically identical with his d on. executi in it from though very distant
Table of Categor ies"


Of Quanti ty Unity Pluralit y Totality


Of Quality Reality Negatio n Limitat ion

Of Relation e Of Inheren ce and Subsistence

(s'llbstlmtia et accidens)
Of Causalit y and Dependence (cause and effect) Of Commu nity (reciprocity between agent and patient)

Of Modali ty Possibil ity - Impossi bility Existenc e - Non-ex istence Necessi ty - Conting ency


T7er7J1 i;(fen

, gedachte Functionen

copy of the The fol1owing notes precede the ensuing table of the categories in Kant's first edition: Categor"Logical functions are only forms for the relation of concepts in thinking. the to regard in ed determin are s intuition certain ies are concepts, through which e.g., what must thetic unity of their consciousness as contained under these functions; be thought as subject and not as predicate." (E XLII, p. 24; 23:25) "On the use of the categories in the division of a system. "On the analytic of the categories and the predicables. represen"On a characteristic of concept'; of intellectual, empirical, and pure sensible

tations. 23:25) " - Lex origi17aria: concept of the understanding.". (E XLIII, p. 24; , Relation


Section rH. On the pure concept s of the underst anding

Now this is the listing of the unci~r~~~n'c1Ill,gC()m:~IJlliJnltS<:,:H{l:""'''L'':C~.",y" ac;ccmnt

At ,,,hi"h

it. Thi2~~iv:i.sioI1jssystematically generatc.;cl namely JP_e. J,!c1ll ty_iQr judging is. for ihliihng), and has not arisen rhapsodically from a haphaz ard search pure concepts, of the completeness of which one could never be since one would only infer it throug h inducti on, withou t rej:1e~:tlJog in this way one would never see why just these and not other concep ts should inhabit the pure underst anding . Aristotle's search for tllese damental concepts was an effort worthy of an acute man. But since had no principle,d he rounde d them up as he stumbl ed on them, and first got up a list of ten of them, which he called catego ries (precllc aments). Subsequently he believed that he had found five more which he added under the name of post-predicaments. But his table had holes. Further , it also included several modi of pure sellsibility ubi, situs, as well as prius, simul,)e as well as an empirical one (7170tUS)! which do not belong in this ancestral registryg of the understanding; derivative concepts were also included among ones (actio, passio),h and several of tlle latter were entirely missing. For the sake of the primary concepts it is therefo re still necessary to remark that the categories, as the true ancestr al concep ts i of pure understanding, also have their equally pure derivat ive) concep ts, could by no means be passed over in a complete system of transce ndental philosophy, but with the mere mentio n of I can be satisfied in a merely critical essay. Let me be allowed to call these pure but derivative concepts the predicahles of pure unders tanding (in contras t to the If one has the original and primitive concepts, the derivative and ternate ones can easily be added, and the family tree k of pure standing fully illustrated. Since I am concern ed here not

A 8I




" The words "of synthesis" are stricken in Kant's copy of the first edition (E XLIV, p. 24; 2346). .\ Object Princip


That is, the concepts of when, where, and position, and the relations of priority and simotion



action, passion Sta771771Z,egriffe emphasized only in the first edition.

k 8W7J1771br11f!H


Doctrine of Elements. Pt. H. Div. 1. Bk I. Ch. I the principlesa for a completeness But one could for anothe r system, I reserve ks in hand, textboo ical ontolog the readily reach this aim if one took predicables the inated subord y, causalit of y e.g., under the categor of presthose nity, commu of that under ; passion of force, action, of genthose ty modali of ments predica resistance; under the ence d combine ies categor The on. so eration, corrupt ion, alteration, mulgreat a yield other each with or either with the modis of sensibility concepts, to take note of which and, as far as titude of derivative a but possible, completely catalogue would be a useful and not unpleasant here dispensable effort. I deliberately spare myself the definitions of these categories in this 3 In the seA83 treatise, althoug h I should like to be in possession of them.' nt in fesufficie is that degree the to ts quel I will analyze these concep of system a In up. g workin am I lation to the doctrin e of method B I09 would they here but me; of these pure reason one could rightly demand only distract us from the chief point of the investigation by arousing be referred to anothe r occasion can doubts and l withou t detract ing from our essentia aim. In any case, from the little a comple te lexicon with that I have here adduce d it becomes clear e but even easy to possibl only be not all the requisite definitions ry to fill them necessa merely is it produc e. The headings already exist; make it easy will one, present the out, and a systematic topic, such as and at the belongs y properl t concep not to miss the place where every empty.b still is that any make it easy to notice same time <

Subtle considerations about this table of categories could be made, the which could perhap s have considerable consequences with regard to uncomis table this that For reason. scientific form of all cognitions of hy monly useful, indeed indispensable in the theoret ical part of philosop insofar science a of whole the for for completely outlinin g the plan acas it rests on a priori concepts, and dividin g it mathematically in from dent self-evi already is d les, cordan ce with detenn inate princip the fact that this table completely contains all the elemen tary concepts
P1711cipien Inserted in Kant's copy of the first edition: e. "What are categories? - - That they extend only to objects of experienc "1. vVhence do they arise? e?" (E XLV; pp. 24-5; 23:25) "2. How are they valid a priori of objects of experienc This explains how Kant can refer edition. second the in , Sections I I and 12 were added 1786. to the lVIetaphysical FOllndations ofNatural Science, not published until



Section Ill. On the pure concepts of the understanding of the understanding, indeed even the form of a system human understanding, consequently that it gives instruction the moments, indeed even of their order, of a planned speculative science, as I have elsewhere given proof.* Now here are several of these remarks. The first is that this table, which contains classes the understanding, can first be split into two divisions, is concerned with objects of intuition (pure as well as errlplflcal), second of which, however, is directed at the existence of (either in relation to each other or to the understanding). I will call the first class the mathematical categories, second, are dynamical ones. As one sees, the first class has no correlates, to be met with only in the second class. Yet this difference must have a ground in the nature of the understanding. Second remark: that each class always has the same of categories, namely three, which calls for reflection, since otherwise a priori division by means of concepts must be a But here first two in its third category always arises from the combination of class. Th\!~~ allness (totality) is nothing other than unity,/limitation is nothing other than reality cOlnbinE~d comrri.unity is the causaHty of a substance in the reciprocal del:ennUla':: cion of others, finally necessity is nothing other than the existence that is given by possibility itself. But one should not think that the third category is therefore a merely derivative one and not an ancestral concept of pure understanding. For the combination of the first second in order to bring forth the third concept requires a special act of the understanding, which is not identical with that act performed in the first and second. Thus the concept of a number (which belqngs to the category of allness) is not always possible wherever the concepts of tude and of unity are (e.g., in the representation the infinite); or influence, i.e., how one substance can be the cause of something in another substance, is not to be understood immediately by cOJmb,ining concept of a cause and that of a substance. From this it is clear a special act of the understanding is requisite for this; and likewise in the other cases. Third remark: The agreement of a single category, of community, which is to be found under the third title, with the form of a disjunctive judgment, which is what corresponds to it in tlle table of logical functions, is not as obvious as in the other cases. In order to be assured of this agreement one must note in all disjunctive judgments the sphere (the multitude of everything that is conMetap,bysi;cal Foundations a/Natural Science. 215


B I 12

B 110

Doctrine of Elements. Pt. II. Div. 1. Bk I. Ch. I divided into parts (the suboris represe nted as a since none these can be contain ed under any dinated concepts), other, they are though t of as coordi nated with one another, not subin ordina ted, so that they do not determ ine each other unilate rally, as dithe of r membe one (if ate a series, but recipro cally, as in an aggreg vision is posited, all the rest are excluded, and vice versa). Now a similar connec tion is though t of in an entiret y of things, of since one is not subord inated, a as effect, under another , as the cause eouslv simultan other the with its existence, but is rather coordi natedb ~ fuld reciprocally as cause with regard to its determ ination (e.g., in other), each repel also yet body, the parts of which reciprocally attract to which is an entirely differen t kind of connec tion from that which is conto ground (of effect to be found in the mere relatione of cause e sequence), in which the conseq uence does not reciprocally determin (as latter the with whole a the ground and therefo re does not constit ute the world- creator with the world). The unders tanding follows the same proced ure when it represe nts the divided sphere of a concep t as when I 3 s it thinks of a thing as divisible, and just as in the first case the member sphere, one in ted of the division exclude each other and yet are connec so in the latter Case the parts are represe nted as ones to which existence are (as substances) pertain s to each exclusively of the others, and which yet connec ted in one whole.



hy But there is also yet anothe r chapter in the transce ndental philosop that, anding underst the of of the ancients that contains pure concepts acalthoug h they are not reckon ed among the categories, nevertheless in objects, of ts concep priori cording to them should also count as a s, categorie the of r numbe which case, however, they would increase the which cannot be. These are expoun ded in the propos ition, so famous alamong the scholastics: quodlibet ens est unum, verum, bonum. d Now very be to out turned has though the use of this principlee for inferences in meager (they have yielded merely tautological propositions), so that sics metaphy in place a it modem times it has been customary to grant courtesy, nevertheless a though t that has sustained italmost solely self so long, no matter how empty it seems, always deserves an investiits gation of its origin, and justifies the conjecture that it must have


Verhiiltnis Every being is one, true, and good. Princips


Section Ill. On the pure concept s of the underst anding

ground in some rule of the underst anding , which, as so often happen s, has merely been falsely interpr eted. These suppos edly transc:en(jelllta l predicates of things are nothing other than logical requisit es and criteria of all cognit ion of things in general , and ground it in the categor ies of quantity, namely, the categor ies of unity, plurali ty, and totality ; yet these categories must really have been taken as materia l, as DeilOflg lng to the possibility of things itself, when in fact they used in a merely formal sense, as belong ing to the logical re({uirelnents for every cogniti on; thus these criteria of thinkin g were made into propert ies of things in themse lves. In every cogniti on of an nr"Pr1r there is, namely, unity of the concep t, which one can unity insofar as by that only the of the of manifold of cogniti on is though t, as, say, the unity of the theme in a play, a speech, or a fable. Second , truth in respect of the conseq uences. The more true conseq uences from a given concep t, the more in(jicat ion of its objective reality. Onc could caU this the qualita tive pl1LIT2l1 1ty the marks that belong to a concep t as a commo n ground of in it as a magnit ude). Third, finally, perfect ion, which consists in plurality convers ely being traced back to the unity of the concep t, and agreeing comple tely with this one and no other one, which one can call qualitative comple teness (totality). From this it is obvious logical criteria of the possibi lity of cogniti on in general transfo rm the three categories of magnitude,c in which the unity in the generat ion of the magnitude d must be assume d to be comple tely homog eneous , into a principle' with the quality of a cogniti on for the connec tion heterogene ous elemen ts of cogniti on into one conscio usness also. the criterion of the possibi lity of a concep t (not of its object) fis the definition, in which the unity of the concep t, the truth of everyth ing may initially be derived from it, and finally the compl etenes s of thing that is drawn from it, constit ute everyth ing that is necessa ry the production of the entire concep t; or the criteri on a hypoth esis is also the intellig ibility of the assume d ground of explan ation or its unity (without auxiliary hypothe ses), the truth (agreem ent itself and wiu~ experience) of the conseq uences that are derived from it, and finally the comple teness of the ground of explana tion of these consequences, which do not refer us back to anythin g more or less than was already assumed in the hypothe sis, and which merely an;llV1tIGtllv back a posteriori and agree with that which was though t syrlthetI1cal1y

B 114




Doctrine of Elements. Pt. H. Div. 1. Bk I. Ch. I

priori. - The transce ndental

B 116

the categories is thus not comperfect ion, as if it were pleted with the concep ts of unity, a of these concep ts to objects b lacking someth ing, but rather, the relation being entirely set aside, our proced ure with these concep ts is only being with though t under general logical rules for the agreem ent of cogniti on itself.>

Verhalmis Objeete


The Transcendental Analytic Second Chapter On the Deduction ofthe Concepts the Understanding

First Section < 13>a On the principles b of a transce ndenta l deduct ion in general. 14 1urists, when they speak of entitlements and claims, distinguish in a legal matter between the questions about what is lawful d (quid juris)
" Paragraph number added in the second edition. In the first edition, the second chapter of the "Transcendental Analytic," the "Transce ndental Deductio n," is divided into three main sections, the first of which is in turn subdivided into two subsectio ns. Apart from a few minor changes in wording, which will be noted, and dle addition of the section numbers themselves, the two subsections of the first section are retained in the second edition and are identical until the last paragrap h of their second subsectio n, which is replaced by three new paragrap hs in the second edition. The second and third sections of the chapter in the first edition are then replaced by an entirely new second section in the second edition, which is broken up into numbere d paragrap hs. IS through 27. We "ill present all of this material in the following sequence: the first section as it appeared in both editions, with the last paragrap h of the first-edition version followed the last mree paragraphs that replaced it in the second edition; the second and dlird sections as mey appeared in me first edition; then the second section, consistin g of numbered parts IS through 27, as it appeared in the second edition. b Prinripicn , The following notes are inserted here in Kant's copy of the first edition: "Consciousness and inner sense are different. 'I think' is spontane ity and does not depend on any object. The represen tation, however, with which 1 dlink, must be given to me antecedently in intuition (through imagination). With regard to it I am affected." (EXLVI,p. 25; 23'16) "It must be proved that if there were no sensible intuition a priori, and if this were not the form of sensibility in me subject, with which all appearances must be in accord, then: "1. No categorie s would have significance. "2. From mere categorie s no synmetic a priori propositi ons at all would be possible." (EXLVII,p. 25; 23'16) d was Rechtens ist


1. Ch. II Doctrine of Elements. Pt. n. Div. 1. Bk. and since they dem and proof whi ch con cern s the fact or which is to establish the enti tlem ent call the first, of both , emof de titu mul a of 's We make use the legal claim, the ded ucti on. s to n from anyone, and take ourselve ctio obje out with pirical con cep ts even tion ifica sign ed pos e and a sup be justified in gran ting them a sens always have experience ready at we use beca on, wiu."'lout any deducti But ther e are also concepts that han d to prov e thei r objective B I 17 and fate , which circulate with alhave bee n usurped, such as fort une are occasionally called upo n to esIliost universal indulgence, but that juris, and then ther e is not a thei r claim by the que stio n quid ucti on because one can adduce no little emb arra ssm ent abo ut thei r ded rience tlem ent to thei r use eith er from expe A8S dea r legal gro und for an enti or from reason. mixed however, that con stitu te the very a AmongJ:hem~nY con cep ts, for ined dest also an~ are some that fabric of hum an cog niti on, ther e these and , nce) erie exp (completely inde pen den tly of all pur e use r enti tlem ent, since proofs from exalways req uire a ded ucti on of thei lawfulness of such a use, and yet one peri enc e are not sufficient for the be related to objects b that they do mus t kno w how thes e concepts can ther efor e call the explanation of the not derive from any experience. I cts a priori u~eir tran scen den in whi ch concepts can rela te to obje iric al deduction, and distinguish this from the emp tal nce and reHecerie exp uired thro ugh whi ch shows how a con cep t is acq the fact from but ess fuln not the lavv tion on it, and ther efor e con cern s which the possession has arisen. concepts of an entirely different Now we already have two sorts of B II8 r in that they bot h relate to objects kind,c whi ch yet agree with each othe cepts of space and time, as 'forms of com plet ely a priori, nam ely the con cepts of the understanding. To seek sensibility, and the categories, as con ld be enti rely futile work, for what an empirical ded ucti on of them wou their obis precisely t.~at they are related to A86 is distinctive in t-heir natu re r repthei for nce erie exp thin g from jects with out having bor row ed any it mustalways be resentation. Thu s if a transcendental. concepts, as in the case of all cogNevertheless, in the case of these if not for the principled of their niti on, we can search in experience, causes of thei r generation, where the possibility, t-hen for the occasional the first occasion for opening the enimpressions of the senses provide

a bf',ltJnl111t b

Objecte from experience" inserts: "The y are not borr owed Rant 's copy of the first editi on (E XLVTII, p. 25; 23'46 ).


Section 1. On the principl es of a transcen dental deductio n

, tire power of cogniti on to them bringin g which contains two very heterog eneous elements, a matter cognition from the senses 4. cenain form. for orderin g it from the inner source of pure intuitin g and thinking, on the of first brough t into use and bring forth concepts. Such a tracing of the first endeavors of our power of cogniti on to ascend individual percep tions to general concep ts is of great B II9 utility, and the famous Locke is to be thanked for having first op,en(~C! the way for this. Yet a deduct ion of the pure a priori concepts can never be achieved in this way; it does not lie down this at for in regard to their future use,,which should be entirely indepe ndent of experience, an entirely differen t birth certificate than that of an ancestr y from experiences must be produc ed. I will therefo re call this attemp ted physiological derivation,I6 which cannot properl y be called a A87 at all because it concern s a quaestio facti,a the explanation of posses s10n of a pure cognition. It is therefo re dear that only a transcendental and never an empirical deducti on of them can be given, in regard to pure a priori concepts empirical deducti ons are nothin g 'attempts, which can occupy only those who have not grasped the entirely distinctive nature of these cognitions. But now even if the sole manne r of a possible deducti on a priori cognition is conceded, namely that which takes the transcendental path, it is still not obvious that it is unavoidably necessary. We above traced the concepts of space and time to their sources by means ofa transcendental deduction, and explained and determ ined their a priB IZO ori objective validity. Geome try nevertheless follows its secure course through strictly a priori cognitions withou t having to beg for any certification of the pure and lawful pedigree of its fundamental concept of space. Yet the use of the h concep t in this science concern s the external world of the senses, of which space is the pure form of its intuition, and in which therefo re all geometrical cognition is Immed Iately evident because it is ground ed on intuitio n a priori, and the objects are given through the cogniti on itself a priori in intuitio n (as far as their A88 form is concerned). With the pure concep ts of the unders tandin g, however, there first arises the unavoidable need to search for the transcendental deduction not only of them but also of space, for since speak of objects not throug h predicates of intuitio n and sensibility through those of pure a priori thinking, they relate to objects general ly without any conditions of sensibility; and since they are not ground ed in elTJerience and cannot exhibit any object C in a priori intuitio n on which
As in the first edition; the second, declining quaestio, prints quaestionom. The first edition here reads "dieses" instead of the second's "des."


Doctrin e of Elements. Pt.

n. Div. I. Bk. 1. Ch. n

arouse to any experience, they not to ground their synthesis also but use their of limits e the objectiv suspicion beyond it use to us g inclinin ous ambigu space make the concep t of accoun t a transcendental the conditions of sensible intuition, on B 12 I the reader must be conThus above. needed also was deducti on of it ndental deduction transce a such of ty necessi able vinced of the unavoid for he would reason; pure of field the in step before he has taken a single would still around ing wander much after and otherwise proceed blindly, he must But begun. had he which from ce ignoran have to return to the so that ty, difficul le inevitab its outset the from and also clearly underst is itself -matter subject the where ty obscuri of in not compla he runof l remova the over soon too d annoye deeply veiled or become drances, since we must either surrend er completely claims to insights A 89 ies of pure reason in its favorite field, namely that beyond the boundar ation. investig critical this perfect else or of possible experience, In the case of the concepts of space and time, we were able ahove to ns, make compre hensibl e with little effort how these, as a priori cognitio syna e possibl made and , objects to relate must nevertheless necessarily obthetic cogniti on of them indepe ndent of all experience. For since an ty, sensibili of forms pure such of means ject can appear to us only pure a thus are time and space n, i.e., be an object of empirical intuitio obintuitio ns that contain a priori the conditi ons of the possibility of B I 22 validity. e objectiv has them in is synthes jects as appearances, and the The categories of the underst anding , on the contrary, do not repreat sent to us the conditi ons under which objects are given in intuition to having rily necessa t withou us to appear all, hence objects can indeed the be related to functions of the underst anding , and therefo re withou t is y difficult a Thus '7 ons. conditi priori a unders tanding contain ing their ty, sensibili of field the in ter encoun not did revealed here that we e namely how subjec tive condit ions of thinkin g should have objectiv of n cognitio all of lity possibi the of ons validity , i.e., yield conditi A90 objects; for appearances can certainly be given in intuitio n without functions of the underst anding . I take, e.g., the concep t of cause, which ing A signifies a particular kind of synthesis, in which given someth b It is not rule. a to ng accordi posited is B t someth ing entirely differen clear a priori why appearances should contain anythin g of this sort (onc of cannot adduce experiences for the proof, for the objective validity thereis it and trated), demons be to able this a priori concep t must be fore a priori doubtfu l whethe r such a concep t is not perhaps entirely empty and finds no object anywhere among the appearances. For that

Emende d in Kant's copy of the first edition to "posited according to necessarily" (E XLIX, p. 25; 23:46).


an a priori rule, i.e.,


Section I. On the principl es of a transcen dental deducti on

not find them in accord with the conditi ons of its and PV"M7T h,na would then lie in such confusion that, e.g., in the succession of jJnnp:~r ances nothing would offer itself that would furnish a of <:v"tr'P<: l<: and thus corresp ond to the concep t of cause and effect, so that cept would therefo re be entirely empty, nugatory, cance. Appear ancesw ould to our intuition by no means requires the mJJ.c1:1011S0t .ttnnk111g. If one were to think of escaping from the toils of these investigations by saying that experience constantly offers examples of a regularity of appearances that give sufficient occasion for abstrac ting the COJt1C(;p t of cause from them, and thereby at tl1e same time though t to cOlJ.h:rm the objective validity of such a concept, tl1en one has not noticed the concept of cause cannot arise in this way at all, but must either be grounded in the unders tanding comple tely a priori or else be entir,elv surrendered as a mere fantasy of the brain. For this concep t quires that someth ing A be of such a kind that someth ing else B from it necess arily and in accord ance with an absolu tely univers al rule. Appearances may well offer cases from which a mle is possible in accordance with which someth ing usually happens, never a in accordance with which the succession is necess ary; to the sVTltflesis of cause and effect there attaches a dignity that can never be expressed empirically, namely, that the effect does not merely come along with the cause, but is posited throug h it and follows from it. The strict universality of the mIe is therefo re not any propert y of empirical which cannot acquire anythin g more throug h inducti on than compar ative universality, i.e., widespread usefulness. But now the use of the pure concepts of the unders tanding would be entirely altered if one were to treat them only as empirical products.
" Following Erdmann in reading "Einheit" for "Einsicht"; Kant uses "Einheit" in a fashion in the next sentence. b Inserted in Kant's copy of the first edition: "If I were simply to say that without the connection of causes and effects I would not grasp the sequence of alteration s, it would not at all follow from this that this must be precisely as an understa nding needs it to be to grasp it, but I would not be able to explain whence they continuo usly follow one another. Only J would not raise this question if I did not already have the concept of cause and of the necessity of such persistence. A subjective necessity , habit, would make it worse. An implante d necessity would not prove necessity." (E L, pp. 25-6; 2312 6)

;:~~~~~~~~1:~~; a con~~usiontlt~i:(;(lf'sth.~t-th~;{~nd~rsi:3nclinf{ ,"'QuId

wise ~<:Y_"'~lll~_?:()!b~~_]~~i~~f()l~ll~;but ll1" '=---''ll!<_yUDU''! ;,"u with the conditi ons. that the unders tanding rpr'Hi'rp~ t!l~e__ ,:~}';lthe

obje{:J;s~~siQ1~in tlIitionmus~accQrci seJ],sibility_lli?J;Ji~iRJ:hemi~ d a prjorLis dear

B 123






Doctrine of Elements. Pt.

n. Div. 1. Bk. I. Ch. n

the categor ies.

aTrans ition to the transce ndenta l deduct ion

B 125

two possible cases in which synthetic represe ntation and There are as it its objects can come togethe r, necessarily relate to each other, and, ntaeflFese S-the-F nemake bjectalo -itheo were, meet each other: Eithel" . --UoUcPDSsible, or if the represe ntation alone makesci1e object possible
relat1on_J~~~s- Olliy
Arld.. thJS J:'iJ:jle_Ca~,e_'wJJ:h:1PJ)e::l~Glll'Ce III re-


be1o'llgsto s(~nsatI,on. 1..> U~L LLJJ_._'~c LIlt; second, its (for we are not here talking about its as far s objecLa its ce t.produ doesnQ causality by means of the of the eXJl.tlenlre..Js.~cono:xnled, the re.pJ:es.emation is still as ne$ng :z;esOl <:o~rni -lto lulloJ:le lgh PQ1sslIlletbtOl cognithe alone which under ons ~ohj~ct. But there are two conditi tion of an object is possible: first, intuitio n, L~rough which it is given, is but only as appearance; second, cQ...ncept, throug h which an obj~ct has what from dear is It n. A93 though t that corresp onds to this intuitio been said above, however, that the first condition, bnamely that under the mind a priwhich alone objects can be intuited , in fact does lie c therefore necnces appeara All ori as the ground of the form ofobjec ts. because only ity, sensibil of on conditi essarily agree with this given. The and intuited ally empiric be throug h it can they appear, i.e., e,ascon prececl notalsQ do ts concep questio n now is whethe r a priori neverthe, intuited not if be, can ing ditions under which alone someth n of cognitio al empiric all then for , less though t as object in general their t withou since ts, concep necessarily in accord B 126 d presupp osition nothing is possible as ohject of experie nce. Now, the however, all experience contains in additio n to the intuitio n of that object an of t concep a given, is senses, throug h which someth ing is given in intuitio n, or appears; 18 hence concepts of objects in general lie at the ground of all experiential cogniti on :olS:-ipiioricouditions; cononcepts, S~(:l~~~~t~~._the objective validity of the categories, as a priori--r (as far possible nce experie is that rests on necesrelated are then they For ed). as the form of thinkin g is concern of them sarily and a priori to objects of experience, since only by means be ~Qltght at an~----'-~ "~--""~._----~--of can any

should have been No section number appears here in the second edition, but " 14" added to avoid an unnumbe red section between 13 and IS. confused the sinFollowing Erdmann in reading "liegt" for "liegen"; Kant seems to have with the plural, perhaps because of the intervening occurgular antecede nt rence of the plural


Objecten Object



I. On the principl es of a transcen dental deductio n

The transce ndental deducti on of a priori concepts therefo re a principlea toward which entire investigation must namely this: that they must be recognized as a possibility of experiences {wheth er of the intuitio n in them, or of the thinking).1 9 Concep ts that ground of the possibility of experience are necessary son. The unfoldi ng of the experience in they are enl:oumt<"red, however, is not their deducti on (but their illustration), since thereby be only conting ent. VVithout this original relation to B 127 experience, in which all objects of cogniti on are found, their to any object b could not be compre hended at all. '[There are, however, three original sources (capacities or faculties the soul), which contain the~conditions of the possibility of all experience, and cannot themselves be derived from any other of the mincL namely sense, imagin ation, apperc eption . On these are grounded I) the SynOPSi1LQf the manifold a priori throug h sense; 2) the synthesis of this manifold throug h the imagination; 3) the of this synthesis throug h original apperception. In to empirical use, all of these faculties have a transce ndental one, which is concerned solely with form, and which is possible a priori. have discussed this with regard to the senses in the first above, A95 we will now attemp t to unders tand the nature of the two other ones.] d<The famous Locke, from neglect of this consideration, and because B 127 he encountered pure concepts of the unders tanding in experience, also derived them from this experience, and thus proceed ed so incons istently that he thereby dared to make attemp ts at cognitions go far beyond the bounda ry of all experience. David Hume recognized that in order to be able to do the latter it is necessary that these concepts have to have their origin a priori. But since he could hot explain at all how it is possible for the unders tanding to think of concepts in themselves are not combin ed in the unders tanding as necessarily combined in the object, and it never occurre d to him perhap s understanding itself, by means of these concepts, could be the origina tor of the experience in which its objects are encoun tered, he thus, driven by necessity, derived them from experience (namely from a subjective necessity arisen from frequen t association in experience, which is subsequently falsely held to be objective, i.e., custom );e howe'iTer
Principitr71l Object , This paragraph in the first edition is omitted in the second and replaced by three that here follow it. The next three paragrap hs are added in the second edition, replacing the previous onc. e GC7l'ohnhcit



Doctrin e of Elemen ts. Pt.

n. Div. 1. Bk. T.



: 129

subseq uently proceed ed quite consistently in declaring it to be imposand sible to go beyond the bounda ry of experience with these concepts r, howeve ion, derivat cal empiri the principles that they occasion. Thc reality the with led reconci be cannot to which both of them resorte d, of the scientific cogniti on a priori that we possess, that namely of pure by mathem atics and genera l natura l science , and is therefo re refuted a the fact. The first of these two famous men opened the gates wide toenth u' siasm, since reason, once it has authori ty on its side, will not be J:cept the tion; modera of ns endatio recomm within limits by indeter minate to second gave way entirely to skeptic ism, since he believed himself of on decepti a reason be to have discovered in what is generally held see to t attemp an make to about our faculty of cognition. - We are now two whethe r we cannot successfully steer human reason betwee n these entire the open keep still and cliffs, assign its determ inate boundaries, field of its purposive activity. I will merely precede this with the explan ation of the categories. They are concepts of an object in general, by means of which its intuition is regarde d as detenn ined with regard to one of the logical funct tions for judgments. 20 Thus, the function of the catego rical judgmen "All e.g., te, predica was that of the relationship of the subject to the unbodies are divisible." Yet in regard to the merely logical use of the concepts two these of which derstan ding it would remain undete rmined of will be given the functio n of the subject and which will be given that body." a is e divisibl thing "Some the predicate. For one can also say: of Throug h the categor y of substance, however, if I bring the concept experiin n intuitio al empiric a body under it, it is determ ined that its and ence must always be considered as subject, never as mere predicate; likewise with all the other categories.> The Deduc tion of the Pure Concep ts of the Unders tandin g Second Section

t t

t a cl c p n



cl P

bO n the a priori ground s for the possibi lity

of experie nce.

It is entirely contrad ictory and impossible that a concep t should be it genera ted comple tely a pn'ori and be related to an object although

oJ of

bl pr

das Factum

as it appeated in the \VhaL follows is the version of the "Transee ndenta1 Deductio n" of the present chapfirst edition, where it is divided into the second and third sections second section, single a by replaced be will sections two these ter. In the second edition, below. divided into subsections numbere d from IS to 27. See BI29-69



Section n. Grounds of the possibility of experience <A> neither belongs itself within the concep t of possible experience nor cOnsists of elemen ts of a possible experience. For it would then have no content, csince no intuitio n would corresp ond to it though intuitio ns in general, throug h which objects can be given to us, constit ute the field or the entire object of possible experience. concep t was not relateda to the latter would be only the logical form for a C011CC:p t, but not the concep t itselfth rough someth ing If there are pure a priori concepts, therefore, they can contain nbthing empirical; they must nevertheless be a priori conditions for a possible experience, as that alone on its objec;ti,Te reality can rest. Hence if one wants to know how pure concep ts of the understan(iing ate-possible, one must inquire what are the a priori on the possibility of experience depend s and that ground it even if One abstracts from everyth ing empirical in the appearances. A concep t expresses this formal and objective conditi on of experience unlversal1v and sufficiently would be called a pure concep t of the underst anding . Once I have pure concep ts of the underst anding , I can also up objects that are perhap s impossible, or tlut are perhap s possible in themselves but cannot be given in any experience since in connec tion of their concepts someth ing may be omitted that yet necessarily belongs to the conditi on of a possible experience (the concep t of a spirit), or perhaps pure concepts of the unders tanding will be extended further than experience can grasp (the concep t of God). But the elemen ts for all apriori cognitions, even for arbitrar y and absurd fantasies, cannot indeed be borrowed from experience (for then they would not be a priori cognitions), but must always contain the pure a priori conditions of a possible experience and of an object of it, for otherwise not only would nothing at all be though t through tl1em, but also data would not even be able to arise in thinkin g at all. Now these concepts, which contain a priori the pure thinkin g in every experience, we find in the categories, and it is already a sufficient deduction of them and justification of their objective validity if we can A97 prove that by means of them alone an object can be though t. But since in such a though t there is more at work than the single faculty of thinking, namely the underst anding , and the unders tanding itself, as a faculty of cognition that is to be related to objects,b also requires an elucidation of the possibility of this relation, we must first assess not the empiric al but the transcendental constit ution of the subjective sources that comprise the a priori foundations for the possibility of experience. If every individual represe ntation were entirely foreign to other, as


Doctrine of Elements. Pt. H. Div. 1. Bk. 1. Ch. II <Pi> it were isolated separated from then there would never arise anything like cognition, is a whole of compared and connected representations. If therefore I ascribe a synopsis to sense, because it contains a manifold in its intuition, a synthesis must always correspond to this, and receptivity can make cognitions possible only if combined with spontaneity.ltrhis is now the ground of a threefold synthesis, which is necessarily foUnd in cognition: that, namely, of the apprehension of the representation~, as modifications of the mind in intuition; of the reproduction of them in the imagination; and of their recognition in the concept. ! Now these direct us toward three subjective sources of cognition, which make possible even the understanding and, through the latter, experience as an empirical product of understanding.


Preliminary reminder The deduction of the categories is connected with so many difficulties, and necessitates such deep penetration into the primary grounds of the possibility of our cognition in general, that in order to avoid the longwindedness of a complete theory and nevertheless not to omit anything in such a necessary inquiry, I have found it more advisable to prepare than to instruct the reader in the following four numbers, and only then to represent the exposition of these elements of the understanding systematically in the immediately following third section. For this reason the reader should until then not be deterred by the obscurity that is initially unavoidable in a path that is thus far entirely unexplored, but which as I hope, be completely illuminated in that section.


On the synthesis of apprehension in the intuition. Wherever our representations may arise, whether through the influence of external things or as the effect of inner causes, whether they have originated a priori or empirically as appearances - as modifications of the mind they nevertheless belong to inner sense, and as such all of our cognitions are in the end subjected to the formal condition of inner sense, namely ~, as that in which they must all be ordered, connected, and brought into l"~l~tions. This is a general remark on which one must ground everything that follows. 22 Every intU!t1on contains a in itself, which however would not be represented as such if theinind did not distinguish th~_tirne in the succession of impressions on one another; for as contained ~one



The third section, beginning at AIl 5.


Section H. Grounds of the possibility of experience <A> momen t no represe ntation can ever be anythin g other unity. Now in order for unity of intuitio n to come this m~mitold (as, say, in the represe ntation of space), it is necessary thr-.QI!.gh and thed':to t~~~J:g~th~1:l1b~J!1aIJ,if()Idness, which action I the synthe sis of appreh ension , since it is aimed directly at intuition, which to be sure provides a manifold can never effect as such, and indeed as contain ed in one represe ntation , the occurrence of such a synthesis. Now this synthesis of appreh ension must also be exercised a i.e., in regard to represe ntation s that are not empirical. For withcm t we could have a priori neither the represe ntation s space nor since these can be generat ed only throug h the synthesis of the that sensibility in its original receptivity provides. We therefo re have a pure synthesis of apprehension.


On the synthes is of reprod uction in the imagin ation. It is, to be sure, a merely empirical law in accordance which representations that have often followed or accompanied one anothe r are finally associated with each other and thereby placed in a connec tion in accordance with which, even withou t the presence of the object, one of these represe ntation s brings about a transition of the mind to the other in accordance with a constan t nue. This law of rerlrodu,cu()n, howf':ve r. presupposes that the appearances themselves are subject to such a rule, and that in the manifold of their represe ntation s an accompaniment or succession takes place accordi ng to certain rules; for out that our empirical imagin ation would never get to do anythin g suitable to its capacity,a and would thus remain in interio r of the mind, like a dead and to us unknow n faculty. If were now red, now black, now light, now heavy, if a human being were now changed into this animal shape, now into that one, if on the longest the land were covered now with fruits, now with ice and snow, then my empirical imagin ation would never even get the to think of heavy cinnabar on the occasion of the represe ntation of color or if a certain word were attribu ted now to this thing, now to or if one and the same thing were sometimes called this, sometimes without the govern ance of a certain rule to which the appearances are already subjected in themselves, then no empirical synthesis of reproduction could take place. There must therefo re be someth ing that itself makes possible this re-




produc tion the appearances by being the a priori ground of a necesof them. One soon comes upon this if one re;;.a1ls sary synthetic rather the mer~ ' themselves, that to determinations down come end the in s, our represe ntation even ouipur est a that strate demon can we if inner.s ense. Now they contain the as insofar except on cogniti no priori inmitio ns provide e a thoroug hpossibl makes that ld manifo sort of combin ation of the imagination the of is synthes this then going synthesis of reprodu ction, principles,a priori a on nce experie all to prior would be ground ed even power, this of is synthes ndental transce pure a and one must assume the which that (as nce experie all of lity which ground s even the possibi is it Now oses). presupp rily necessa nces reprodu cibility of the appeara one from time the of think or t, though in line obvious that if I draw a noon to the next, or even want to represe nt a certain numbe r to myself, I must necessarily first grasJ4'me of these manifold representation~fter anothe r in my though ts. But if I were always to lose the precedin~repthe line, the precedi ng parts of time, or resenta tions (the first parts the successively represe nted units) from my though ts and not~ reprewhole no then , . ones. ng followi I proceed to the the sentati on and none of the previously mentio ned though ts, not even could time, and space of s ntation most fundam ental represe purest ever arise. The synthesis of appr..~;)ion is therefo re inseparably combined the with the synthesis of reprodu ction. And since the former constitutes (not general in on cogniti all of lity possibi transce ndental ground orthe reonly of empirical cogniti on, but also of pure a priori cognition), the n tran~~e tl1~ among belongs ation produc tive synthesis ofthe imagin this c:ill also will we this to with dental actions of the mind, and faculty the trar~~~_~~t~ltaC:lll.ty.2Lt:hE:.. !JJ}

s~ '0~.3\

Doctrin e of Elements. Pt. II. Div. I. Bk. I. Ch. II <A>


3 On the synthes is of recogn ition in the concep t.

W-i4o ut consciousness as repwhJlD'lC.e.J:h.o.llghLamQJ.!l~m1:l~f()Ie,!ll1 . repr()d1J.ctiQl}

ion in resenta tions would be in vain. For it would be a new~!esentat b through tI1e"'act to all at belong not would our current state, which it had been gradually generat ed, and its manifold would never constit ute a whole, sins~itw~ lack the unity that only consciousness can obtain for it. If, it): counting,)I forget that the units that now h..Q~r
~~~-~-~.-<~,./ ~

cr. b Actus; up to this point Kant has been using the word Hm'7diJl11

Pr-in ([~~ien


Section II. Grounds of the possibility of experience <A>

before my senses were successively added to each I would not cognize the generation of successive addition of one to the other, and consequently I would not nize the number; for this concept consists solely in the COJ~c:iOllsness The word "concept" itself could already it is this one consciousness unifies that cessively intuited, and then also reproduced, into one rq)re:seJlltalticlll. This consciousness may often only be weak, so that we connect it with the generation of the representation only in the effect, but not in the act b itself, i.e., immediately; but regardless of these differences one consciousness must always be found ,v even lacks conspicuous without that concepts, and with them cognition of objects, tireJy impossible. And here then it is necessary to make understood is meant the expression "an object of representations." We have said appe;lr;ll1CeS themselves are nothing but representations, must not be regarded in themselves, in the same way, as objects (OlltSJlde the power of representation)!What does one mean, if one of an object corresponding tb and therefore also distinct from cognition? It is easy to see that this object must be thought as something in general = X, since outside of our cognition we have nothIng that we could set over against this cognition as corresponding to it. We find, however, that our thought of the relation of cognition to its object carries something of necessity with since namely the iEegarded as that which is opposed to our cognitions being detelmJme~d at pleasure or arbitrarily rather than being determined a sofar as they are to relate to an object our cognitions must neceshave that sarily agree with each other in relation to it, unity that constitlle:~_th~C:()l1ceptofanobje'::t: 23 It -is dear~however, that since we have to do of our representations, and that X which corresponds to (the object), because it should be something distinct from all of our representations, is nothing for us, the unity that the object makes necessary can be nothing other than the formal_:::o':C"J. __,..~ th~c:Ql1c:jQllm(~ssjllthles;yn th_e~is of.!he l1l.~llifold.~()Lth~ ..r~~~.l1te!!Qm: Hence we say that we cognize the object if we have effected synthetic unity in the manifold of intuition. But this is impossible if the intuition could not have been produced through a function of synthesis in accordance with a rule that makes the reproduction of the manifold necessary a priori and a concept in which this manifold is united possible. Thus we think of a triangle as

A !O4




Doctrin e of Elements. Pt.

n. Div. I. Bk. I. Ch. II <A>

compo sition of three straigh t lines being conscious of an object to which such an intu.ition can aI, ng accordi a rule in accordance nI!e determ ines every manifold, of u~, ways be exhibited. Now this unity of apperce ption possible, the make that ons and limits it to conditi ntation of the object = J{,'represe the is this concep t of and . triangle a of tes predica those which I think throug h ct or obscure it imperfe r howeve t, concep a ~l cogniti on requires ,106 something always is latter the ed concern may be; but as far a;; its form is t of body concep the Thus rule. a as serves that general, and someth ing of means by nces appeara outer of on cogniti our serves as the rule for be can it er, Howev it) h throug t though is that ld the unity of the manifo of/ a rule of intuitio ns only if it represe nts the necessary reprodu ction con, the in unity ic synthet the hence ns, the maillfold of given intuitio ing sciousness of them. Thus in the case of the percep tion of someth of ntation represe u~e ry necessa makes outside of us the concep t etc. shape, of , trability impene of extension, and with it that Every necessity has a transce ndental conditi on as its ground . A tran, scendental ground must therefo re be found for the unity of the con, sciousness in the synthesis of the manifold of all our intuitions, hence a ob, also of the concepts of objects in general, consequently also of all any of think to ible imposs be would jects of experience, withou t which it object for our intuitions; for the latter is nothing more than the some, thing for which the concep t expresses such a necessityb of synthesis. Now this original and transce ndental conditi on is nothin g other than 24 in the transce ndenta l apperc eption . The consciousness of oneself A 107 ion percept l interna in state our of s accordance with the determ ination is merely empirical, forever variable; it can provide no standin g or abiding self in this stream of inner appearances, and is customarily called rinner sense or empiri cal apperc eption . That which should necessa such as of t though be cannot al identic cally ily be represe nted as numeri exthroug h empirical data. There must be a conditi on that precedes all a such make should which e, possibl perienc e and makes the latter itself transce ndental presupp osition valid. Now no cognitions can occur in us, no connec tion and unity among the them, withou t that unity of consciousness that precedes all data of alone is objects of ntation represe all intuitions, and in relation to which possible. This pure, original, unchan ging consciousness I will now alname transce ndenta l apperc eption . That it deserves this name is namely unity, e objectiv purest the ready obvious from this, that even that of the a priori concepts (space and time) is possible only through the relation of the intuitio ns to it. The numeri cal unity of this appera

Objecte Following Erdmann , reading NnJ'!17"p"rlirrp,oit for NOl'lnu,md,'z.


Section H. Ground s of the possibility of experience <A>

ception therefo re grounds concepts a pnorl, Just as of space and time ground s the intuitio ns of sensibility. Just this tnmsce ndental unity apperception, however, makes out of A lO8 all possible appearances that can ever come togethe r in one experie nce a connection of all of these represe ntation s accordance with laws. 2s For this unity of consciousness would be impossible if in cogniti on of the manifold the mind could not become conscious of the function by means of which this manifold is syrlthetl,cally cI)mbiI led into one cogniti on. Thus the original and necessary consciousness of the identity of oneself is at the same time a consciousness of an pm,.., 1I " necessary unity of the synthesis of appearances in accordance concepts, i.e., in accordance with rules that not only make necessarily reproducible, but also thereby determ ine an object for their intuition, i.e., the concep t of someth ing in which they are necessa rily connected; for the mind could not possibly think of the itself in the manifoldness of its representations, and indeed this a priori, if it did not have before its eyes the identity of its action, subjects all synthesis of appreh ension (which is to a transcendental unity, and fir?t makes possible their connec tion in accordance with a priori rules. furthe r, we are now also able to our concepts of an object in general more correctly. All representations, as representations, have their object, and can themselves be of other representations in turn. Appearances are the only objects that can be given to us immediately, and that in them which is immediately re- A lO9 lated to the object is called intuitio n. However, these appearances are not things in themselves, but themselves only represe ntation s, which in turn have their object, which therefo re cannot be further intuited us, and that maY,therefore be called the non-empirical, i.e., translcerlde:nt:al object= X 26 / (fhe pure 'concep t of this transce ndental object (which in all of our cognitions is really always one and the same = X) is that which in all of our' empirical concep ts in general can provide relation to an object, i.e., objective reality. Now this concep t cannot contain any determ inate intuition at all/ and therefo re concern s nothing but that must be encdun tered in a manifo ld of cogniti on insofar as it stands in relation to an object. This relation , however, is nothin g other than the necessary unity of consciousness, thus also of the synthesis of the manifold through a commo n function of the mind for combin ing it in one representation. Now since this unity must be regarde d as necessary a (since the cogniti on would otherw ise be withou t an object), relation to a transce ndental object, i.e., the objective reality of our empirical cognition, rests on the transce ndental law that all appearances, A I 10 insofar as objects are to be given to us throug h them, must stand under rules of their synthet ic unity, in accordance with which their re233

Doctrine of Elements. Pt. II. Div. I. Bk. 1. Ch. II <A> nce lationo in empirical intuitio n is alone possible, i.e., that in experie papperce they must stand under conditi ons of the necessary unity of condiformal tion just as in mere intuitio n they must stand under the that tions of space and time; indeed, it is throug h those conditi ons every cogniti on is first made possible.

Provisi onal explan ation of the possibi lity of the categor ies as a priori cogniti ons. There is only one experience, in which all percept ions are represented one as in thoroug hgoing and lawlike connec tion, just as there is only b of relation all space and time, in which all forms of appearance and ces, experien t being or non-be ing take place. If one speaks of differen the they are only so many percept ions insofar as they belong to one and of unity ic same universal experience. The thorou ghgoin g and synthet it and nce, percept ions is precisely what constitutes the form of experie lccoris nothing other than the synthetic_ unity ofilieap pearan ces-inl
~!tc:~"With c:oncepts.

Unity of synthesis in accordance with empirical concepts would be entirely conting ent, and, were it not ground ed on a transcendental fill ground of unity, it would be possible for a swarm of appearances to in But it. up our soul withou t experience ever being able to arise from since ar, that case all relation of cogniti on to objects would also disappe l the appearances would lack connec tion in accordance with universa but t, though t and necessary laws, and would thus be intuitio n withou never cognition, and would therefo re be as good as nothing for us. The a priori conditi ons of a possible experience in general are at the ce. same time conditi ons of the possibility of the objects of experien nothare d Now I assert that the catego ries that have just been adduce ing other than the condit ions of thinkin g in a possib le experience, the just as space and time contain the condit ions Qf the intuiti on for for s concept very same thing. They are therefo re also fundam ental e C therefor thinkin g objects in general for the appearances, and they to wanted have a priori objective validity, which was jnst what we really know. However, the possibility, indeed even the necessity of these cateall gories rests on the relation that the entire sensibility, and with it also everywhich possible appearances, have to the original apperce ption, in hthing is necessarily in agreem ent with the conditions of the thoroug

Verhiiltni, Verhiiltnis

, Objecte


Section H. Grounds of the possibility of experience <A>

going unity of self-consciousness, i.e., must stand tions of synthesis, namely of the synthesis in accordance concepts, as that in which alone apperception can demonstrate a priori its thoroughgoing and necessary identity. Thus the concept of a cause is nothing other than a synthesis (of that which in the series with other appearances) in accordance concepts; that sort of unity, which has its rule a priori, subjects appearances to itself, thoroughgoing and universal, hence necessary of consciousness would not be encountered in the manifold DerC<~D hons. But these would then belong to no experience, and a quently be without an object,a and would be nothing representations, i.e., less than a dream. All attempts to derive these pure concepts of the understanding from experience and to ascribe to them a merely empirical origin are therefore entirely vain and futile. I will not mention e.g., the of a cause brings the trait of necessity with it, no experience at all can yield, for experience teaches us that one appearance cu:stom2tnly t,ollows another, but not that it must necessarily follow nor ference from a condition to its consequence can be made a entirely universally. But that empirical rule of association, one must assume throughout if one says that everything in series of octhat currences stands under rules according to which nothing is not preceded by something upon which it always follows - on I ask, does this, as a law of nature, rest, and how is this association even possible? The ground of the possibility of the association of the manifold, insofar as it lies in the object,b is called the affinity of the m2lmJlol(j. I ask, therefore, how do you make the thoroughgoing affinity of the appearances (by means of which they stand under constant laws and must belong under them) comprehensible to yourselves? On my principles it is easily comprehensible. All possible appearances belong, as representations, to the whole possible self-consciousness. But from this, as a transcendental representation, numerical identity is inseparable, and certain a priori, because nothing can come into cognition except by means of this original apperception. Now since this identity must necessarily enter into the synthesis of all the manifold of appearances insofar as they are to become empirical cognition, the appearances are thus subject to apriori conditions with which their synthesis (of apprehension) must be in thoroughgoing accord. Now, however, the representation of a universal condition in accordance with which a certain m:mitold whatever kind) can be posited is called a role, and, if it must be so posited, a law. All appearances therefore stand in a thoroughgoing con" Object b Objecte

A I 12

A I 13


Doctrine of Elements. Pt. H. Div. 1. Bk. 1. Ch. II <A>



nection according to necessary laws, and hence in a transcendental affinity, of which the empirical affinity is the mere consequence. That nature should direct itself according to our subjective ground of apperception, indeed in regard to its lawfulness even depend on this, may sound quite contradictorf and strange. But if one considers that this nature is nothing in itself but a sum of appearances, hence not a thing in itself merely a multitude of representations of the mind, then one will not be astonished to see that unity on account of which alone it can be called object a of all possible experience, i.e., nature, solely in the radical faculty of an our cognition, namely, transcendental apperception; and for very reason we can cognize this unity a priori, hence also as necessary, which we would certainly have to abandon if it were given in itself independently of the primary sources of our thinking. For then I would not know whence we should obtain the synthetic propositions of such a universal unity of nature, since in this case one would have to borrow tlIem from the objects of nature itself. But since this could happen only empirically, from that nothing but merely contingent unity could be drawn, which would fall far short of the necessary connection that one has in mind when one speaks of nature. Of the Deduction of the Pure Concepts of the Understanding Third Section On the relation b of the understanding to objects in general and the possibility of cognizing these a priori. \Vhat we have expounded separately and individually in the previous section we will now represent as unified and in connection. The possibility of an experience in general and cognition of its objects rest on three subjective sources of cognition: sense, imagination, and apperception; each of these can be considered empirically, namely in application to given appearances, but they are also elements or foundations a priori that make this empirical use itself possible. Sense represents the appearances empirically in perception, the imagination in association (and reproduction), and apperception in the empirical consciousness of the identity of these reproductive representations with the appearances through which they were given, hence in recognition. But pure intuition (with regard to it as representation, time, the form of inner intuition) grounds the totality of perception a the pure synthesis of the imagination grounds association a priori; and pure apObject






Section In. On the relation of understanding to objects <A> percept ion, i.e., the thoroug hgoing identity of oneself in represe ntation s, ground s empiric al conscio usness a 27 Now if we wish to follow the inner ground of this connec tion of representa tions up to that point in which they must all come in order first to obtain unity of cogniti on for a possibl e eXl)erleUlce, we must begin with pure apperce ption. All intuitio ns are us and do not the least concern us if they cannot be taken into consciousness, whethe r they influen ce it directly or indirect ly, this alone is cogniti on possibl e. 28 W~ ..are cOllsci0lIs of the thorollghgo ing identity of ourselv es with regard to all represe ntation s can ever belong to our cogniti on, as a necessa ry conditi on of possibility of all represe ntation s (since the latter represe nt someth ing in me only insofar as they belong with all the others to one conscio usness, hence they must at least be capable of being connec ted in it). This ciplea holds a priori, and can be called the transce ndenta l princip l,c" of the unity of all the manifo ld of our represe ntation s also in intuition). Now the unity of the manifo ld in a subject is synthet ic; apperce ption therefo re yields a princip lec of tlle synthet ic unity of manifold in all possibl e intuitio n.* This synthet ic unity, howeve r, presupp oses a synthes is, or include s and if the former is to be necessa ry a priori then the latter must also
* One should attend carefully to this proposition, which is of great


All 8

All representations have a necessary relation to a possibl e empirical consciousness: for if they did not have this, and if it were entirely impossible to become conscious of them, that would be as much as to say that they did not exist at all. All empirical consciousness, however, has a necessary relation to a transcendental consciousness (preceding all particular experience), namely the consciousness of myself, as original apperception. It is th\;refore absolute ly necessary that in my cognition all consciousness belong to ohe consciou sness (of myself). Now here is a synthetic unity of the manifold (of consciousness) that is cognized and that yields the ground for synthetic propositions concerning pure thinking in exactly the same way that space and time yield such propositions concerning the form of mere intuition. The syntheti c proposition that every different empirical consciousness must be combine d into a single self-consciousness is the absolutely first and synthetic principl e of our thinking in general. But it should not go unnoticed that the mere representation I in relation to all others (tlle collective unity of which it makes possible) is the transcendental consciousness. Now it does not matter here whether this representation be clear (empirical consciousness) or obscure, even whether it be actual; but the possibility of the logical form of all cognition necessarily rests on the relationship to this apperception as a fuculty.



, Princip Princip


Doctrin e of Elements. Pt.

n. Div. I. Bk. I. Ch. n <A>

Thus the transcendental unity of apperc eption is rea synthesis a of lated to the pure synthesis of the imagination, as an a priori condition But on. the possibility of an compos ition of the nlanifold in a cogniti prionly the produc tive synthe sis of the imagin ation can take place a nce. experie of ons ori; for the reprod uctive synthesis rests on conditi is The principlea of the necessary unity of the pure (productive) synthes posthe of of the imagination prior to apperce ption is thus the ground 29 sibility of all cognition, especially that of experience. Now we call the synthesis of the manifold in imagin ation transcen but nothing s dental if, withou t distinct ion of the intuitio ns, it concern a priori, and the unity of this synthesis the connec tion of the is called transce ndental if it is represe nted as necessary a priori in relathe tion to the original unity of apperception. Now since this latter is of unity ndental ground of the possibility of all cognitions, the transce cognie the synthesis of the imagin ation is the pure form of all possibl tion, throug h which, therefore, all objects of possible experience must

A I 20

be represe nted a priori. The unity of apperc eption in relatio n to the synthe sis of the imagin ation is the unders tandin g, and this very same unity, in relation unto the transce ndenta l synthe sis of the imagination, is the pure derstan ding. In the unders tanding there are therefo re pure a the cogniti ons that contain the necessary unity of the pure synthesis of are r, howeve imagin ation in regard to all possible appearances.3 These, ently consequ ; the catego ries, i.e., pure concepts of the underst anding the empirical power of cogniti on of human beings necessarily contains an underst anding , which is related to all objects of the senses, though only by means of intuitio n, and to their synthesis by means of imagiexnation, under which, therefore, all appearances as data for a possible possible lo perienc e stand. Now since this relation of appearances experience is likewise necessary (since withou t it we could not obtain us any cogniti on at all throug h them, and they would thus not concern catethe at all), it follows that the pure underst anding , by means of b gories, is a formal and synthetic princip le of all experiences, and that appearances have a necess ary relatio n to the unders tandin g. Now we will sct the necessary connec tion of the underst anding with the appearances by means of the categories before our eyes by beginning from beneath, namely with what is empirical. The first thing that is given to us is appearance, which, if it is combin ed with consciousness, is called percept ion (without the relatione to an at least possible conus, sciousness appearance could never become an object of cognition for

, Verhaltni,


Section Ill. On the relation of understanding to objects <A> and would therefore be nothing for us, and since it ality in itself and exists only in cognition it be nothing since every appearance contains a manifold, thus different pe:rce:ptiorls by themselves are encountered dispersed and separate in the a combination of them, which they cannot have in sense itself, is therefore necessary. There is thus an active faculty of the synthesis of manifold in us, which we call imagination, and whose action Immediately upon perceptions I call apprehension.* For the is to bring the manifold of intuition into an image;a it must therefore antecedently take up the impressions into its activity, i.e., them. It is, however, clear that even this apprehension would bring forth no image and no connection of the impressions were there not a subjective ground for calling back a perception, from which the mind has passed on to another, to the succeeding ones, and for exhibiting entire series of perceptions, i.e., a reproductive imagination, which is then also merely empirical. Since, however, if representations reproduced one another wilthclUt distinction, just as they fell together, there would in turn be no determinate connection but merely unruly heaps of them, and no cognition at all would arise, their reproduction must thus have a rule in accordance with which a representation enters into combination in ination with one representation rather than with any others. This subjective and empirical ground of reproduction in accordance with rules is called the association of representations. But now if this unity of association did not also have an objective ground, so that it would be impossible for appearances to be apprehended by the imagination otherwise than under the condition of a possible synthetic unity of this apprehension, then it would also be entirely contingent whether appearances fit into a connection of human cognitions. For even though we had the faculty for associating perceptions, it would still remain in itself entirely undetermined and contingent whether they were also associable; and in case they were not, a multitude of perceptions and even an entire sensibility would be possible in which much empirical consciousness would be encountered in my mind, but separated, and without belonging to one consciousness of
* No psychologist has yet thought that the imagination is a necessary ingredi-

A 121

A 122

A I 20

ent of perception itself. This is so partly because tl1is faculty has been limited to reproduction, and partly because it has been believed that the senses do not merely afford us impressions but also put them together, and produce images of objects, for which without doubt something more than the receptivity of impressions is required, namely a function of the synthesis of them.
" Bild


Doctrin e of Elemen ts. Pt. II. Div. I. Bk. I. Ch. II <A>

because I ascribe all For however, is myself, ption) can I say of apperce l origina (of usness percept ions to one conscio therefo re be an must There them. of us conscio am all percept ions that I to all empira tood unders be can Lhat one i.e., , objective ground indeed even ity, possibil the rests which on ation, imagin icallaw s of the law, namely, a nces, appeara all h throug ng extendi the necessity of a law le in associab are that sense of as hout for regardi ng them throug ion connect hgoing thoroug a of laws al univers to themselves and subject apof tion associa all of ground e objectiv this call in reprodu ction. I exre anywhe this ter encoun never can we But . affinity pearances their unity of apperc eption with regard to all cept in the principle of to me. In accorda nce with this principle belong cogniti ons that are to come into the mind or be apprehended must er all appearances whatev ent with the unity of apperception, agreem in are in such a way that they ic unity in their connection, synthet t withou ible which would be imposs ry. necessa ely which is thus also objectiv The objective unity of all (empirical) consciousness in one conA 12 3 origina l apperception) is thus the necessary condition sciousness appearances (near even of all possible percept ion, and the affinity of the imagination in is synthes a of uence or remote ) is a necessary conseq that is ground ed a priori on rules. The imagin ation is therefo re also a faculty of a synthesis a priori, on accoun t of which we give it the name of productive imagination, and, insofar as its aim in regard to all the manifold of appearance is nothing the further than the necessary unity in their synthesis, this can be called certainly re therefo is It ation. imagin the of transce ndental functio n by strange, yet from what has been said thus far obvious, that it is only the even that ation imagin the of n functio means of this transce ndental lataffinity of appearances, and with it the association and throug h the exently consequ and laws, with nce accorda in ter finally reprodu ction of s concept no them t withou for ; possible perienc e itself, become nce. experie an into e would converg objects at For the standin g and lasting I (of pure apperception) constitutes the to correlate of all of our representations, so far as it is merely possible all-eman to belongs usness conscio all and become conscious of them, bracing pure apperce ption just as all sensible intuitio n as representation belongs to a pure inner intuition, namely that of time. It is this apperA 124 its ception that must be added to the pure imagination in order to make altion, imagina the of is synthes the itself in function intellectual. For s though exercised a priori, is nevertheless always sensible, for it combine triangle. a of shape the e.g., n, intuitio in s the manifold only as it appear u Throug h the relation of the manifold to the unity of apperception,



Section Ill. On the relation of understanding to objects <A>

however, concepts that belong to the understanding can come only by means of the imagination in relation to the sensible mtmtl0Jl. We therefore have a pure imagination, as a fundamental of the human soul, that grounds all cognition a priori. By its means we into combination the manifold of intuition on the one side and dition of the necessary unity of apperception on the extremes, namely sensibility and understanding, must necessarily nected by means of this transcendental function of the Im.agimlti()n, since otherwise the former would to be sure yield appearances but no objects of an empirical cognition, hence there would be no experience. Actual experience, which consists in the apprehension, association (the reproduction), and finally the recognition of the appearances, contains in the last and highest (of the merely elements of experience) concepts that make possible the formal of experiel1cy with it all objective validity (truth) of empirical cognition)These grounds of the recognition of the manifold, so far as concern merely the form of an experience in general, are now those categories/On them is grOlmded, therefore, all formal in sis of the imagination, and by means of the latter also of its el1"lpiJriC<ll use (in recognition, reproduction, association, and appr,ehen:sIo,n) to the appearances, since the latter belong to our consciousness at and hence to ourselves only by means of these elements of cognition. ;rhus we ourselves bring into th~ appearances that order and larity in. them ~at we call nature,yand mor~over we would not be to find It there If we, or the nature of our mmd, had not OfjtgIJrJ.aJJ} it there. For this unity of nature should be a necessary, i.e., certain unity of the connection of appearances. But how should we be to establish a synthetic unity a priori if subjective grounds of such a unity were not contained a priori among the originalsources of cognition in our mind, and if these subjective conditions were not at the same time objectively valid, being the grounds of the possibility of cognizing any object a in experience at all?b
., Object
[, Question mark added. At this point, the following note is inserted in Kant's copy of the first edition: "That the laws of nature really have their origin in the understanding, and are just as litde to be encountered outside it as space and time are, is already proved the in any case already acknowledged assertion that we cognize them and as necessary; for if, on the contrary, they had to be borrowed from outside, we could cognize them as contingent. But then what sort of laws are those? No greater and no less than is necessary in order to bring appearances into a general connection with one conscionsncss, only in order to cognize objects as such - for that is the form of their intuition and at the same time the condition of their unity in apperception given, and given apriori." (E LI, pp. 26-7; 23'26-7) Erdmann observes that this is the only substantial note in Kant's copy of the first-

A 125




Doctrine of Elements. Pt.

n. Div. L Bk. 1. Ch. II <ID

A 127

A 128

We have above explained the unders tanding in various ways - through a spontan eity of cogniti on (in contras t to the receptivity the sensibil of also or ts, concep of faculty a or g, thinkin ity), through a faculty for judgments - which explanations, if one looks at them properly, come of down to the same thing. Now we can characterize it as the faculty essence. its to closer comes and , fruitful more is roles. This designation us Sensibility gives us forms (of intuition), but the underst anding gives of aim the with nces appeara the through rules. It is always busy poring a (and e objectiv are they as far so Rules, them. finding some sort of rule in laws. thus necessarily pertain to the cogniti on of objects) are called particonly are these nce, experie h throug laws Althou gh we learn many ular determ ination s of yet higher laws, the highest of which (under and which all others stand) come from the underst anding itself a priori, appearthe provide must rather but nce, experie are not borrow ed from posances with their lawfulness and by that very means make experience rules making for faculty a merely not thus sible. The underst anding is for throug h the compar ison of the appearances; it is itself the legislation all, at nature any be not would there anding nature, i.e., withou t underst with i.e., synthetic unity of the manifold of appearances in accordance in only exist but us, outside occur rules; for appearances, as such, cannot exan in on cogniti of object the as our sensibility. The latter, however, of perience, with everything it may contain, is possible only in the unity dental transcen the is r, howeve ption, apperce ption. The unity of apperce ce. ground of the necessary lawfulness of all appearances in an experien repof ld manifo a to regard with ption apperce This very same unity of the resentations (that namely of determ ining it out of a single one) is ces appearan All . anding underst the is rules rule, and the faculty of these and as possible experiences, therefore, lie a priori in the understanding, ty sensibili the in lie they as just it, from lity possibi receive their formal their as far as latter the through e possibl only as mere intuitions, and are form is concerned. Thus as exaggerated and contrad ictory as it may sound to say that the the unders tanding is itself the source of the laws of nature, and thus of apand correct eless neverth is n assertio an formal unity of nature, such propria te to t.~e object, namely experience. To be sure, empirical laws, ndas such, can by no means derive their origin from the pure understa cannot nces appeara the of ldness manifo urable ing, just as the immeas . be adequa tely conceived throug h the pure form of sensible intuition pure the of s ination determ lar particu only are But all empirical laws
gave up hope of imedition deduction, from which he infers that Kant in fact very early proving the deductio n by minor changes. ... " in Kant's copy Changed to "Rules, so far as they [represent] existence as necessary of the first edition (E LIl, p. 27; 23:46).


Section Ill. On the relation of understa nding to objects <Pt>

laws of the underst anding , under which and in accordance norm they are first possible, and the appearances assume a just as, regardless of the variety of empirical form, all ap]?e~lra:nC("s must nevertheless always be in accord with the pure form The pure unders tanding is thus in the categories the law of thetic unity of appearances, and thereby first and originally experience possible as far as its form is concern ed. But we not to accomplish more in the transce ndental deduct ion of the categories than to make compre hensibl e this relationa of the to sensibility and by means of the latter to all objects of experience, to make compre hensibl e the objective validity ofits pure concepts, and thereby determ ine their origin and truth. Summ ary represe ntation of the correct ness and unique possibi lity deduct ion of the pure concep ts of the unders tanding . If the objects with which our cogniti on has to do were things in themselves, then we would not be able to have any a priori concep ts of at all. For whence should we obtain them? If we take ject b (without even investig ating here how the latter known to us), then our concep ts would be merely emlpiJricll and not a concepts. If we take them from ourselves, then that which is merely in us cannot determ ine the constit ution of an object distinc t from our represe ntation s, i.e., be a ground why there should be a that corresponds to someth ing we have in our though ts, and this representation should not instead be empty. But if, on the rn"tr:lr v we have to do everywhere only with appearances, it is not possible but also necessary that certain a priori concep ts precede empirical cogniti on of objects. For as appearances they constit ute an object that is merely in us, since a mere modific ation of our sensibil ity is not to be encoun tered outside us at all. Now even represe ntation - that all these appearances and thus all objects with which we can occupy ourselves are all in me, i.e., determ ination s my self- expresses a thorou ghgoin g unity of them in one and the same apperception as necessary. The form of all cogniti on of objects (throug h which the manifold is though t as belong ing to one object),c howeve r, also consists in this unity of possible consciousness. Thus the way in which the manifold of sensible represe ntation (intuiti on) belongs to a
" Verbiiltnis



, zu Eincm Object


Doctrin e of Elemen ts. Pt. H. Div. I. Bk. 1. Ch. II <A>



as its intellectual cogmtl 0n consciousness precede s objects on of cogniti formal a an itself constit utes form, is synthes Their ries). (catego are they in general, insofar as relain s ntation represe all of unity the ation, imagin throug h the pure tion to origina l apperce ption, preced e all empirical cogniti on. Pure necessary concep ts of the unders tanding are therefo re possible, has to do on cogniti our because nce, experie a priori in relation to s, ourselve in lies lity possibi nces, with nothing but appeara enis object) an of ntation represe the whose connec tion and thus must precede all experience and first counte red merely in us, from this ground, is concern ed. form its as make it possible as categories has the of ion deduct our among the only possible one been conduc ted.


Ofthe Deduction ofthe Pure the Understanding Second Section Transcendental deduction concepts ofthe understanding a'32

B 129

< 15 On the possibility of a combination in general.

The manifold of representations can be given in an intuition that is merely sensible, i.e., nothing but receptivity, and the form of intuition can lie a p1'iori in our faculty of representation being anything other than the way in which the subject is affected. Yet combination (conjunctio) of a manifold in general can never come to us through the senses, and therefore cannot already be contained in the pure form of sensible intuition; for it is an act b of the of the power of representation, and, since one must call latter unde'rsltarlding, in distinction from sensibility, ~ll combination, we are conscious of it or not, whether it is a combination of the m:Il11loJd tuition or of several concepts, and in the first case non-sensible intuition, is an action of the we would designate with the general title synthesis in order at the same til,Ile to draw attention to the fact that we can represent nothing as combined in the object C without having previously combined it ourselves, and that among all representations combination is the one that is Qot given through objects d but can be executed the it,df, since it is an act e of its self-activity. One can here easily see this action must originally be unitary! and equally valid for all COlmt1ln:ltH)n,
" In the second edition, the following IS through 27 replace the second and third sections of the "Transcendental Deduction" in the first edition (A95 to A130). b Actus Object d Objecte , Actus in modern German this is used only in idioms connoting being in agreement or harmony; perhaps Kant meant to write einzig, i.e., unique.

B 130


Doctrine of Elements. Pt. n. Div. 1. Bk. 1. Ch. n. <B> seems to be its opposite, in fact and that the dissolution (analysis) sly always presupposes it; for where the unders tanding has not previou h throug only for g, anythin combin ed anything, neither can it dissolve as ntation represe of power the to given it can someth ing have been combin ed. But in additio n to the concep t of the manifo ld and of its synthesis, the of concep t of combin ation also carries with it the concep t of the unity unity tic synthe of the manifold. Combi nation is the represe ntation e, of the manifold. * The represe ntation of this unity cannot, therefor ntar.eprese to added being by the combin ation; arise cion of the manifold, it first makes the concep t of combin ation possiconcep ts of combin ation a priori, is ble. 33 This unity, which precede s d not the former categor y of unity ( IQ); for all categories are grounde the thus ation, combin these in but on logical functions in judgme nts, alunity of given concep ts, is already though t. The categor y therefore (as unity this scek re therefo ready presupposes combin ation. We must contains itself which that in qualitative, 12) someplace higher, namely the ground of the unity of different concep ts in judgme nts, and hence of the possibility of the underst anding , even in its logical use.



On the origina l-synth etic unity of apperc eption. for The I think must be able to accomp any all my representations; be otherw ise someth ing would be represe nted in m.e that could' not would ntation represe the that say to as much as is tJ.~ought at all, which repeither be impossible or else at least would be not..~ing for me. That n. intuitio called is g thinkin all to prior given be can resenta tion that in Thus an manifold of intuitio n has a necessary relation to the Lthink this the same subject in which this manifo ld is to be encoun tered. But as d regarde be cannot it i.e., neity, sponta of acta an is represe ntation disbelong ing to sensibility. I call it the pure apperc eption , in order to eprion, lapperc origina the also or one, cal empiri tinguis h it from the since it is that self-consciousness which, because it produc es the representatio n I think, which must be able to accomp any allot:herL:iud therefore * vVhether the representations themselves are identical, and whether

B 132

B 131

one could be thought through the other analytically, does not come into consideration here. The consciousness of the one, as far as the manifold is con. cerned, is still always to be distinguished from the consciousness of the ot,~er, here.' issue at is that sness and it is only the synthesis of this (possible) consciou


Section n. Trans. deduction of pure concepts of understanding <B> consciousness is one the same, cannot by any further representation. I also call its the trans;ceJl:u1~:ntal .unity of self-consciousness in order to designate the possibility of a priori cogniti!?Jn from it. For the manifold representations that are in a certain intuition would not together be my representations if did not all together belong to a self-consciousness; i.e., as my representations (even if I am not consciousof as such) must necessarily be in accord with the condition under which can stand together in a universal self-consciousness, because otherwise would not throughout belong to me. From this original combination much may be inferred. Namely, this thoroughgoing identity of the apperception a manifold given in intuition contains a synthesis of the representations, and is possible only through the consciousness of this synthesis. For empirical consciousness that accompanies different representations is itself dispersed and without relation to the identity of the subject. The latter relation therefore does not yet come about my accompanying each representation with consciousness, but rather my adding one representation to the other and being conscious of their synthesis. Therefore it is only because I can combine a of given representations in one consciousness that it is possible for me to represent theidentity of the conscic::fusness in these representations itself, i.e., the analytical unity of apperception is only possible under the presup~ position of some synthetic one. *,34 The thought that these representabOnS given in intuition all together belong to me means, accordingly, the same as that I unite them in a self-consciousness, or at least can unite them therein, and although it is itself not yet the consciousness of the synthesis of the representations, it still presupposes the possibility QLthe latter, i.e., only because I can comprehend in a consciousness do I call them all together my representations; for wise I would have as multicolored, diverse a self as I have representa* The analytical unity of consciousness pertains to all common concepts as such, e.g., if I think of red in general, I thereby represent to myself a feature that (as amark) can be encountered in anything, or that can be combined with other representations; therefore only by means of an antecedently conceived possible synthetic unity can I represent to myself the analytical A representation that is to be thought of as common to several must be regarded as belonging to those that in addition to it also have something different in themselves; consequently they must antecedently be conceived in synthetic unity with other (even if only possible representations) before I can think of the analytical unity of consciousness in it that makes it into a conceptlLl" nis. And thus the synthetic unity of apperception is the highest to which one must affix all use of the understanding, even the whole of logic after it, transcendental philosophy; indeed this faculty is the understanding itself.
w~ich in


B 134



Doctrine of Elements. Pt. II. Div. I. Bk. I. Ch. H. <B> intions of which I am conscious. Synthe tic unity of the manifo ld of apof identity the ground the thus is tuitions , as given a . percep tion itself, which precede s a priori all my determ inate thinking were it as cannot and r, howeve , objects the in lie not Combi nation does first be borrow ed from them throug h percep tion and by that means the of on operati an only rather is but , anding taken up into the underst B 135 comof faculty the than further nothing itself underst anding , which is bining a priori and bringin g the manifold of given represe ntation s under unity of apperce ption, which princip le is the suprem e one in the whole of human cognition.3 5 Now this princip le of the necessary unity of apperce ption is, to be as sure, itself identical, thus an analytical propos ition, yet it declares without n, intuitio an in given ld manifo the necessary a synthesis of be which that thoroug hgoing identity of self-consciousness could not maninothing , ntation represe simple a as I, though t. For throug h the fold is given; it can only be given in the intuitio n, which is distinct from it, and though t throug h combi nation in a consciousness. An understandin g, in which throug h self-consciousness all of the manifold would at the same time be given, would intuit; ours can only think and must seek the intuitio n in the senses. I am therefo re conscious of the identical self in regard to the manifold of the represe ntation s that are given tato me in an intuitio n because I call them all togethe r my represen conam I that say to as much as is that But tions, which constit ute one. scious a priori of their necessary synthesis, which is called the original synthet ic unity of apperce ption, under which all represe ntation s given of a to me stand, but under which they must also be brough t by means B 136 synthesis.

17 the synthe tic unity of apperc eption The princip le is the suprem e princip le of all use of the unders tanding . intuitio n in relation to J_Jle..sllI)reme princip le of the possibility of the sells1blllty was, accordi ng to the Transc endenta l Aesthetic, that all and ~pace 1I1:a!11tqJLd Qf sensibility stand under the formal conditionsQt cr time. The suprem e priQciple of all intuitio n in relation to the~;nd of ns condjtiQ standin g is that all the manifold of intuitio n stand under 6 All the manifo ld repre~~e origina l synthet ic unity of apperce ption. *,3
a* Space and time and all their parts are intuitions, thus individual represent the (see es themselv in contain they along with the manifold that tions Transcendental Aesthetic), thus they are not mere concepts by means ofwhich the same consciousness is contained in many representations, but rather are it; many representations that are contained in one andjn the consciousness of conof unity the ently consequ and they are thus found to be composite,

Section II. Trans. deduction of pure concepts sentations of intuition stand

un,derstandilng <B>

first pnmc:lpJle second insofar as consciousness; without that th()U~~ht or cognized through them, since the given rej)re:seJtlt2ltic)lls would not have in common the acta of I and not grasped together in a self-consciousness. thereby Understanding is, generally speaking, the of cogniti~:>n:s. These consist in the determinate relation of given to an object. b An object,c however, is that in the concept fold of a given intuition is unitedY Now, howc:ver. representations requires unity of consciousness them. Consequently the unity of consciousness is stitues the relation of representations to an object, thus their obJec:tn'e validity, consequently is makes into cognitions on which even the possibility of understanding rests. The first pure cognition of th!( understanding, thcTe:tore, the whole of the rest of its use is grounded, and tllat is at the same time also entirely independent from all conditions of sensible is the principle of the original synthetic unity of the mere form of outer sensible intuition, space, is not yet cog-n.luon it only gives the manifold of intuition a priori a Fv,~'nlH"" COJ~l1JltlC)ll. But in order to cognize something in space, e.g., a I must and thus synthetically bring about a determinate combination of the given manifold, so that the of this action is at tlle same time the unity of consciousness (in the concept of a line), ject d (a determinate space) first cognized. is therefore an objective condition of order to cognize rather s()l11etl1ing under which every intuition must ~~c()JJ1e,an obj~(:tfor_!1!e, since in any other way, synthesis, the manifold would not be united in one consciousness. This last proposition is, as we said, itself analytic, to be sure, it makes synthetic unity into the condition of all thinking; for it says nothing more than that all my representations in any given intuition must stand under the condition under which alone I can ascribe sciousness, as synthetic and yet as original, is to be found in them. This singularity of theirs is important in its application (see 25).
, Actus b Object , Object d Object , Object f Object




Doctrine of Elements. Pt. n. Div. 1. Bk. 1. Ch. n. <B> them to the identic al self as my represe ntation s, and thus can grasp them togethe r, as synthetically combin ed in an apperce ption, through the general expression I think. a This principle, however, is not a princip le for every possible underapperce ption in the reppure standin g, but only for one throug h whose alL That understanding at given resenta tion I am nothing manifold is at ld throug h whose self-consciousness the manifo of intuitio n would ntareprese whose h throug the same time be given, an unders tanding tion the objectsb of this represe ntation would at the same time exist, the would not require a special acre of the synthesis of the manifo ld for merely which , anding underst human unity of consciousness, which the thinks, but does not intuit, does require . But for the human understandin g it is unavoidably the first principle, so that the human understandin g cannot even form for itself the least concep t of another possible underst anding , either one that would intuit itself or one that, t while possessing a sensible intuitio n, would possess one of a differen kind than one ground ed in space and time.


18. Vilhat objecti ve unity of self-co nsciolis ness is.

The transc endent al unity of apperc eption is that unity through of which all of the manifo ld given in an intuitio n is united in a concept d It is called objecti ve on that accoun t, and must be distinthe object. guished from the subjec tive unity of consciousness, which is a detern is minati on of inner sense, throug h which that manifo ld of intuitio become can I er \Vheth ation. combin a such for empiric ally given e empiri cally conscious of the manifo ld as simulta neous or successiv emthe Hence ons. conditi al empiric or , stances circum depend s on the tapirical unity of consciousness, throug h association of the represen tions, itself concer ns an appearance, and is entirely contingent. The n in pure form of intuitio n in time, on the contrary, merely as intuitio general , which contain s a given manifo ld, stands under the original the unity of consciousness, solely by means of the necessary relation of synpure the h throug thus think, I one the to n manifo ld of intuitio thesis of the unders tanding , which ground s a priori the empirical synof thesis. That unity alone is objectively valid; the empirical unity deapperce ption, which we are not assessing here, and which is also has concreto, in ons conditi given under former, the rived only from

B 140

Princip Objecte CActus d Object



Section n. Trans. deduction of pure concepts of understanding <B> merely subjective validity. One person combines a certain word with one thing, another something else; unity of consciousness in that which is empirical is not, with regard to that which is given, necessarily and universally valid.

19 The logical form all judgments consists in objective unity of the apperception of the concepts contained therein.3 8
I have never been able to satisfY myself with the explanation that the gicians give of a judgment in general: it is, they say, representation of a relationa between two concepts. Withput quarreling what is mistaken in this explanation, that in any case it fits gorical but not hypothetical and disjunctive judgments two do not contain a relationb of concepts but of judgments themselves) (though from this error in logicl many troublesome consequences arisen),*,39 I remark only that it is not here determined wherein this relatione consists. If, however, I investigate more closely the relationd of given cognitions in every judgment, and distinguish relation, as something belonging to the understanding, from the relatione in accordance with laws of the reproductive imagination (which has subjective validity), then I find that a judgment is nothing other than the to given cognitions to the objective unity of apperception. 40 aim of the copulafis in them: to distinguish the objective representations from the subjective. For this word designates tion of the representations to the original apperception and its necessary unity, even if the judgment itself is hence contingent,
* The widespread

B 141

B 142

doctrine of the four syllogistic figures concerns only the categorical inferences, and, although it is nothing more than an art for surreptitiously producing the illusion of more kinds of inference than that in the first figure by hiding immediate inferences (consequentiae among the premises of a pure syllogism, still it would not have achieved any special success by this alone if it had not succeeded in focusing attention exclusively on categorical judgments as those to which all others have to be related, which according to 9, however, is false.


, Ve7'hiiltnisses b Verhiiltni., , Verhiiltnis d Here Kant uses Bez:ich:'I71f[ when he might have used Verhiiltni.>. , Vcrhii!tniFe f Vcrhiiltni.cwiirtcbc71


Doctrine of Elements. Pt. II. Div. 1. Bk. I. Ch. II. <B> not mean to say that to be sure, I e.g., "Bodies are heavy." these represe ntation s necess arily belong to one anothe r in the empirto one anothe r in virtue of ical intuitio n, but rather that they ns, the necess ary unity of the apperce ption in the synthesis of intuitio all of ination determ e principles a of the objectiv i.e., in accordance prinwhich them, represe ntation s insofar as cogniti on can come from the principle of the transce ndental unity of derived ciples b are in this way does there arise from this relation' a apperce ption. judgtn ent, i.e., a relation that is objecti vely valid, and that is suffis in ciently distinguished from the relation of these same representation with nce accorda in subjective validity, e.g., be which there the latter I could only say "If I nce accorda In tion. associa of laws is carry a body, I feel a pressure of weight," but not "It, the body, comare s ntation heavy," which would be to say that these two represe of bined in the object,d i.e., regardless of any difference in the condition r (howeve ion the subject, and are not merely found togethe r in percept often as that might be repeated).


sensibl e intuitio ns stand under the categor ies, as conditi ons under which alone their manifo ld can come togeth er in one conscio usness. The manifold that is given in a sensible intuitio n necessarily belongs this under the original synthetic unity of apperce ption, since through unthe of action That 17). ( e possibl alone is the unity of the intuitio n tarepresen given of ld manifo the derstanding, however, throug h which an under t brough is ts) concep or ns tions (wheth er they be intuitio 19). ( nts judgme of n functio logical apperc eption in general, is the e , Theref ore all manifold, insofar as it is given in one empirical intuition t, judgmen for ns functio logical the of is determ ined in regard to one by means of which, namely, it is brough t to a consciousness in generaL for But now the catego ries are nothing other than these very functions with ined determ is n intuitio given a judging, insofar as the manifold of regard to them ( 13).4' Thus the manifold in a given intuitio n also necessarily stands under categories.

translate further oc, Verhiiltnisse; the further occurren ces of "relation " in this sentence currence s of Verhiiltnis.

instead of merely , Einer. Not ordinaril y capitalized, suggesting the translatio n "one"
"an. n



Section n. Trans. deduction of pure concepts of understanding <B>

Remark. A manifold that is contained in an intuition that I call mine is represented as belonging to the necessary unity of self-consciousness through the synthesis of the understanding, this takes means of the category.* This indicates, therefore, that consciousness of a given manifold of one a intuition stands under a a priori self-consciousness, just as empirical intuitions stand pure sensible one, which likewise holds a priori. - In the above pn)pC)Slnon, therefore, the beginning of a deduction of the pure concepts categories arise the understanding has been made, in which, since independently from sensibility merely in the understanding, I must abstract from the way in which the manifold for an empirical intuition is given, in order to attend only to the unity that is added to the intuition through the understanding bfmeans of the category. In the sequel ( 26) it will be shown from the way in which the intuition is given in sensibility that its unity can be none other than the one category prescribes to the manifold of a given intuition in general according to the preceding 20; thus by the explanation of its b in regard to all objects of our senses the aim of the deduction be fully attained. In the above proof, however, I still could not abstract from one namely, from the fact that the manifold for intuition must . .>1,.,0""'1,, given prior to the synthesis of understanding and Inc[epen(1el1ltly it; how, however, is here left undetermined. For if I an understanding that itself intuited (as, say, a divine understanding, which would not represent given objects, but whose representation the objects would themselves at the same time be given, or produced), then the categories would have no significance at all regard to such a cognition. They are only rules for an understanding whose entire capacityc consists in thinking, i.e., in the action of bringing synthesis of the manifold that is given to it in intuition from to the unity of apperception, which therefore cognizes at all



* The ground of proof rests on the represented unity of intuition


which an object is given, which always includes a synthesis of the manifold that is given for an intuition, and already contains the relation of the latter to unity of apperception.
" Eimr, again capitalized.

The antecedent is probably "the category" in the preceding clause, but it could also be "the unity," and thus the translation has been left ambiguous.

;" VC7771rigC77


Doctrine of Elements. Pt. n. Div. 1. Bk. I. Ch. n. <B> orders the materia l for cognition, the inonly combines itself, a object. b But for the petuition, which must be given to it throug h of culiarity of our underst anding , that it is able to bring about the unity through only and ies categor the of means by apperc eption a priori ofprecisely this kind and numbe r of them, a further ground may be these ly precise have we why for offered fered just as little as one can be the and no other functions for judgme nt or for why space and time are n. sole forms of our possible intuitio



The catego ry has no other use for the cogniti on of things than its applica tion to objects of experie nce. To think an object and to cogniz e an object are thus not the same. For two compo nents belong to cognition: first, the concept, through which an object is though t at all (the category), and second, the intuthe ition, throug h which it is given; for if an intuitio n corresp onding to then it would be a though t as far as its concep t could not be given at form is concerned, but withou t any object, and by its means no cognition of anythin g at all would be possible, since, as far as I would know, nothin g would be given nor could be given to which my though t could be applied. Now all intuitio n that is possible for us is sensible (Aesthetic), thus for us thinkin g of an object in general throug h a pure conthis cept of the unders tanding can become cogniti on only insofar as either concep t is related to objects of the senses. Sensible intuitio n is pure intuitio n (space and time) or empirical intuitio n of that which, B 147 throug h sensation, is immed iately represe nted as real in space and time. ns Throug h determ ination of the former we can acquire a priori cognitio ed, of objects (in maL~ematics), but only as far as their form is concern as appearances; whethe r there can be things that must be intuited in this are form is still left unsettle d. Conseq uently all mathematical concepts not by themselves cognitions, except insofar as one presupposes that the there are things that can be present ed to us only in accordance with form of that pure sensible intuitio n. Thing s in space and time, howacever, are only given insofar as they are percept ions (representations tation. compan ied with sensation), hence throug h empirical represen are The pure concepts of the underst anding , consequently, even if they only applied to a priori intuitio ns (as in mathematics), provide cognition insofar as these a priori intuitions, and by means of them also the concepts of the underst anding , can be applied to empirical intuitions. ConfUr sich



Section H. Trans. deduction of pure concepts of understanding <B> sequently the ('~tegories do not afford us cognition of things means mthl'ough their possible application to for the possibility of empirical cngniti()n. This, however, is called experience. The qtegories consequently have no otheI" use for the cognition of things except insofar as these are taken as objects of possible experience.

B 148

The above proposition is of the greatest importance, for it determines the bollndaries of the use of the pure concepts in ~egard to objects, just as the Transcendental Aesthetic de1tennirled boundaries of the use of the pure form of our sensible intuition. Space and time are valid, as conditions of the possibility objects can given to us, no further than for objects of the senses, hence only for experience. Beyond these boundaries( they do not represent at all, for they are only in the senses and outside of them have no The pure concepts of the understanding are free from this lurutatron and extend to objects of intuition in general, whether the latter be similar to our own or not, as long as it is sensible and not intellectual. But this further extension of concepts beyond our sensible intuition does not get us anywhere. For they are then merely empty concepts of jects,a through which we cannot even judge whether the latter are possible or not - mere forms of tl10ught without objective reality - since we h:lyeavailable no inwition to the synthetic unity of ception, which they alone contain, could be applied, and 'detEimine an object. Our sensible and empirical intuition alone can provide them with sense and significance. Thus if one assumes an object b of a non-sensible)ntuition as given, one can certainly represent it through all of the predicates that already lie in the presupposition that nothing belonging to sensible intuition pertains to it: thus it is not extended, or in space, that its is not a time, that no alteration (sequence of determinations in time) is to be encountered in it, etc. But it is not yet a genuine cognition if I indicate what the intuition of the object C is not, without being to say what is then contained in it; for then I have not represented the possibility of an object d for my pure concept of the understanding at since I cannot give any intuition that would correspond to it, but could only say that ours is not valid for it. But what is most important here is that not even a single category could be applied to such a thing, e.g., the
, Ohjecten Object ( Object d Objects



Doctrine of Elements. Pt. II. Div. T. Bk. 1. Ch. II. <B> concept of a substance, i.e., that of something that could exist as a subject but never as a mere predicate; for I would not even know whether there could be anything that corresponded to this determination of thought if empirical intuition did not give me the case for its application. But more of this in the sequel.



On the application of the categories to objects of the senses in general. The pure concepts of the understanding are related through the mere understanding to objects of intuition in general, without it being determined whether this intuition is our own or some other but still sensible one, but they are on this account mere forms of thought, through which no determinate object is yet cognized. The synthesis or combination of the manifold in them was related merely to the unity of apperception, and was thereby the ground of the possibility of cognition a priori insofar as it rests on the understanding, and was therefore not only transcendental but also merely purely intellectual. But since in us a certain form of sensible intuition a priori is fundamental, which rests on the receptivity of the capacity for representation (sensibility), the understanding, as spontaneity, can determine the manifold of given representations in accord with the synthetic unity of apperception, and thus think a priori synthetic unity of the apperception of the manifold of sensible intuition, as the condition under which all objects of our (human) intuition must necessarily stand, through which then the categories, as mere forms of thought, acquire objective reality, i.e., application to objects that can be given to us in intuition, but only as appearances; for of these alone are we capable of intuition a priori. This synthesis of the manifold of sensible intuition, which is possible and necessary a priori, can be called figurative (synthesis speciosa), as distinct from that which would be thought in the mere category in regard to the manifold of an intuition in general, and which is called combination of the understanding (synthesis intellectualis); both are transcendental, not merely because they themselves proceed a priori but also because they ground the possibility of other cognition a priori. Yet the figurative synthesis, if it pertains merely to the original synthetic unity of apperception, i.e., this transcendental unity, which is thought in the categories, must be called, as distinct from the merely intellectual combination, the transcendental synthesis of the imagination. Imagination a is the faculty for representing an object even without its presence in intuition. Now since all of our intuition is sensible,



Here Kant uses both large type and spacing for extra emphasis.


Section n. Trans. deduction of pure concepts of understanding <B> the imagination, on account subjective COlldltlon alone it can give a corresponding intuition to the COllcepts standing, belongs to sensibility; but insofar as its synthesis is still an exercise of spontaneity, which is determining and not, like sense, determinable, and can thus determine the form sense a cordance with the unity of apperception, is to extent a faculty for determining the sensibility a priori, its synthesis of intuitions, in accordance with the categories, must be the transcendental synthesis of the imagination, which is an effect of the understanding on sensibility its first application at the same time others) to objects of the intuition IS POSSIIDle the ground of us. As figurative, it is distinct from the synthesis wi1thclUt any imagination merely through the understanding. Now insofar as the imagination is spontaneity, I also ~ccasionaIly call it imagination, and thereby distinguish it from the rel)rc~diJlctJtve nation, whose synthesis is subject solely to empirical ,,.,.,.,,p>!,,th"Qp of association, and that therefore contributes nothing to the eXl)lanatlon not of the possibility of cognition a priori, and on that account in transcendental philosophy but in psychology.


Here is now the place to make intelligible the that must have struck everyone in the exposition of the form of inner sense ( namely how this presents even ourselves to consciousness as we appear to ourselves, not as we are in ourselves, since we intuit ourselves only as we are internally affected, which seems to be cOlltradlctclry, since we would have to relate to ourselves passively; for this reason it is customary in the systems of psychology to treat sense as the same as the faculty of apperception (which we distinguish)Y That which determines the inner sense is the understanding its original faculty of combining the manifold of i.e., of ing it under an apperception (as that on which its very possibility rests). Now since in us humans the understanding is not itself a faculty of intuitions, and even if these were given in sensibility cannot take them up into itself, in order as it were to combine the manifold of its own intuition, thus its synthesis, considered in itself"a alone, is nothing other than the unity of the action of which it is conscious as such even out sensibility, but through which it is capable of itself determining sensibility internally with regard to the manifold that may be given to it in accordance with the form of its intuition. Under the designation of a transcendental synthesis of the imagination, it therefore exer, fUr sich



Doctrin e of Elemen ts. Pt. II. Div. 1. Bk. I. Ch. II. <B>

B 154

cises that action on the passive subject , whose faculty it is, about which we rightly say that the inner sense is thereby affected. Apperthe ception and its synthet ic unity is so far from being the same as ation, combin all of source inner sense that the former, rather, as the sensible intuitio n of objects a in general , to the manifold applies to inner of intuiti ons in genera l, under the name of the categor ies; withbut n, sense, on the contrary, contain s the mere form of intuitio and thus it does not yet contain out combin ation of the manifo ld in throug h the which is possible any determ inate intuitio n at h the tranthroug ld conscio usness of the determ ination of the manifo unce scende ntal action of the imagin ation (synthe tic influen of the e figurativ the derstan ding on the inner sense), which I have named

synthesis. We also always perceiv e this in ourselves. We cannot think of a line t dewithou t drawin g it in though t, we cannot think of a circle withou all at space of scribin g it, we cannot represe nt the three dimens ions same the at withou t placing three lines perpen dicular to each other g a point, and we cannot even represe nt time withou t, in drawin of ntation straigh t line (which is to be the externa l figurative represe d manifol time), attendi ng merely to the action of the synthesis of the thereby and throug h which we successively determ ine the inner sense, attendi ng to the succession of tllis determ ination in inner sense. *.b Motion , as action of the subject (not as determ ination of an object), B I 55 from t abstrac we if conseq uently the synthesis of the manifo ld in space, with this manifo ld in space and attend solely to the action in accordance conthe es which we determ ine the form of inner sense, first produc find not does cept of succession at all. The unders tanding therefo re but sense, inner some sort of cornhin ation of the manifo ld already in to is think I that I produc es it, by affecti ng inner sense. But how the inof kinds other nt differ from the I that intuits itself (for I can represe the tuition as at least possible) and yet be identic al with the latter as thinkand ence same subject, how therefo re I can say that I as intellig
thus also not * Motion of an object" in space does not belong in a pure science, but a d cognize be cannot in geometry; for that somethi ng is movable acri pure a is space, a of tion descrip as motion, But only through experience. general of the successive synthesis of the manifol d in outer intuition in even but y geometr to through product ive imagination, and belongs not only to transcen dental philosophy.


Objecte Objects Object Actus


Section n. Trans. deductio n of pure concepts of underst anding <B>

ing subject cognize my self as an object a that is though t, as am also given to myself in intuitio n, only, like other phenom ena, not as I am for the unders tanding but rather as I appear to myself, is no more and no less difficult than how I can be an object b for m general and indeed one of intuitio n inner perceptions. But it really must be so can be dearly shown, if one lets space count as a mere pure form of the appearances of outer sense, fact although it is not itself an object of outer intuitio n at cannot be made representable to us except under the image of a line, insofar as we draw it, without which sort of present ation we could not its measure at all, or likewise from the fact that we must the determ ination of the lengu~ of time or also of the positions in time for all inner percept ions from that which presents external things to us as alterable; hence we must order t~ determ ination s of inner sense as appearances in time in just the same way as we order those outer sense in space; hence if we admit about the latter that we cognize jects C by their means only insofar as we are externally affected, then we must also concede that through inner sense we intuit ourselves as we are interna lly affected our selves, i.e., as far as inner intuitio n is concerned we cognize our own subject only as appearance not in accordance with what it is in itself.*,43



In the transcendental synthesis of the manifold represe ntation s in general, on the contrary, hence in the synthetic original unity of apperc~ption, I am conscious of myself not as I appear to myself, nor as am in myself, but only that I am. This repres entatio n is a thinkin g, not an intuitin g. Now since the cogmt ion of ourselves, in addition to the action of thinkin g that brings the manifold of every possible intuition to the unity of apperception, a determ inate sort of int:uit:io ][l, through which this manifold is given, is also required, my own existen ce
* I do not see how one can find so many difficulties in the fact that inner sense

is affected by ourselves. Every aetd of attentio n can give us an example of this. In such acts the underst anding always determi nes the inner sense, in accordance with the combin ation that it thinks, to the inner intuitio n that corresponds to the manifold in the synthesis of the understa nding. How much the mind is common ly affected by this means, everyone will be able to perceive in himself.

156 157

" Object b Object , Objecte d Actus


Doctrin e of Elemen ts. Pt. H. Div. I. Bk. 1. Ch.

n. <B>

the determinamere illusion), is not indeed appearance the form occur in corresp ondenc e existence*,44 can tion B 158 d manifol the particu lar way in which of inner sense, accordi ng to cogno have and I therefo re that I combin e is given in inner appear to myself. I as but am, I as myself of nition cognitionoLoneself, from re t..~erefo is oneself of ess sciousn constit ute the thinkin g of an object" regardless of the categories in genera l throug h combin ation of the manifold in an apperception. as for the cogniti on of an objectb distinc t from me 1also need an C catintuitio n additio n to the thinkin g of an object in genera l (in the cogthe for I determ ine that general concept, so egory), throug h the consciousllt.:ss, or in I also need which I think myself, an intuitio n of the manIfold in I determ ine this though t; and I exist as an intelliconscious of its faculty for combin ation but which, gence that is in regard to u~e manifo ld that it is to combine, is subject to a limiting B 159 combination inconditi on that it calls inner sense, which can make that d lie entirely tuitable only in accordance with tempor al relations can therethat and proper, tanding unders the of ts concep outside of the to an inregard with itself to appears it as merely itself fore still cognize the tuition (which is not intellectual and capable of being given through were unders tanding itself), not as it would cognize itself if its intuitio n intellectuaL
existence is * The I think expresses the act" of determi ning my existence. The the mani.e., it, ne thereby already given, but the way in which I am to determi given. thereby yet not is it, to ng ifold that I am to posit in myself as belongi form, given an in d grounde is which , required is For that self-intuition able. determin the of ity receptiv the to belongs and sensible is which i.e., time, deterthe give would Now I do not have yet another self-intuition, which before mining in me, of the spontan eity of which alone I am conscious, even to be is which that gives time as way same the in , the act! of determination e self-activ a of that as e existenc my ne determi cannot I thus determi ned, dethe of i.e., , thought being, rather I merely represen t the spontan eity of my able, i.e., terminin g, and my existence always remains only sensibly determin the reais eity spontan this Yet nce. appeara an of e determi nable as the existenc ence. intellig an myself son I call



Object Objects , Object d Zeitverhiiltl11~'S(l1

, Actus fActus


Section n. Trans. deduction of pure concepts of understanding <B>

Transc endent al deduct ion of the use of the pure concep ts of the unders tanding possibl e expene nce.

In the metaph ysical deduct ion45 the origin of the categor ies in general was establis hed throug h their comple te coincid ence with the universal logical functio ns of thinkin g, in the trallls:c endenta l dedlJ!ccion, however, their possibi lity as a priori cogniti ons of objects of an intuition in general was exhibit ed ( 20, 2 Now the possibi lity cognizing a priori throug h catego ries whatev er objects may come fore our senses , not as far as the form of their intuitio n but far as the laws of their combin ation are concern ed, thus the possibi lity of as it were prescri bing the law to nature and even making possible, is to be explained. For if the categor ies not serve in way, it would not become clear why everyth ing may ever come before our senses must stand under the laws that arise a p1'iori from understanding alone. First of all I remark that by the synthe sis of appreh ension I stand the compos ition of the ma.uifold in an empiric al through which percept ion, i.e., empiric al conscio usness of it (as appearance), become s possible. We have forms of outer as well as inner sensibl e intuitio n a priori in the represe ntation s of space and time, and the synthes is of the apprehension of the manifo ld of appeara nce must always be in agreem ent with the latter, since it can only occur in accorda nce u~is But space and time are represe nted a priori not merely as forms of serlsil)! e intuition, but also as intuitio ns themse lves (which contain a m,mltoJ d), and thus with the determ ination of the unity of manifo ld in (see the Transce ndental Aesthetic).*,46 Thus even unity of the
* Space, represented as object (as is really required in geometry), contains more

B r60

B 16r

than the mere form of intuition, namely the compre hensiona of the manifold given in accordance with the form of sensibility in an intuitiv e representation, so that the form of intuitio n merely gives the manifold, but the formal intuition gives unity of the representation. In the Aesthetic I ascribed this unity merely to sensibility, only in order to note that it precedes all concept s, though to be sure it presupposes a synthesis, which does not belong to the senses but through which all concepts of space and time first become possible . For since through it (as the understanding determines the sensibility) space or time are first given as intuitions, the unity of this a priori intuition belongs to space and time, and not to the concept of the understanding ( 24)'




Doctrine of Elements. Pt. H. Div. 1. Bk. I. Ch. H. <B> sis of the manifold, outside or us, hence also a combination with which everything is to be represented as determined in space or time must agree, is already given a along with (not in) these intuitions, as condition of the synthesis of an apprehension. But this synthetic unity can be none other than that of the combination of the manifold of a given intuition in general in an original consciousness, in agreement with the categories, only applied to our sensible intuition. Consequently all synthesis, through which even perception itself becomes possihle, stands under the categories, and since experience is cognition through connected perceptions, the categories are conditions of the possibility of experience, and are thus also valid a priori of all objects of experience.




Thus if, e.g., I make the empirical intuition of a house into perception through apprehension of its manifold, my ground is the necessary unity of space and of outer sensible intuition in general, and I as it were draw its shape in agreement with this synthetic unity of the manifold in space. This very same synthetic unity, however, if I abstract from the form of space, has its seat in the understanding, and is the category of the synthesis of the homogeneous in an intuition in general, i.e., the category of quantity,a with which that synthesis of apprehension, i.e., the perception, must therefore be in thoroughgoing agreement.* If (in another example) I perceive the freezing of water, I apprehend two states (of fluidity and solidity) as ones standing in a relationb of time to each other. But in time, on which I ground the appearance as inner intuition, I represent necessary synthetic unity of the manifold, without which that relatione could not be determinately given in an intuition (with regard to the temporal sequence). But now this synthetic unity, as the a priori condition under which I combine the manifold of an intuition in general, if I abstract from the constant form of my inner intuition, time, is the category of cause, through which, if! apply it to my sensibility, I determine everything that happens in time in * In such a way it is proved that the synthesis of apprehension, which is empirical, must necessarily be in agreement with the synthesis of apperception, which is intellectual and contained in the category entirely a It is one and the same spontaneity that, there under the name of imagination and here under the name of understanding, brings combination into the manifold of intuition.



Relation , Relation


Section n. Trans. deduction of pure concepts of understanding <B>

general as far as its relationa is concerned..


such an occurrence, hence the occurrence itself, as far as possible efception is concerned, stands under the concept of the relationb fects and causes, and so in all other cases.

Categories are concepts that prescribe laws a priori to appearances, thl!S tonatureas the sum total of all appearances materialiter spectata),c and, since they are not derived from nature and do not follow it as their pattern (for they would otherwise be merely empirical), question now arises how it is to be conceived that nature must them, i.e., how they can determine a priori the combination of the manifold of nature without: dEiiving from the latter. Here is the solution to this riddle. It is by no means stranger that the laws of appearances in nature must agree with the understanding and its a priori form, i.e., its combining the manifold in general, than that the appearances selves must agree with the form of sensible intuition a priori. For laws exist just a little in the appearances, but rather exist only to the subject in which the appearances inhere, insofar as it has understanding, as appearances do not exist in themselves, but only relative to the same being, insofar as it has senses. The lawfulness of things in themselves would necessarily pertain to them even without an understanding that cognizes them. But appearances are only representations of things that exist without cognition of what they might be in themselves. As mere representations, however, t&ey stand under no law of connection at except that which the connecting faculty prescribes. Now that connects the manifold of sensible intuition is which depends on understanding for the unity of its syntllesis and on sensibility for the manifoldness of apprehension. Now since possible perception depends on the synthesis of apprehension, but latter itself, this empirical synthesis, depends on the transcendental one, on the categories, all possible perceptions, hence everything that can ever reach empirical consciousness, i.e., all appearances of nature, as far as their combination is concerned,47 stand under the categories, on which nature (considered merely as nature in general) depends, as original ground of its necessary lawfulness (as natura fOrmaliter spectata).d The pure faculty of understanding does not suffice, however, to
, Relation



Verhiilt71i'ses , "Nature regarded materially," i.e., nature in the sense of its material. d "Nature formally regarded," i.e., nature considered with regard to its form rather than its matter.


Doctrine of Elements. Pt. II. Div. 1. Bk. I. Ch. II. <B>

\ laws prescri be to .the appeara nces throug h mere categor ies a priori apof ess lawfuln as l, genera beyond those on which rests a nature in emn concer they because laws, lar pearanc es in space and time. Particu from pirically determ ined appeara nces, cannot be comple tely derive d be must nce Experie them. under the categories, althoug h they all stand expeabout but all; at laws lar added in order to come to know particu experience in general, and about what can be cognize d as an object of tion. rience, only those a priori laws offer instruc

Result of t.1Us deduct ion of the concep ts of the underst anding . <;ogWe cannot think any object except throug h categories; we cannot correthat ns intuitio h throug except t though is that nize any object our intuitio ns are sensible, and this spond to those concep ts. Now is empirical. Empiri ral c:ognition, given, is object its cogniti on, so far as no a priori cogniti on is possible uently Conseq nce. however, is experie B 166 e experience.* possibl of objects of for us except solely to objects of experience, merely limited is which on, But this cogniti nce; rather, with regard experie from ed borrow all t is not on that accoun of the understanding, ts concep pure the as well as to the pure intuitio ns tered in us a priencoun be to are that on cogniti there are elemen ts of agreem ent of ary necess a which in ways tw:o ori. Now there are t: either the though be can objects its of ts concep the experie nce with ts make the concep these 0]:" e possibl ts concep these experie nce makes ies (nor categor tlle with case the not is first The e. experie nce possibl B 167 indehence ts, concep priori a are they for n); intuitio with pure sensible be a would origin al empiric an of n assertio (the nce penden t of experie remains way second the only uently Conseq a).a aequivoc sort of generatio



me and disad* So that one may not prematurely take issue with the worriso

catvantageous consequences of this proposition, I will only mention that the sensible our of ns egories are not restricted in thinkin g by the conditio that intuition, but have an unbounded field, and only the cogniti on of objects absence the in ; intuition requires object,b the of nation we think, the determi of the latter, the thought of the object' can still have its true and useful conexsequences for the use of the subject's reason, which, however, cannot be object, the of nation determi the to pounded here, for it is not always directed thus to cognition, but rather also to that of the subject and its willing.
supThe generation of one sort of thing out of something essentially different, e.g., the posed generation of flies from rotting meat.

Object , Object


Section n. Trans. deductio n of pure concept s of under'st andmg <B>

(as it were a system of the epigen esis 48 of pure reason): ,vHnpl!w ground s the aB experience in general from the siclc; of the underst anding . But more about make experience possible, and which principles of its pV""HLn uLy yield in their application to appearances, will be in loIlovvingchapter on the transce ndental use of power If someone still wanted to propos e a middle way belwe:en two, already named ways, namely, that the categories were neither th()u~~ht a priori first princip lesa of our cogniti on nor drawn from experience, but were rather subjective predispositions implanted in us along with our existence our author in such a that their use would agree exactly with the laws of nature along experience runs (a kind of prefon nation -system49 of pure reason) , then (besides the fact that on such a hypothesis no end can seen to how far one might drive the presupp osition of predete rmined predispositions for future judgme nts) this would be decisive against the supposed middle way: that in such a case the categories lack the necessi ty that is essential to their concept. For, e.g., the concep t of cause, which asserts the necessity of a conseq uent under a presupp osed condition, would be false if it rested only on a subjective necessity, arbitrarily implan ted in us, of combin ing certain empirical represe ntations according to such a rule of re13tion. b I would not be able to say that the effect is combin ed with the cause in the object" (i.e., necessa rily), but only that I am so constit uted that I cannot think of this representation otherwise than as so connec ted; which is precisely the skeptic wishes most, for then all of our insight throug h the suppos ed objective validity of our judgme nts is nothing but sheer illusion, and there would be no shortag e of people who would not conced e this jective necessity (which must be felt) on their own; one not be able to quarrel with anyone about that merely depends on the way in which his subject is organized. Brief concep t of this deduct ion.


It is the exhibition of the pure concep ts of the with them of all theoret ical cogniti on a priori) as princip lesd the possibility of experience, but of the latter as the determ ination of appearances in space and time in general . - and the latter, from the
Pri17cipicn Verhiiltnirres

B r69




Doctrin e of Elements. Pt. II. Div. I. Bk. 1. Ch. H. <B>

of apperc eption, as the form principlea of the origina l synthet ic time, as origina l forms of space to of the unders tanding in relation sensibility.

I hold the division into paragraphs to be necessary only this far, because we have been dealing with the elemen tary concepts. Now that we will represe nt their use, the exposition may proceed in a continu ous fashion, withou t this division.>




Transcendental Second The Anal ytic

/'i'Jl]'fll 1,/TU'

General logic is constru cted on a plan that corresp onds precisely with the division of the higher faculties of cognition. are: understan ding, the power of judgm ent, and reason . In its analytic doctrine accordingly deals with concep ts, judgm ents, inferen ces, corresponding exactly to the functions and the order of those powers of mind, which are compre hended under the broad designation of understanding in general. Since merely formal logic, so conceived, abstracts from all conten t cognition (wheth er it be pure or empirical), and concern s itself with the form of thinkin g (of discursive cognition) in general, it can include in its analytical part the canon for reason, the form of which its secure precept, into which there can be a priori insight mere analysis of the actions of reason into their momen ts, withou t into consideration tlle particn lar nature of the cogniti on about it is employed. Transcendental logic, since it is limited to a determ inate content , namely that of pure a priori cognitions alone, can ot imitate general ll logic in this division. For it turns out that the tranS'c endent al use reason is not objectively valid at all, thus does not belong to the logic of truth, i.e., the analytic, but rather, as a logic of illusion , requires a special part of the scholastic edifice, under the name of the transce ndental dialect ic. Understanding and the power of judgme nt accordingly have their canon of objectively valid, thus tme use in transce ndental logic, therefore belong in its analytical part. Only reason in its att,empts make out someth ing about objects a priori and to extend cognition yond the bounds of possible experience is wholly and entirely dialect ical, and its illusory assertions do not fit into a canon of the sort the analytic ought to contain. The analytic of princip les will accordingly be solely a canon for the power of judgm ent that teaches it to apply to appearances the concepts of the understanding, which contain the conditi on for rules a



B 1711 A132

Doctrine of Elements. Pt. n. Div. 1. Bk. II reason,a as I take the actual princip les of the unders tanddesigna tion a doctrin e of the use of I ing as judgm ent, throug h which this enterpr ise may be more prepower cisely designated. transce ndenta l power of judgm ent in genera l. of rules, then the unders tanding in general is explained as the rules, i.e., of ing of subsum is the the power of datae given a de1:enmlllmg ",'hem,;r someth ing stands under of power the for at s Genera l logic contains no precept legis)b or ts abstrac it since For moreov er cannot contain them. judgme nt, business the but it to s from aU conten t of cogniti on, nothing remain of analytically dividing the mere form of cogniti on into concepts, judgachieving formal rules for all use of )/ B172 ments, and inferences, and the underst anding . Now if it wanted to show generally how one ought ng to subsum e under these rules, i.e., distinguish whethe r somethi again once except not happen stands under them or not, this throug h a rule. But just because this is a rule, it would demand another alinstruc tion for the power of judgme nt, and so it become s clear that and ted instruc though the unders tanding is certainly capable of being that equipp ed throug h rules/t he power of judgme nt is a special talent to specific is what also is cannot be taught but only practiced.{Thus this any by good made be so-called mother -wit, the lack of which cannot school; for,c althoug h such a school can provide a limited understanding with plenty of rules borrow ed trom the insight of others and as it were corgraft these onto it, nevertheless the faculty for making use of them a such of absence the in rectly must belong to the student himself, and safe is aim this for him one might prescri be to natural gift no rule an therefo re, a judge, or a statesman, can have physici A *,d misuse. tram 173 4/B

;3/ B 172

stupidity, * The lack of the power of judgment is that which is properly caned

34/ B 173

and such a failing is not to be helped. A dull or limited head, which is lacking nothing but the appropriate degree of understanding and its proper concepts, g may well be trained through instruction, even to the point of becomin se(the nt judgme of power the learned. But since it would usually still lack who in cunda Petri),' it is not at all uncommon to encounter very learned men

Ursacbc case of the given law r In the first edition, "since." first edition (E, p. 27), but nevertheless d Kant struck this footnote from his copy of the second. the let it remain in , the compani on of Peter



many fine pathological, juridical, or political rules in his head, of he can even be a thorough teacher, and yet can easily 111 application, either because he is lacking in nanlral power of Judlg-nler\t (though not in understanding), and to be sure understands the universal in abstracto but" cannot distinguish whether a case in concreto beLor\gs under it, or also because he has not received adequate training judgment through examples and actual business. is also and great utility of examples: that they sharpen the power insight of For as far as the correctness and precision of standing is concerned, examples more usually do it some dalua,ge, since they only seldom adequately the condition of rule (as casus in tcr771inis)b and beyond this often weaken the effort of the unclerstanc1mg to gain sufficient insight into rules in the universal and mc1cllertdE:ntiy of the particular circumstances of experience, thus in custom us to use those rules more like formulas Thus examples are the leading-strings of the power of judgment, he who lacks the natural talent for judgment'" can never But now although general logic can give no precepts to the power of judgment, things are quite different with transcendental logic, so that it even seems that the latter has as its proper business to correct and secure the power of judgment in the use of the pure unclerst:anc1mg through determinate rules. For although for expansion of the hence as a the understanding in the field of pure cognitions a doctrine, philosophy seems entirely unnecessary or rather lU-suttel~. since after all its previous attempts little or no has won, yet as critique, in order to'avoid missteps in judgment (lapsus judici)d the use of the few pure concepts of the understanding that we have, losophy with all of its perspicacity and art of scrutiny is called up (even thqpgh its utility is then only negative). /But the peculiar thing about transcendental philosophy is this: that in addition to the rule (or rather the general condition for rules), is given in the pure concept of the understanding, it can at the same time indicate a priori the case to which the rules ought to be cause of the advantage that it has in this regard over sciences (except for mathematics) lies just here:

174 135


the use of their science frequently give glimpses of tllat lack, which is never to be ameliorated.
, "but" added in the second edition. b I.e., as a limiting case. , Fallowing Erdmann in reading "dcnelben" instead of "dcssclben, " thus taking "the power of judgment" as its antecedent. d lapses of judgment


Doctrine of Elements. Pt. II. Div. I. Bk. n



cepts are to related to their objec~ a priori, hence its objective validity cannot be established a posteriori/for would leave that digof theirs entirely untouched; rather it must at the same time offer a general but sufficient characterization of the conditions under which objects in harmony with those concepts can be given, for otherwise they would be content, and thus would be mere logical forms and not pure concepts of the understanding. This transcendental doctrine of the power of judgment will contain two chapters: the first, which deals with the sensible condition under which alone pure concepts of the understanding can be employed, i.e., the schematism of the pure understanding; and the second, which deals wit.~ those synthetic judgments that flow a priori from pure concepts of the understanding under these conditions and ground all other cognitions a priori, i.e., with the principles of pure understanding.


aThe TranscendentallJocrrlne of the Power ofJu dgme nt (or Anal ytic First Chapter On the schematism b pure concepts ofthe

In all subsum ptions o( an object under a COJt1C(:pt the former must be homog eneous with the must contain that which is represe nted in the object is to sumed under it, for that is just whatis meant the expression "an object is contain ed under a concep t." Thus the empirical concep t of a plate has homog eneity with the pure geomet rical concep t of a circle, for the roundn ess that is though t in the former can be in the latter. Now pure concep ts of the underst anding , however, in compa nson with empirical (indeed in general sensible) intuitio ns, are entm:lv unhomogeneous, and can never be encoun tered in any intuitio n.


The following notes pertainin g to the general argumen t of the next section are all inserted on AI37 in Kant's copy of the first edition: "We cannot think any intuition s or relations [Verha!t17isse] of intllition s for the categories~ rather they must be given in experience. Thus all principles pertain merely to possible experience, since this is possible only in accordan ce with the form of the unity of understanding." (E UII, p. 27; 23:27) "The incompre hensibili ty of the categories stems from the fact that we cannot have insight into the synthetic unity of apperception." (E Uv, p. 27; 23:27) "The schema of time a line." (E IN, p. 27; 23:27) "The possibility of an object [Oijects] of the concept of the understa nding, e.g., a cause or r077177lCrcill7ll, cannot be thought a priori, consequently only an experience can be thought with the condition s under which it can become experienc e in combina tion with the concept ofthe understa nding." (E LVI, p. 27; 23:27) b Kant's copy of the first edition adds this note: "The synthesis of the understa nding is called thus if it determin es the inner sense in accordance with the unity of appercep tion." (E LVII, p. 27; 23'17)


Doctrine of Elements. Pt. n. Div. I. Bk. n. Ch. I how is the subsum ption of the latter under the former, thus the applisay cation of the categor y to appearances possible, since no one would the h throug intuited be also lE 177 that the category, e.g., causality, could and senses and is contain ed in the appearance? This question, so natural of e doctrin ndental transce a makes import ant, is really the cause which possithe show to , namely order, in the power of judgme nt necessary, bility of applying pure concep ts of the unders tandin g to appearances the in general. In all other sciences, where the concep ts throug h which eneous heterog and t differen so not object is though t in general are to from those that represe nt it in concreto, as it is given, it is unnecessary latter. the to former the of tion offer a special discussion of the applica in Now it is clear that there must be a third thing, which must stand on nce appeara the and hand homog eneity with the categor y on the one latthe other, and makes possible the application of the former to the emg anythin t (withou pure be ter. one hand and sensibl e on the other. on pirical) Such a r~reS~JJ1::l1:i()l}j:;the transce ndenta Jschem a. of The concep t of the unders tanding conta{ns pure synthet ic unity d manifol the of on conditi the manifo ld in general. Time, as the formal an s contain s, ntation of inner sense, thus of the connec tion of all represe a priori manifold in pure intuitio n. Now a transce ndental time-deterits minatio n is homog eneous with the catego ry (which constitutes on is it But priori. a rule a BI78 unity) insofar as it is univer sal and rests on ance insofar as time is AI39 the other hand homog eneous with the appear manifold. Hence an the of contain ed in every empirical represe ntation applica tion of the categor y to appearances becomes possible by means the of the transce ndental time-de termina tion which, as the schema of latter the of ption subsum concep t of the underst anding , mediate s the under the former. After what has been shown in the deducti on of the categories, hopefully no one will be in doubt about how to decide the question, whether also these pure concepts of the underst anding are of merely empirical or experie possibl a of ons of transce ndental use, i.e., whether, as conditi ns ence, they relate a priori solely to appearances, or whether, as conditio in objects to d extende be of the possibility of things in general, they can seen have we For ity). sensibil themselves (withou t any restrict ion to our there that concep ts are entirely impossible,a and cannot have any signifat icance, where an object is not given either for them themselves or cannot they uently conseq least for the elemen ts of which they consist, they pertain to things in themselves (withou t regard to how and whethe r sensibilour of ation modific the may be given to us) at all; that, further,

(E LVIII, p. 28; Altered in Kant's copy of the first edition to "are for us without sense" 23:4 6).


On the schematism of pure concepts of un(iecstaIIdiJLIg.a

is the way m objects are given to us; to the function of concepts a przorz, m category, must also contain a formal conditions (namely of the inner sense) that contain the general C011dluon which alone the category can be applied to any pure condition of the sensibility, to formal cept of the understanding is restricted, the schema the understanding, and we will call the procedure of the underst:al1(img with these schemata the schematism of the pure undejrst:uHlmg. The schema is in itself always only a product of the but since the synthesis of the latter has as its aim no intultJon rather only the unity in the determination of sensibility, the schema is to be distinguished from an image. Thus, if I five in a row, ..... , this is an image of the number five. On the contrary, if I only think a number in general, which could be five or a hUJildt'ed, thinking is more the representation of a method for representing a multitude (e.g., a thousand) in accordance with a certain conc(~pt image itself, which in this case I could survey and compare with the concept only with difficulty. Now t}.is representation of a general cedure of the imagination for providing a concept its image is I the schema for this concept. In fact it is not images of objects but schemata that ground our pure sensible conceptsY No image of a triangle would ever be adequate to the concept of it. For it would not attain the generality of the C011ce:pt, which makes this valid for all triangles, right or acute, etc., but always be limited to one part of this sphere. The schema of the tmmgle can never exist anywhere except in thought, and signifies a synthesis of the imagination with regard to pure shapes in less does an object of experience or an image of it eyer to pirical concept, rather the latter is always related schema of the imagination, as a rule for the determination of our intuition in accordance with a certain general concept. concept of a dog signifies a rule in accordance with which my imagination can specify the shape of a four-footed animal in general, being restricted to any single particular shape that e;-;:perience offers me or any possible image that I can exhibit in concreto. )Ihis schematism of our understanding with regard to appearances and their mere form is a art in the depths of the human soul, whose true operations we can divine from nature and lay unveiled before our eyes only with ditticlllt:yi We can say only this much: the image is a product of the errlpiric:al ulty of productive imagination, the schema of sensible concepts
" In the first edition the right-hand heading here changes to "On the Schematism of the Categories,"




Doctrin e of Elemen ts. Pt.

A. 142

n. Div. 1. Bk. n. Ch. I

and as it were a monog ram of pure aprias figures in space) is a in accordance with which the imori imagination, throug h which must be connec ted with the which but ages first become possible, never fully congru ent, always lves themse in are concept, to which they te. The schema of a pure designa they that only by means of the schema is someth ing that can y, contrar the on , concep t of the underst anding only the pure syntherather is but all, at never be brough t to an image concepts in general, to ng accordi sis, in accord with a rule of produc t of the ndental transce a is es, which the category express inner sense in the of ination determ the s imagination, which concern a form (time) in regard to its of ons conditi general, in accordance with all represe ntation s, insofar as these are to be connec ted togethe r apriori in one concep t in accord with the unity of apperception. Rather than pausing now for a dry and boring analysis of what is required for transce ndental schemata of pure concepts of the understandof ing in general, we would rather presen t them accordi ng to the order these. with the categories and in connec tion for outer sense is magnit udes The pure image of B182 time. The pure is it , general in space; for all objects of the senses t of the unconcep a as r, howeve magni tude schem a izes the summar that ntation derstanding, is numbe r, which is a represe nUffiThus . another to unit ) successive additio n of one (homogeneous of a d manifol the of is synthes the of ber is nothing other than the unity A 143 b I generat e time itself in the because , homog eneous intuitio n in general appreh ension of the intuitio n. 53 Realityc is in the pure concep t of the unders tanding that to which a sensati ond in general corresponds, that, therefo re, the concep t of which of in itself indicates a being (in time). Negati on is that the concept thus two the of ion opposit The time). (in which represents a non-be ing takes place in the distinct ion of one and the same time as either a filled or an empty time. Since time is only the form of intuitio n, thus of obis jects as appearances, that which corresp onds to the sensation in these (thingves themsel in things as , objects the transce ndental matter of
to "seiner," perhaps In his copy of the first edition, Kant changed this from "ihrer" to "inner sense" (E nation" "determi from nt antecede its change to intending thereby LIX, p. 28; 23'46). dadunh, daft Inserted in Kant's copy of the first edition: real of the rep' "Sensatio n is that which is really empirica l in our cognition , and the n therefore lies onto Sensatio time. fOnll, their to contrast in sense inner of ons resentati from another \<1th side all a priori cognition . Only therein, how one sensation differs [sic]" (E LX, p. quantity. their of not but degrees, priori a the beyond quality, regard to 28; 23:27)

, Realitiit


On the schematism of pure concepts of understanding hood,a reality). Now every sensation has a degree or magnitude, the same time, i.e., inner sense through which it can more or less in regard to the same representation of an object, until it ceases in nothingness (= 0 = negatio). Hence there is a relationb and connection between, or rather a transition from reality to negation, that makes every reality representable as a quantum, and the schema of a reality, as the quantity of something insofar as it fills time, is this continuous uniform generation of that quantity in time, as one descends in time from the sensation that has a certain degree to its disappearance or gradually ascends from negation to its magnitude. The schema of substance is the persistence of the real in i.e., representation of the real as a substratum of empirical time-determination in general, which therefore endures while everything else changes. (Time itself does not elapse, but t.~e existence of that which is changeable elapses in it. To time, therefore, which is itself unchangeable and lasting, there corresponds in appearance that which is unchangeable in existence, i.e., substance, and in it alone can the succession and taneity of appearances be determined in regard to time.) The schema of the cause and of the causalityc of a thing in general is the real upon which, whenever it is posited, something else always lows. It therefore consists in the succession of the manifold insofar as it is subject to a rule. The schema of community (reciprocity), or of the reciprocal causality of substances with regard to their accidents, is the simultaneity of the determinations of the one with those of the other, in accordance with a general rule. The schema of possibility is the agreement of the synthesis of various representations with the conditions of time in general (e.g., since opposites cannot exist in one thing at the same time, they can only exist one after another), thus the determination of the representation of a thing to some time. The schema of actualityd is existence at a determinate time. The schema of necessity is e the existence of an object at all times. 54 Now one sees from all this that the schema of each category contains and makes representable: in the case of magnitude, the generation (synthesis) of time itself, in the successive apprehension of an object; in the case of the schema of quality, the synthesis of sensation (perception) with the representation of time, or the filling of time; in the case of the
" Sachheit



added in the second edition.


Doctrine of Elements. Pt. H. Div. I. Bk. H. Ch. I the percept ions among li~emselves relation,a the schema of time-de termina tion); fia to all time (i.e., in accorda nce ies, time itself, as the categor its and ty modali in the schema of an object belongs to how and r whethe of ination correla te of the determ time-d etennipriori a but g nothin re therefo are time. The schema ta ng to the accordi , concern these rules, nation s in accorda nce the rime, of t conten the eries, time-s order of the categor ies, the C posall to regard in time of total sum order of rime, and finally the B 185 sible objects. From this it is clear that the schema tism of the underst anding to throug h the transce ndental synthes is of imagin ation comes down inner in n intuitio of ld manifo the all of nothing other than the unity sense, and thus indirec tly to the unity of apperce ption, as the function ta of that corresp onds to inner sense (to a receptiv ity). Thus the schema for ns conditio sole and true the are the concep ts of pure unders tanding A 146 and ance, signific with thus 'l objects, to providi ng them with a relation al use, hence the categor ies are in the end of none but a possibl e empiric synof rules general to nces appeara since they merely serve to subject the of t accoun (on unity ry necessa thesis throug h ground s of an a priori ption), apperce l origina an in usness necessa ry unifica tion of all conscio in one and thereby to make them fit for a thoroug hgoing connec tion experie nce. exAll of our cogniti ons, howeve r, lie in the entirety of all possible truth al empiric all s precede which truth, perienc e, and transce ndental and makes it possible, consists in the general relation to this. But it is also obvious that, althoug h the schema ta of sensibility first limit realize the categor ies, yet they likewise also restrict them, i.e., B 186 sensiin , (namely tanding unders the outside them to conditi ons that lie sensibility). Hence the schema is really only the phenom enon, or the us est (Numer y. categor the with ent agreem in ble concep t of an object, peret s constan enon, phaenom realitas sensatio

durabile rerum substantia phaenomenon - 11l"t{'rnij'a~. ne,ces:,ittIS P,b([(;110mCj'1il if we etc.).e Now if we leave aside a restrict ing conditi on, it may seem as


pure amplify the previou sly limited concep t; t..1}us the categor ies in their for hold should ity, sensibil of ons conditi significance, withou t any repremerely ta schema their of instead are, things in general , as they a sigsenting them how they appear , and they would therefo re have

Relation Verhiiltnis

Objecte [of the] pht:norncnon, , "Numher is the quantity [ofthe] phenome non, sensation the reality non, ctemiD'the constancy and the enduranc e of things the substance [of the] phenome etc." na, necessity [of] phenome


On the schematism of pure concepts of understanding nificance independent of all schemata and extending In fact, even after abstraction from every sensible condition, significance, but only a logical significance of the mere unity of representations, is left to the pure concepts of the understanding, but no and thus no significance is given to them that could yield a concept a of the object. b Thus, e.g., if one leaves out the sensible determination persistence, substance would signify nothing more than a something that can be thought as a subject (without being a predicate of something else). Now out of this representation I can make nothing, as it shows me nothing at all about what determinations the thing that is to count as such a first subject is to have. Without schemata, therefore, catenot gories are only functions of the understanding for concepts, represent any object. This significance comes to them from sensibility, which realizes the understanding at the same time as it restricts it.
" Changed in Kant's copy of the first edition to "cognition" (E LXI, p. 28; 23:46).






Power ofJudgment Analytic ofPrinciples) Second Chapter principles ofpure System understanding

In the previous chapter we have conside red the transce ndental power of judgme nt only in accorda nce with the general conditi ons under which alone it is authori zed to use the pure concep ts of the undersysstandin g for synthet ic judgme nts. Now our task is to exhibit in actually tanding unders the that nts judgme the ation tematic combin brings about a priori subject to this critical warnin g, for which our table e. of the categor ies must doubde ss give us natural and secure guidanc must For it is precisely these whose relation to possible experience constit ute all pure cogniti on of the unders tanding a priori, and whose all relationa to sensibility in general will, on that very accoun t, display ely transce ndenta l princip les of the use of the unders tanding complet B 188 and in a system. A priori princip les bear this name not merely because they contain in are themselves the ground s of other judgme nts, but also because they Yet not themselves ground ed in higher and more general cognitions. this h althoug For proof. all beyond them elevate not does this proper ty A 149 could not be carried further objectively, but rather ground s all cognie tion of its object,b yet this does not preven t a proof from the subjectiv from sources of the possibility of a cogniti on of an object in general ion being possible, indeed even necessary, since otherw ise the proposit ious surreptit merely a being of on suspici t greates the would raise assertion. Second, we will limit ourselves merely to those princip les that are re-

Verbdltni, Objects


Section 1. On the highest principl e of all analytic judlgIIlents

lated to the categories. principles a of therefore, according to which space and time are possibility of all things as appearances, as well as the restrict ion of these principles, namely that they cannot be related to in th(~msejives, do not belong within our confine d of investigation. the mathematical principles do not constitu te any part system, since they are drawn only from intuitio n, not from the pure co]ac(;pt understanding; yet their possibility, since they are sYJlthletj,c a pri07'i judgments, necessarily finds a place here, not in order to prove their correctness and apodictic certainty, which is not at all necessa ry, but only to make compre hensibl e and to deduce the such evident cognitions a priori. But we must also speak of the principle of analytic Judlgnlertts, trast, to be sure, to that of synthetic judgments, with erly concerned, since precisely this contras t free latter from all misund erstand ing and lay their particular nature before our eyes. The System of the Princip les Pure Unders tandin g First Section On the suprem e princip le of all analyti c judgme nts. 55 'iVhatever the conten t of our cogniti on may be, and however it may be related to the object,b the general though to be sure condition of all of our judgme nts whatsoever is that they do not C011tr:lClIICt t.1}emselves; otherwise these judgme nts in themselves (even regard to the objecty are nothing . But even if is no contrad iction within our judgment, it can nevertheless combin e 'concepts in a way not entailed by the object, or even withou t any ground being given to us either a priori or a posteriori that would justify such a judgme nt, and for all that a judgme nt may be free of any interna l contrad iction, it can still be either false or groundless. Now the propos ition that no predicate pertain s to a thing that contradicts it is called the principled of contradiction, and is a general though merely negative criterio n of all truth, but on accoun t it belongs merely to logic, since it holds of cognitions merely as cogni-




Object Object Satz: ordinarily translated as "proposition," it will be translated as "principl e" in the phrase "Satz des througho ut this section.


Doctrine of Elements. Pt. H. Div. 1. Bk II. Ch. H regard to their content , and says that contrations in general, diction entirely annihilates and cancels them. . J311t ouecan alsoma ke a positive use of it, i.e., not merely to ban falsehood and error (insofar as it rests on contradiction), but also to cognize e or affir~~ ...".."'J."~ th~ judgm~nt is analyti c, whethe r it be negativ acndyin sufficie d mative, its truth must always be able to be cognize that of y contrar the . cordan ce with the principle of contrad iction. For the which as a concep t already lies and is though t in the cogniti on of necesmust itself t concep objectO is always correct ly denied, while the b sarily be affirmed of it, since its opposite would contrad ict the object. B 191 must also allow the princip le of contra diction to count as the universal and completely sufficient principl~c of an analytic cogniti on; but its authori ty and usefulness does not extend beyond this, as a sufficient criterio n of truth. For that no cogniti on can be opposed conto it withou t annihil ating itself certainly makes this principled into a our of truth the of ditio sine qua non, but not into a determ ining ground A 152 part ic synthet cognition. Since we now really have to do only with the of our cognition, we will, to be sure, always be careful not to act contrary to this inviolable principle, but we cannot expect any advice from it in regard to the truth of this sort of cognition. There is, however, still one formula of this famous principle, although denude d of all conten t and merely formal, which contains a synthesis that is incautiously and entirely unnecessarily mixed into it. This impossible for someth ing to be and not to be at the same is: time." In additio n to the fact that apodictic certain ty issuperffiiously yet append ed to this (by means of the word "impossible"), which must by affected is be unders tood from the propos ition itself, the propos ition someis which the conditi on of time, and as it were says: "A thing = A, B 192 be thing = E, cannot at the same time be non-B, althoug h it can easily young is who bou~ (E as well as non-B) in succession." E.g., a person cannot be old at the same time, but one and the same person can very t.~e well be young at one time and not young, i.e., old, at another. Now limit not must principle of contrad iction, as a merely logical principle, e its claims to tempor al relations. Hence such a formula is entirely conA 153 abtrary to its aim. The misund erstand ing results merely from our first ently subsequ stractin g a predica te of a thing from its concep t and connec ting its opposite with this predicate, which never yields a cond tradicti on with the subject, but only with the predica te that is combine secthe and first with it synthetically, and indeed only when both the

Objects Objecte

d e

Satz ZeirJC1-hiiltnisse


Section n. On the highest principle of synthetic judgments

IS and predicate are affirmed at the same time. If I say "A person unlearned is not learned," the condition at same rime must for unlearned at one time can very well be learned at another " the time. But if I say that "No unlearned person is propositlonls analytic, since the mark (of unlearnedness) is now compriseditithe and then negative pf()pclsition follows immediately from the principle contradiction, wrthcmt condition at the same time having to be added. is then the cause why I have above so altered the formula of it that the nature of an analytic proposition is thereby clearly expressed.


Of the System of the Principles of Pure Undel"standmg' Second Section On the supreme principle of all synthetic Juc!lgrrlents) The explanation of the possibility of synthetic judgments is a nu)hlprn with which general logic has nothing to do, indeed whose name it need not even know. But in a transcendental logic it is the most Import21l1t business of all, and indeed the only business if the issue is tbe fY".'~Hj!r ity of synthetic a priori judgments and likewise the conditions the domain of their validity. For by completing this task transcendental logic can fully satisfy its goal of determining the domain and boundaries of pure understanding. In the a1(alyticjuclgll1ent I remain with the given concept in to asdiscern something about it. If it is an affirmative judgment, I cribetoJ:l1is concept already thought in it; if it is a negativ:ejud~ent, I only exclude the opposite of this it. In sYI!theticjllclgment~,J1()'eYer, I am to go beyond the ordenoconsider something entirely different is thc)U~;ht it as in a relation a to it, a relationb which is therefore never one identity, or one where neither the nor the error of the judgment can be seen 111 the judgment itself. If it is thus conceded that one must go beyond a concept in a thing is necorder to compare it synthetically with another, essary in which alone the synthesis of two conccpts can originate. But now what is this third thing, as the medium of all synthetic judgments? -rhefeis onlyone totalityc in which all of our representations are contained, namely inner sense and its a priori form, time. of representations rests imagination, (which is requisite for judgment), on the unity of apperception. Herein
" Verbiilmis b Verbiiltnis





Doctrine of Elements. Pt. n. Div. L Bk. n. Ch. n therefo re is the possibility of synthetic judgments, and, since all three representations, also the possibility of contain the sources of a sought, indeed on these grounds they be to pure synthetic judgme nts, on of objeCts is to come about which cogniti a if ry will even be necessa represe ntation s. the of is rests solely on the synthes reality, i.e., to be related to an obe objectiv have If a cogniti on is to in that object, the object must sense and ance ject, and is to have signific t that the concep ts are empty, Withou way. be able to be given in some t but not in fact cognized though sure, be to has, and throug h them one played with repremerely rather but g, thinkin this h anythin g throug only mediately, meant again not is this if object, an sentations. To give nothing other is n, intuitio in iately immed ed but it is rather to be exhibit be actual or this er (wheth nce experie to ntation than to relate its represe ts are from concep these as pure as time, and still possible). Even space nted in represe are they that is it as certain as everything empirical and validity e objectiv t withou be still would the mind comple tely a priori, and withou t sense and significance if their necessary use on the objects of experience were not shown; indeed, their represe ntation is a mere schema, which is always related to the reproductive imagination that no calls forth the objects of experience, withou t which they would have ion. distinct t withou ts concep all significance; and thus it is The possibi lity of experie nce is therefo re that which gi":<osalLof our cognitions a priori objective realiw. Now experience rests on the synof thetic unity of appearances, i.e., on a synthesis accordi ng to concepts even not would it which t withou , general in the object of appearances fit be cogniti on but rather a rhapsod y of perceptions, which would not conhly thoroug a of rules with nce accorda togethe r in any context in and nected (possible) consciousness, thus not into the transcendental a of eiples hasprin re therefo nce Experie ption. necessary unity of apperce the in of rules l genera namely priori, a it its form which ground condiry necessa as reality, e objectiv synthesis of appearances, whose But tions, can always be shown in experience, indeed in its possibility. imentirely are itions propos priori a ic apart from this r.elation synthet obipllre namelY thing, third no have possible, since they would then establish could ts concep their of unity ic ject/ in which the synthet objective re~lity. Thus althoug h in synthetic judgme nts we cognize a priori so much about space in general or about the shapes that the productive imagination draws in it that we really do not need any experience for this, still this cogniti on would be nothing at all, but an occupa tion with a mere n figmen t of the brain, if space were not to be regarde d as the conditio
object." b rcinen Gegenstand; Erdmann suggests kei71cm Gegenstand, i.e., "no


Section Ill. Systematic representation of all synthetic principles

of the appearances which constitute the matter outer experience; hence those pure synthetic judgments are related, although ately, to possible experience, or rather to its possibility itself, and on that alone is the objective validity of their synthesis grounded. Thus since experience, as empirical synthesis, is in its possibility only kind of cognition that gives all other synthesis reality, as a c()gnition also possesses truth (agreement with the object)" insofar as it contains nothing more than what is necessary for the syrltlJletlc unity.Qf experience in. general. The suptemeprinciple b ofalls)lI1thetic judgmeI1ts is, therefore: object stands wider ihenc.;cessary conditions of the. synthetic Ullity manifold of intuition in a possible experience. In this way synthetic a priori judgments are possible, if we relate the formal conditions of a priori intuition, the synthesis of the imagination, and its necessary unity in a transcendental apperception to a possible cognition of experience in general, and say: possibility of experience in general are at the same time COJndJlUC)ilS possibility oftlleobjects of experience, and on this account have jective validity in a synthetic judgment a priori." Of the System of the Principles of Pure Understanding Third Section Systematic representation of all synthetic principles of pure understanding. That there are principles anywhere at all is to be ascribed solely to pure understanding, which is not only the faculty of rules in regard to that which happens, but is rather itself the source of the principles in accordance with which everything (that can even before us as an object) necessarily stands under rules, since, without such rules, appearances could never amoUllt to cognition of an object corresponding to them. Even laws of nature, if they are considered as principles of the empirical use of the understanding, at the same time carry with them an expression of necessity, thus at least the presumption of determination by grounds that are a priori and valid prior to all experience. 57 But with, Object
! PriJlri,~;;U771


B 198


The following two notes are entered in Kant's copy at A IS8: "How are the objects determined in accordance with the concept a priori?" (E LXII, p. 28; 23'18) "The [principles] can never be proved from mere concepts, as if they dealt with things in themselves, but can only be proved from the possibility of the perception of things." (E LXIII, p. 29; 23'18)


Doctrine of Elements. Pt. II. Div. I. Bk. n. Ch. II laws of nature stand under higher principles of the unout exception the latter to particu lar cases of appearderstanding, as they ance. Thus these higher principles alone provide the concept, which contains the conditi on and as it were the exponents for a rule in general, while experience provides the case which stands under the rule. regard merely empirical be no danger that one There can , or vice versa; for the anding principles as principles of the pure underst the latter, and whose ishes necessity according to concepts that distingu ly it may hold, general how matter no lack in every empirical proposition, are, however, There on. confusi this t is easily perceived, can easily preven y ascribe to properl not eless neverth that I may pure principles a concepts pure from derived not the pure understanding, since they are ndundersta the of means by gh but rather from pure intuitions (althou Mathets. concep of faculty the is ing); the understanding, however, ce, matics has principles of this sort, but their application to experien apriic synthet such of lity possibi thus their objective validity, indeed the ori cogniti on (its deduction) still always rests on the pure understanding. Hence I will not count among my principles those of mathematics, but I will include those on which the possibility and objective a priori as validity of the latter are ground ed, and which are thus to be regarded to ts concep from the principle of these principles,a and that proceed the intuitio n but not from the intuiti on to concepts. In the application of the pure concepts of underst anding to possible s8 dyexperience the use of their synthesis is either mathem atical or exthe to partly n, intuitio the to namica l: for it pertains partly merely , intuition of ons conditi priori a istence of an appearance in general. The ce, experien e possibl a to regard however, are necessary throug hout in b while those of the existence of the objects of a possible empirical intuHence the principles of the ent. ition are in themselves only conting ry, i.e., apodictic, while necessa y mathematical use win be unconditionall carry with them the also sure, be the principles of the dynamical use, to condition of emthe under only charact er of an a priori necessity, but and indirectly; 1y mediate only pirical thinkin g in an experience, thus e that is,charevidenc iate immed consequently these do not contain the in relatidn to ty certain al univers acteristic of the former (though their judged at the better be will this Yet ). experience is not thereby injured conclusion of this system of principles. gives us entirely natural direction for the The table of these principles are nothing other than rules of table
giving objective validity to als P7i71ripiurn diesel' G17md'iitze, i.e., as the general principle the propositi ons of mathema tics which are themselves synthetic a priori "Transce ndental Aesthetic." Objecte

BI99 AI60

B200 AI61

according to the


Section IH. Systematic representation of all synthetic principles

the Qbj.e.ctjx~JJ.~"QLJ;11<:;Gqtt:;gQIit:. All principles of the pure standing are, accordingly,59


Axioms of intuition.

Anticipations of perception. 4 Postulates of empirical thinking in general. a

3 Analogies of experience.

I have chosen these titles with care, in order not to leave unnoted distinctions with respect to the evidence and the exercise of these ciples. But it will soon be shown that as far as tlle evidence as well as a priori determination of appearances according to the categories of magnitude and quality are concerned (if one attends solely to the of the latter), their principles are importantly distinct from those of two others; while the former are capable of an intuitive certainty, the latter are capable only of a discursive certainty, though in both cases they are capable of a complete certainty. I will therefore call the former the mathematical and the latter the dynamical principles.*,b,60 But one
* [Note added in the second edition:] <All combination is either compositionc (compo,itio) or connectiond The former is the syn, The following note is inserted in Kant's copy of lhe first edition: "r. Axioms of Intuition. F0177Wt. pure mathematics - pura applied mathematics Mathematics 2. Anticipations of Perception. Real. Perception is the consciouS77css ofan appearance (beftre any concept) I. Physical 3. Analogies of EXperience} Physiology { 4. Postulates of empirical 2. Metaplrysir'al Thinking in general Sensation not
experience" (E LXIv, p. 29; 2 pS)

B 201

A 162

B 201

in Kant's copy of the first edition to: "the physiological principles" (E Lxv, p. ;9; 23:46), though obviously this change was not incorporat.ed into the second edit.ion.


Doctrine of Elements. Pt.

n. Div. I. Bk. n. Ch. n

B 202

I here have in the principles of mathematics just as in the one case as principles of general (physical) dy" namics in the other, but rather have in mind only the principles of the pure understanding in relationa to the inner sense (without distinction of the representations which are given therein), through which the former principles all acquire their possibility. I am therefore tiding them more respect to their application than on account of their content, and I now proceed to the consideration of them in u"'e same order in which they are represented in the table.


<Axioms of Intuition.>b

the first edition:]

Principle of pure understanding: All appearances are, as regards their intuition, extensive magnitudes.

the second edition:]

<Their principle is: All intuitions are extensive magnitudes.>c,d,61

All appearances contain, as regards their form, an intuition in space and time, which grounds all of them a priori. They cannot be apprethesis of a manifold of what does not necessarily belong to each other, as e.g., the tvvo triangles into which a square is divided by the diagonal do not of themselves necessarily belong to each other, and of such a sort is the synthesis of the homogeneous in everything that can he considered mathematically (which synthesis can be further divided into that of aggregation and of coalition, of which the first is directed to extensive magnitudes and the second to intensive magnitudes). The second combination is the synthesis of that which is manifold insofar as they necessarily belong to one another, as e.g., an accident belongs to some substance, or the effect to the cause - thus also as represented as unhomogeneous but yet as combined a priori, which combination, since it is not arbitrary, I call dynamical, since it concerns the combination of the existence of the manifold (which can again be divided into the physical combination of the appearances with one another and the metaphysical, their combination in the a priori faculty of cognition).>



Verhiiltnis In the first edition: "On the Axioms ofIntuition." Griiflen. In ~J.,is section, Mnt uses the word "Griifle" as the German equivalent for both quantitas and quantum, as is shown by his parenthetical inclusion of the Latin words. According to C. C. E. Schmid's Worterbuch zum .leichteren Gcbmuch der Kantisrhcn Sch17ften Cl ena: Cracker, 1798), Griifle as quantitas refers primarily to the pure concept


Section Ill. Systematic representation of all synthetic principles

hended, therefore, i.e., taken up into empirical consciousness, through the synthesis of the manifold through the tions of a determinate space or time are generated, i.e., composition of that which is homogeneous a and the consciousness of the synthetic unity of this manifold (of the homogeneous). Now the consciousness of the homogeneous manifold b in intuition in general, insofar as through it the representation of an object C becomes possible, is the concept of a magnitude (Quanti). even of an object,d as appearance, is possible only through the same SYlt1tlletlc unity of the manifold of given sensible intuition which the is 111 unity of the composition of the homogeneous the concept of a magnitude, i.e., the appearances are all malgul1ttldes, and indeed extensive magnitudes, since as intuitions in space or time they must be represented through the same synthesis as that tlll'ough which space and time in general are determined.> I call an extensive magnitude that in which the representation of the parts makes possible the representation of the whole necessarily precedes the latter),62 I cannot represent to myself any no matter how small it may be, without drawing it in thought, i.e" successively generating all its parts from one point, and thereby first sketching this intuition, It is exactly the same even time. I think therein only the successive progress from one moment to another, where through all parts of time and their addition a determi-




of quantity, while GrOfte as quantum refers to "eine Griij1e in concreto" 298, 300). This distinction can be marked in English as that between "quantity" and "magnitude." However, we will follow our practice in earlier sections, using "magnitude" as the translation of GrOfte and reserving "quantity" for Quomitiit. d The following notes are inserted in Kant's copy of the first edition at the start of this section: "One must subsume the perceptions under the categories. But one can infer nothing from those categories themselves, but only from the possibility of perception, which can only happen through the determination of time and in time, in which the act [Actus] that determines the intuition is possible only in accordance with a category." (E LXVI, p. 29; 23: 28 ) "Since we can all arrange perceptions only through apprehension in time, but this is a synthesis of the homogeneous, which the concept of magnitude corresponds to in the unity of consciousness, we cannot cognize the objects of outer and inner sense other,,~se than as magnitudes in experience. Limitation of the concept of magnitude." (E LXVII, p. 30; 23'18-9) , The heading "Proof" and the following paragraph were added in the second edition. Glei"h{;wti!,!;en, syntactically singular but semantically plural, thus meaning "homogeneous units"; see the expression Gleichartigen (deT Einheiten) at A 164/B 205 below. des Gleichartigen


Doctrine of Elements. Pt. H. Div. 1. Bk. H. Ch. H n in nate magnit ude of time is finally generated. a Since the mere intuitio n is intuitio all appearances is either space or time, every appearance as ve successi h an extensive magnit ude, as it can only be cognized throug acare nces appeara All synthesis (from part to part) in appreh ension. cording ly already intuited as aggregates (multitudes of antecedently but given parts),b which is not the case with every kind of magnitude, as us by ended rather only with those that are represe nted and appreh

B 204

extensi ve.C On this successive synthesis of the productive imagination, in the generat ion of shapes, is ground ed the mathem atics of extension (geomits axioms, which express the conditions of sensible intuition etry) a priori, under which alone the schema of a pure concep t of outer appearance can come about; e.g., betwee n two points only one straight are line is possible; two straigh t lines do not enclose a space, etc. These such. as ) the axioms that properl y concer n only magnit udes (quanta But concer ning magnit ude (quantitas), i.e., the answer to the question "How big is someth ing?", althoug h various of these propositions are synthet ic and immed iately certain (indc77Zonstrabilia), there are nevA 164 or erthele ss no axioms in the proper sense. For that equals added to I since subtrac ted from equals give an equal are analytic propos itions, magam immed iately conscious of the identity of one genera tion of a B 205 nitude with the other; but axioms ought to be synthet ic a priori propothe sitions. The self-evident propos itions of numeri cal relation,d on geomof those contrary, are to be sure, synthetic, but not general, like etry, and for that reason also cannot be called axioms, but could rather be named numeri cal formulas. That 7 + 5 = 12 is not an analytic proposition. For I do not think the numbe r 12 either in the representation of of 7 nor in that of 5 nor in the represe ntation of the combination' not is two the of the two (that I ought to think this in the additio n here at issue; in the case of an analytic propos ition the questio n is only

an extensive magniInserted in Rant's copy of the first edition: "Hence the concept of merely to our ini.e., , extension is there wherein that to merely pertain not tude does of the time length the with ce accordan in e tuition. Satisfaction has extensive magnitud ly] according [intensive intensive e magnitud has also it although spent, agreeably is that to the degree of this agreeableness." (E LXVIII, p. 30; 23:29) the first edition (E LXIX, p. 30; 23'46). b These words are stricken in Kant's copy of can never take up a manifold as such "We , Inserted in Kant's copy of the first edition: But since we do not intuit these for time. and space in so doing without n perceptio in general in accordance manifold eous themselves, we must take up the homogen 23:29) 30; p. LXX, (E e." magnitud of concepts d Zahbcrhd ltnis


Section Ill. Systematic representation of all synthetic


whether I actually think predicate in the representation ject). Although it is synthetic, however, it is still a singular proposition. Insofar as it is only the synthesis of that is homogeneous in (of units) that is at issue here, the synthesis here can take a single way, even though the subsequent use of these numbers is general. If! say: "With three lines, two of which taken together are greater than the third, a triangle can be drawn," then I have mere function of the productive imagination, which draws the lines greater angle. The or smaller, thus allowing them to abut at any number 7,a on the contrary, is possible in only a single way, and likewise the number D, which is generated through the synthesis former with 5. Such propositions must therefore not be axioms (for otherwise there would be infinitely many of but rather numerical formulas. This transcendental principle of the mathematics of appearances yields a great expansion of our a priori cognition. For it is this alone makes pure mathematics in its complete precision to objects of experience, which without this principle would not be so obvious. and has indeed caused much contradiction. Appearances are not in themselves. Empirical intuition is possible only through the pure intuition (of space and time); what geometry says about the latter is therefore undeniably valid of the former, and evasions, as if objects of senses did not have to be in agreement with the rules of construction in space (e.g., the rules of the infinite divisibility of lines or angles), must cease. For one would thereby deny all objective validity to space, and with it at the same time to all mathematics, and would no longer know why and how far they are to be applied to appearances. The synthesis of spaces and times, as the essential form of all intuition, is that which at the same time makes possible the apprehension tIle appearance, thus every outer experience, consequently also all cognition of its jects, and what mathematics in its pure use proves about the also necessarily valid for the latter. All objections to this are chicanery of a falsely instructed reason, which erroneously of freeing the objects of the senses from the formal condition of our sensibility, and, though they are mere appearances, represents them as objects in themselves, given to the understanding; in which case, certainly, nothing synthetic could be cognized of iliem a priori at not even through pure concepts of space, and tlle science iliey determine, namely geometry, would not itself be possible.


13 206


B 207

. Inserted in Kant's copy of the first edition: "in the proposition 7 + 5 = 30-1; 23'46).


(E LXXI, pp.


Doc trine of Elements. Pt.


n. Div. 1. Bk. n. Ch. n

<An ticip atio ns of Per cep tion .>a

the first edition:J all perc epti ons , as suc h, The prin cip le, whi ch anti cipa tes sati on, and the real , sen app eara nce s the run s thus: In obje ct (rpa lltay the whi ch corr esp ond s to it in tud e, i.e., a deg ree. gni inte nsiv e ma pb,apl101.1le1'101l), has an [In the second edition:] obje ct In appearances the real, which is an <Its principl 63 ee.> degr a nitu de, i.e., of the sen sati on, has inte nsiv e mag Pr oof ss, i.e., one in which ther e is at the Per cep tion is empirical consciousne , as objects of perception, are not same time sensation. Appearances space and time (for these cannot be pur e (merely formal) intuitions, like efor e also con tain in addition to the perceived in themselves). The y ther objecrd in general (thr oug h which intu itio n the materials for some e is represented), i.e., the real of the som ethi ng existing in space or tim esentation, by which one can only sensation, as mer ely subjective repr cted, and which one relates to an obbe conscious that the subject is affe irical consciousness to the pure conjecr" in general. Now from the emp B 208 ible, whe re the real in the former sciousness a gradual alteration is poss formal (a priori) consciousness of the enti rely disappears, and a merely ;64 thus ther e is also possible a synmanifold in space and tim e remains nitu de of a sensation from its beginthesis of the gen erat ion of the mag arbi trar y magnitude. Now since ning, the pur e intu itio n == 0, to any e representation, and in it neither sensation in itself is not an objectiv e is to be enc oun tere d, it has, to be the intu itio n of space nor that of tim yet it still has a mag nitu de (and insure, no extensive magnitude, but which the empirical consciousness deed thro ugh its apprehension, in hin g == to its given measure), thus can grow in a cert ain tim e from not esp ond ing to which all objects! of it has an inte nsiv e mag nitu de, corr sensation, mus t be ascribed an inperception, insofar 3S they con tain of influence on sense.> tens ive mag nitu de, i.e., a degree which I can cognize and deterOne can call all cog niti on thro ugh A 166 irical cog niti on an anticipation, and min e a priori wha t belongs to emp
eb is:

ions of Perc eptio n." In the first edition: "The Anticipat

Princip second edition. wing para grap h were added in the , The heading "Proof" and the follo
d e

Objecte Object


Section IH. Systematic representation of all synthetic principles

without doubt this is the significance EpicufUs used expression 1TPOA "IJ<[IL<;. 65 But since there is something in the appearances that is never cognized a priori, and which hence also constitutes the real the sensadifference between empirical and a priori cognition, tion (as matter of perception), it follows it is really cannot be anticipated at all. On the contrary, we would nations in space and time, in regard to shape as ticipations of appearances, since they represent a that which may always be given a posteriori in experience. But if it were supposed that there is something which can be cognized a priori in every sensation, as sensation in general (without a particular one being given), would deserve to be called an anticipation in an unusual sense, since it seems strange to anticipate experience precisely in what concerns its how matter, which one can draw out of it. And this is stand. an instant I Apprehension, merely by means of sensation, fills do not take into consideration the succession of many sensations). As something in the appearance, the apprehension is not a successive synthesis, proceeding from the parts to the whole representation, it therefore has no extensive magnitude; the absence of sensation in same moment would represent this as empty, thus = o. Now in empirical intuition which corresponds to the sensation is pblle71071lCnrm); that which corresponds to its absence is negation = o. Now, however, every sensation is capable of a diminution, so that it can decrease and thus gradually disappear. 66 Hence between in appearance and negation there is a continuous nexus of many possible intermediate sensations, whose difference from one another is always smaller than the difference between the given one and zero, or comj:llet:e negation. That is, the real in appearance always magnitude, is not, however, encountered in apprehension, as this takes place means of the mere sensation in an instant and not through successive synthesis of many sensations, and thus does not proceed from the parts to the whole; it therefore has a magnitude, but not an extensive one. Now I call that magnitude which can only be apprehended as a and in which multiplicity can only be represented through apIJro,XHJrlation to negation = 0, intensive magnitude. Thus every reality in appearance has intensive magnitude, i.e., a degree. If one regards reality as cause (whether of the sensation or of another reality in appearance, e.g., an alteration), then one calls the degree of reality as cause a "moment," e.g., the moment of gravity, because, indeed, the degree designates only that magnitude the apprehension of which is not successive but instantaneous. But I touch on this here only in passing, for at present I am not yet dealing with causality. Accordingly every sensation, thus also every reality in appear291

A 167



B 2 IQ


B 2 II

Doctrin e of Elements. Pt. H. Div. 1. Bk. II. Ch. H

ance,a however small it may be, has a degree, i.e., an intensive magnican still always be diminished, and betwee n reality and tude, negatio n there is a continu ous nexus of possible realities, and of possible smaller perceptions. Every color, e.g., red, has a degree, which, howeve r small it may be, is never the smallest, and it is the same with warmth , with the momen t of gravity, etc. is The propert y of magnit udes on accoun t of which no part of them time and Space ity. continu their called is simple) is part the smallest (no enare quanta continua,b because no part of them can be given except as way a dosed betwee n bounda ries (points and instants), thus only in such of that this part is again a space or a time. Space therefo re consists only spaces, time of times. Points and instants are only boundaries, i.e., mere ns places of their limitation; but places always presupp ose those intuitio that that limit or determ ine them, and from mere places, as compo nents be could be given prior to space or time, neither space nor time can A I 70 the composed. Magnit udes of this sort can also be called flowing , since a is ion generat their in ation) imagin tive produc synthesis (of 1J~e by progres s in time, the continu ity of which is customarily designated B2I 2 ng"). ("elapsi g" the expression "floWin appearances whatsoever are accordingly continuous magnitudes, either in their intuition, as extensive magnitudes, or in their mere perof ception (sensation and thus reality), as intensive ones. If the synthesis many of te aggrega an is it then ted, interrup is nce appeara of the manifold appearances, and not really appearance as a quantum, which is not generated through the mere continuation of productive synthesis of a certain kind, but through the repetition of an ever-ceasing synthesis. IH call thirby teen dollars a quantu m of money, I do so correctly insofar as I mean ous that an amoun t of a mark of fine silver, which is to be sure a continu te constitu could part each but t smalles the is part no which in magnitude, by if a coin that would always contain material for still smaller ones. But the term "thirtee n round dollars" I mean so many coins (whatever their a amoun t of silver might be), then it would not be suitable to call this quantu m of dollars, but it must instead be called an aggregate, i.e., a number of coins. Now since there must still be a unity ground ing every numA 171 ber, appearance as unity is a quantum, and is as such always a continuum. ely, are Now if all appearances, considered extensively as well as intensiv c continuous magnitudes, then the propos ition that all alteration (transihas a degree any Inserted in Kant's copy of the first edition: "I do not say that all reality 3 I; 23:29) p. LXXII, (E e." magnitud extensive an has thing every that more than continuous magnitudes just like that of Inserted in Kant's copy of the first edition: "[ The] possibility of which, be given othcannot nding, understa the of all other objects [Objeete] of pure concepts p. 3 I; 2 P9) LXXIII, (E itself." in le cognizab not is It . intuition sensible in than erwise



Section ITl. Systematic representation of all synthetic principles

tion of a thing from one state into another) is also continuous proved here easily and with mathematical self-evidence, a an alteration in general did not lie entirely beyond the boundaries transcendental philosophy and presuppose empirical principles.a For the understanding gives us no inkling a priori that a cause is possible alters the state of things, i.e., detennines them to the opposite of a certain given state, not merely because it simply does not give us insight into possibility of this (for this insight is lacking in many a priori cognitions), but rather because alterability concerns only certain determinations of appearances, about which experience alone can teach us, while their cause is to be found in the unalterable. But since we have before us here ing that we can use except the pure fundamental concepts of experience, in which there must be nothing at all empirical, we cannot anticipate general natural science, which is built upon certain fullclamllental experiences, without injuring the lmity of the system. Nevertheless, wc are not lacking proofs of Ole great influence that our principle has in anticipating perceptions, and even in making good their absence insofar as it draws the bolt against all the false inferences that might be drawn from that. If all reality in perception has a degree, between and negation there is an infinite gradation of ever lesser degrees, and if likewise every sense must have a determinate degree of receptivity for the sensations, whether then no perception, hence also no experience, is possible immediately or mediately (through whatever detour in inference one might want), would prove an entire absence of everything real in appearance, i.e., a proof of empty space or of empty time can never be drawn from experience. For, first, the entire absence of the real in sensible intuition cannot itself be perceived, and, second, it cannot be deduced from any single appearance and the the degree of its reality, nor may it ever be assumed for the explanation of that. For even if the entire intuition of a determinate space or time is real through through, i.e., no part of it is empty, yet, since every reality has its degree that can decrease to nothing (emptiness) through infinite steps while the extensive magnitude of the appearance remains unaltered, it must infinitely different degrees with which space or time is filled, and the intensive magnitude in different appearances can be smaller or greater even though the extensive magnitude of the intuition remains identical. We will give an example of this. Nearly all natural philosophers,b since they perceive a great difference in the quantity of matter of dittenent sorts in the same volumes (partly through the moment of gravity, or weight, partly through the moment of resistance against other, moved
, hincipic77 Naturlcbrcr

B 2 13


B 2 14


B 2 15


Doctrin e of Elemen ts. Pt.

n. Div. 1. Bk. n. Ch. n


B 2 16



matter), unanimously infer from this that this volume (extensive magnimatter, althoug h to be sure tude of the appearance) must be empty in mathein different amounts. But who anlOng these for the most part infertheir that realized ever matical and mechanical students of nature so make they which osition, ence rested solely on a metaphysical presupp (1 space in real the that much pretense of avoiding? - for they assume al empiric are these since cannot call it here impenetrability or weight, concepts), is everyw here one and the same, and can be adifferentiated this only according to its extensive magnitude, i.e., amoun t. Against and nce experie in ground no presupposition, for which they can have proof, ndental transce a oppose which is therefore merely metaphysical, I but which, to be sure, will not explain the variation in the filling of space, osipresupp the of ty which still will entirely obviate the alleged necessi astion that the difference in question cannot be explained except by the granting least at of sumpti on of empty spaces, and which has the merit the underst anding the freedom also to think of this difference in another ry way, if the explanation of nature should make some hypothesis necessa combe can spaces for this end. For there we see that, althoug h equal pletely filled with different matters in such a way that in neither of them is there a point in which the presence of matter is not to be encountered, nevertheless everything real has for the same quality its degree (of resistance or of weight) which, withou t diminu tion of the extensive magnitude or amount,b can become infinitely smaller until it is transformed e.g. into emptiness and disappears. Thus an expansion that fills a space, in without can, nce) warmth , and likewise every other reality (in appeara degree in e decreas the least leaving the smallest part of this space empty, as infinitely, and nonetheless fill the space with this smaller degree just no by is here aim My wen as anothe r appearance does with a larger one. means to assert that this is how it really is concern ing die specific gravity of the variety of matters, but only to establish, on the basis of a prinan ciple of pure underst anding , that the nature of our perceptions makes real the that assume to explanation of this sort possible, and that it is false in appearance is always equal in degree and differs only in aggregation on and its extensive magnitude, especially when this is allegedly asserted the basis of a principle of underst anding a priori. Nevert heless there must always be someth ing striking about this anto ticipati on of percep tion for a researc her who has become accustomed some and s, cautiou transce ndental conside ration and thereby become reservation is aroused about the fact that the unders tanding can anticidepate a synthet ic propos ition of the sort which that concer ning the ity possibil the about gree of evelyth ing real in appearance is, and thus


b lVlenge


Section Ill. Systematic representation of all synthetic principles

of the inner variation of the sensation itself if one abstracts pirical quality, and it is therefore a question not of solutllon, how the understanding can assert something synthetic a priori about pearances, and indeed anticipate them in that which is empirical, namely what pertains to sensation. The quality of sensation is always merely and cannot represented a priori at all (e.g. colors, taste, etc.). But the corresponds to sensations in general, in opposition to the negation = 0, represents something whose concept in itself contains a being, and does not signify anything except the synthesis in an consciousness in general. In inner sense, namely, the empirical consciousness can raised from up to any greater degree, so that the very same extensive magnitude of intuition (e.g., an illuminated surface) can excite as great a sensation as an aggregate of many other (less surfaces together. One can therefore abstract entirely from the extensive tude of appearance and yet represent in the mere sensation in one 1110ment a synthesis of uniform increase from up to the given enlplrICal consciousness. All sensations are thus, as such, given only a posteriori,1l but their property of having a degree can be cognized a priori. It is remarkable that we can cognize a priori of all magnitudes in general a single quality, namely continuity, but that in (the real of appearances) we can cognize a nothing more their intensive quanrity,b namely that they have a degree, and everything else is left to experience.

r 76



3 <Analogies of Experience.>c,6 7

[In the first edition:] Their general principle is: As regards their.existence, appearances stand a priori under rules of determination of their relation to each other one time. [In the second edition:] <Their principled is: Experience is possible only through the representation of a necessary connection ofperceptions.>

Experience is an empirical cognition, i.e., a cognition that detem1ines an object! through perceptions. It is therefore a synthesis of perceptions,
, Following Erdmann, reading "a postcri077" instead of "a priori." b QlI.7mitiit , In the first edition: "The Analogies of Experience." d f1'incip , The heading "Proof" and the following paragraph were added in the second edition. f Object


Doctrine of Elements. Pt. n. Div. 1. Bl<. n. Ch. II ic which is not itself contained in perception but contains the synthet constiwhich unity of the manifold of percept ion in one consciousness, extutes what is essential in a cognition of objects a of the senses, i.e., of in Now senses). the of n perience (not merely of the intuitio n or sensatio B 2 19 so ently, conting only r experience, to be sure, perceptions come togethe perthat no necessity of their connection is or can become evident in the b the of sition ceptions themselves, since apprehension is only a juxtapo of ty necessi the of manifold of empirical intuition, but no representation and space in ses juxtapo the combin ed existence of the appearances that it is a cogniti on of obtime is to be encoun tered in it. But since experience d jectsC through perception, consequently the relation in the existence of but the manifold is to be represented in it not as it is juxtaposed in time the ed, perceiv be as it is objectively in time, yet since time itself cannot about come only can determination of the existence of objects e in time prithroug h their combin ation in time in general, hence only through a along ty necessi ori connecting concepts. Now since these always carry of with them, experience is thus possible only through a representation the necessary connec tion of the perceptions.> fThe three modi of time are persist ence, succes sion, and simulta neI 77~ ity. Hence three rules of all tempor al relations of appearances, in accordance with which the existence g of each can be determ ined with regard it possible. to the unity of all time, precede 311 experience and first make necessary the The general princip le of all three analogies rests on B 220 usconscio al unity of apperce ption with regard to all possible empiric pria an is that ness (of perception) at every time, consequently, since g ori ground , it rests on the synthetic unity of all appearances accordin to related is to their relations in time. For the origina l apperce ption inner sense (the sum of all representations), and indeed related a priori
Objecte Zusm711'71cn,tellunp; ( Objecte " will translate Verhdltni, unless otherd Through out this section of the work, "relation 'wise noted. ( Objecte in his copy of the first edif The text common to the two editions resumes here, although written the following two instead and hs paragrap two next the out struck tion Rant had notes: in me so far as I "For the proposit ion that I myself am simultan eous with all time us." (E tautologo be would form, its think it, i.e., with the whole time that I think, or

LXXIV; p. 31; 23:29)

hence the sub"The principle of persistence does not concern things in themselves, appearances. only but tion, appercep i.e., itself, as things of tations ject of the represen of time subject the to even not else, For the concept of time does not apply to anything 23:29) 31; p. itself." (E LXXV; relation [Verbditni.rJ In Kant's copy of the first edition, "existence" is replaced with "the 23'47). 31; p. LXXVI, (E ce" of the real in appearan


Section Ill. Systematic representation of all synthetic principles

to its form, i.e., the relation of the manifold empirical consciousness in

time. Now in the original apperception all of this so as its temporal relations are concerned, is to be unified; for this is what its transcendental unity, under which everything stands that is to belong to my (i.e., my united)a cognition, and thus can become an object for me, asserts a priori. This synthetic unity in the temporal of perceptions, which is determined a priori, is thus ical time-determinations must stand under rules determination, and the analogies of experience, with deal, must be rules of this sort. These principles have the peculiarity that they not concern appearances and the synthesis of their empirical intuition, their existence and their relation to one a.nother with regard to this their existence. Now the way in which something is apprehended in appearance can be determined a priori so that the rule of its synthesis at the same time yields this intuition a priori in every empirical eX:lmlple, i.e., can bring the former about from the latter. Yet the existence appearances cannot be cognized a priori, and even if we could succeed on this path in inferring to some existence or other, we still not able to cognize it determinately, i.e., be able to anticipate that through which its empirical intuition is differentiated from others. The preceding two principles, which I named the mathematical ones in consideration of the fact that they justified applying mathematics to appearances, pertained to appearances with regard to their mere possibility, and taught how both their intuition and the real in their perception could be generated in accordance with rules of a mathematical synthesis, hence how in both cases numerical magnitudes and, with them, the determination of the appearance as magnitude, could be used. E.g., Iwould be able to compose and determine a priori, constmet the degree of the sensation of sunlight out of about 200,000 from the moon. Thus we can call the former principles constitutive. must be different with those



B 22 I




neither axioms nor are to thought of; if a perception is given to us in a temporal to others (even how great though indeterminate), it cannot be said a priori which this other perception is, but only how it is necessarily combined with the first, as regards its existence, in this modus of time. In philosophy analogies signify something very different from what they represent in matheb

Verhiiltnis Principicn


Doctrine of Elements. Pt. H. Div. 1. Bk. II. Ch. n matics. In the latter they are formulas that assert the identity of two relars tions of magnitude,a and are always constit utive, so that if two membe be can i.e., given, thereby of the propor tion are given the third is also two constructed. In philosophy, however, analogy is not the identity of given three from where s, of two qualita tive relation quantit ative membe rs I can cognize and givej.priJ!Ei only the relatio n to a fourth for membe r but not this fourth memo er itself, althoug h I have a rule A 180 analogy An there. it ring . . seeking it in experience and a mark for discove experience will therefo re be only a rule in accordance with which unity as of experience is to arise from perceptions (not as a percept ion itself, the of valid be not will it le empirical intuition in general), and as a princip objects (of the appearances) constit utively but merely regulatively.68 The very same thing will also hold for the postulates of empirical thinkB 223 (of ing in general, which togethe r concern the synthesis of mere intuition ce experien of and , the form of appearance), of perception (of its matter) regulaonly are they (of the relation of these perceptions), namely that es, tive principles, and that they differ from the mathematical principl estabis which y, which are constitutive, not, to be sure, in their certaint i.e., lished a priori in both cases, but yet in the manner of their evidence, . tration) with regard to their intuitiveness (thus also their demons But what must be remem bered about all synthetic principles and especially noted here is this: that these analogies have their sole signifithe cance and validity not as principles of the transcendental use of they hence use, al empiric underst anding but merely as principles of its be can be proven only as such; consequently the appearances must not AI8I a. schemat their under subsumed under the categories per se, but only things were related be to For if the objects to which these principles were in themselves, then it would be entirely impossible to cognize anything ces about them synthetically a priori. Now it is nothing but appearan must es principl priori a all whose complete cognition, to which in the end come down, is only possible experience, and consequently those principles can have as tlleir goal nothing but the conditions of the unity of ernns pirical cognition in the synthesis of the appearances; but these conditio B 224 ndundersta the of t are though t only in the schema of the pure concep ing, and the category contains the function, unrestr icted by any sensible s, condition, of their unity, as of a synthesis in generaL These principle with accord in only therefore, justify us in compo unding the appearances in an analogy with the logical and general unity of concepts, and hence (its n executio its in the principle itself we make use of the category, but its application to appearances) we set its schema in its place, as the key to conng restricti its as use, or rather we set the latter alongside the former, dition, under the name of its formula.


Section HI. Systematic representation of all synthetic principles

First Analogy. Principle of the persistence <of substance.>a,69
[In the first edition:] All appearances contain that which persists (substance) as the object itself, and that which can changeb as its mere termination, i.e., a way in which the object exists.c


its quantum is neither increased nor diminished in nature.>


[In the second edition:] <In all change of appearances substance persists,

der Substanz added in the second edition. das Wimdelbare , The following series of notes is entered at the beginning of the "First Analogy" in Kant's copy of the first edition: "Here it must be shown that this proposition does not pertain to any other substances than those whose alteration is effected only through moving causes, and also consists only in movement, consequently in alteration of relations [Relatio17t1Z]." (E LXXVII, p. 31: 2 no) "All arising and perishing is only the alteration of that which endures (the substance), and this does not arise and perish (thus the world also does not)." (E LXXVIII;

p. 32: 23:3)

"Change can only be perceived through that which persists and its alteration. For the difference of the times in which things are can only be perceived in them as parts of one and the same time. All change is only the division of time. Hence tllere must be something that exists throughout the entire time, since the whole is always the ground of the division. Hence substance is the substratum, and that which is changing is the way in which this exists." (E LXXIX, P.32: 23:30) "Here the proof must be so conducted that it applies only to substances as nomena of outer sense, consequently from space, which exists,~t all time along its determination. "In space all alteration is movement; for if there were something else in tlle relations [Rclatiol1C'1lj, then in accordance with the concept of alteration the subject would persist. Therefore everything in space would have to disappear at the same time." (E LXXX, p.


2 n O)

"If the substance persists, while the accidents change, but the substance, if all aceide11tia are taken away, is the empty sub.ITfmtia!c, then what is it that persists? Now everything that can be distinguished from that which changes in experience is quantity Igro:cseJ, and this can only be assessed ilirough ilie magnitude of tlle merely relative effect in the case of equal external relations [Relatiol1f11] and therefore applies only to bodies." (E LXXXI, p. )2; 23:30- 1 ) "Here alterations must be discussed." (E LXXXII, p. 32; 23:31) d The heading "Proof' and the following first paragraph in the second edition replace the heading "Proof of this first Analogy" and iliis opening paragraph in the first edition: "All appearances are in time. This can determine ilie relation [Verhd!tnisJ in their existence in two ways, insofar as they exist either successively or simultaneously. In the case of the former time is considered as temporal series, wiili regard to ilie latter as temporal domain."


Doctrine of Elements. Pt. n. Div. I. Bk. n. Ch.

appearances are in time, in which, as substra tum (as persiste nt form simult aneity as well as succes sion can alone of inner intuition), nces be represe nted. The time, therefo re, in which all change of appeara sucwhich in that is it since ; change not does and is to be though t, lasts B 225 it. of s ination determ as only nted represe be can cession or simulta neity obthe in is it uently Conseq itself.a by ed Now time cannot be perceiv be jects of percept ion, i.e., the appearances, that the substra tum must or change which in and general in time encoun tered that represe nts of relation the h throug ension appreh in ed simulta neity can be perceiv i.e., real, ing everyth of tum substra the er, Howev it. the appearances to b belongs to the existence of things, is substan ce, of everyth ing dewhich everyth ing that belongs to existence can be though t only as a which to relation in , persists which that termina tion. Conseq uently subalone all tempor al relation s of appearances can be determ ined, is the as which nce, appeara the in real the i.e., stance in the appearance, e, therefor this, Since same. the s remain substra tum of all change neither be also Can nature in m quantu its e, cannot change in existenc increas ed nor diminished.> Our appreh ension of the manifo ld of appeara nce is always succesA 182 sive, and is therefo re always changing. We can therefo re never deteris mine from this alone whethe r this manifold, as object of experience, alwhich it ground not does ing simulta neous or successive, if someth ways exists, i.e., someth ing lasting and persist ing, of which all change and simulta neity are nothing but so many ways (modi of time) in which B 226 are that which persists exists. Only in that which persists, therefore, only the are ion success and neity simulta (for e tempor al relation s possibl emrelations in time), i.e., that which persists is the substr atum of the A 183 termina time-de all alone which by itself, time pirical represe ntation of the as time to ion express general gives nce tion is possible. Persiste acconstan t correla te of all existence of appearances, all change and all apthe only but itself, time affect not does compan iment.' For change in pearances in time (just as simulta neity is not a modus for time itself, If ). another one succeed all rather but neous which no parts are simulta to have would one itself, time to ion success a one were to ascribe such .d think yet anothe r time in which this succession would be possible

fUr sich


by another, i.e., Bcglcit1lng, here connotin g the accompa niment of one state of affairs ce. coexisten or eity" "simultan what Kant is here otherwise calling The following notes are added here in Kant's copy of the first edition: n of successive "The perceptio n of enduranc e is not possible through the perceptio through the relanot also thus time, to series their of relation the of and ations determin temporal space, but tion to another sequence of determin ations, which itself requires a


Section Ill. Systematic representation of all synthetic principles

Only through which persists does existence in different one the temporal series acquire a magnitude, For in mere sequence alone existence is always disappearing and beginning, and never has the least magnitude. Without that which persists there is therefore no temporal relation. Now time cannot be perceived in itself; thus this persisting thing in the appearances is the sut)stlratllm of all time-determination, consequently also the possibility of all synthetic unity of perceptions, i.e., of experience, and in this persisting thing all existence and all change in time can be regarded as a modus of the existence of that which lasts and persists. Therefore in all appearances that which persists is object itself, i.e., the substance (phacnomcnon), but everything that changes or can change belongs only to the way in which this substance or substances exists, thus to their determinations. I find that at all times not merely the philosopher even the common understanding has presupposed this persistence as a substratum of all change in the appearances, and has also always accepted it as indubitable, only the philosopher expresses himself more determinately in saying that in all alterations in the world the substance remains and only the accidents change. But I nowhere find even attempt at a proof of this so obviously synthetic proposition, indeed it only rarely stands, as it deserves to, at the head of the pure and completely a priori laws of nature. In fact the proposition that substance persists is tautological. For only this persistence is the ground for our application of the category of substance to appearance, and one have proved that in all appearances there is something persists, which that which changes b is nothing but the determination of its existence. But since such a proof can never be conducted dogmatically, i.e., from concepts, because it concerns a synthetic a and it was never considered that such propositions are only in relatione to possible experience, hence that they can be proved only through a
through something whose existence is not a series of successions, but which includes these in itself as its determinations, consequently per dllmhilitotcm [through the durability] of substance. "This proof, like all synthetic ones, is proved only from the possibility of perception. It is valid where I cannot perceive substance outside of its alterations; but wherc I cannot perceive it except through these alterations themselves, it is not valid, and I can estimate its endurance and in general the time of its alteration only through outer things, as I, since I think, think my own existence; my persistence is therefore not proved." (E LXXXIII, pp. 32-3; 2nl) "No quantum of substance is possible in the soul. Hence also notlling that one could determine through any predicate and call persistent." (E LXXXIV, P32; 23;31) Exi.rtcnz b daJ' TV177dclhorc

B 227


B 22 8


Doctrine of Elements. Pt. II. Div. 1. Bk. II. Ch. II


it has, to latter, it is no wonde r the possibility of deduct ion empirin it for need the feels one (for experience be sure, ground ed . proved been ical cognition), but has never A_ philoso pher was asked: How much does the smoke weigh? He was burnt replied: If you take away from the weight of the wood of the weight the have you over, left are that the weight of the ashes matthe fire in even that le overtib incontr as smoke. He thus assume d in on alterati an suffers only rather but ter (substance) never disappears a Likewise the propos ition "Nothi ng comes from nothing " is its form. of only anothe r conseq uence of the princip le of persistence, or rather For nces. appeara the in subject proper the the everlasting existence the if that in the appeara nce which one would call substan ce is to be the in ce existen all then tion, termina proper substra tum of all time-de it and it in ined determ be to able be must time past as well as in future if only ce substan of name the nce appeara an alone. Hence we can grant exly perfect even not is which time, all at we presupp ose its existence fupressed throug h the word "persis tence" since this pertain s more to ably insepar is ng persisti of ty necessi ture time. Nevert heless the inner B 229 connec ted with the necessity of always having existed, and the expression may therefo re stand. Gigni de nihilo nihil, in nihilurlZ nil posse rr:,;erti,b A 186 and are two propos itions which the ancients connec ted inseparably, being, erstand misund of out only ed which are now sometim es separat that and lves, themse in things to cause one imagines that they pertain on a the former would be oppose d to the depend ence of the world worry this but ed); concern is ce substan its as suprem e cause (even as far exis unnecessary, for here the issue is only appearances in the field of allow to were we if e possibl be never perienc e, the unity of which would then new things (as far as their substance is concern ed) to arise. For time, of unity the nt represe can alone that ar everyth ing would disappe its namely the identity of the substra tum in which alone all change has than more nothing re therefo is nce thoroug hgoing unity. This persiste the way in which we represe nt the existence of things (in appearance). The determ ination s of a substance that are nothing other than particular ways for it to exist are called acciden ts. They are always real, since they concern the existence of the substance (negations are merely detere). minatio ns that express the non-be ing of someth ing in the substanc (e.g., ce substan in real this to ce existen lar Now if one ascribes a particu B 230 motion , as an acciden t of matter) , then this existence is called "inherence," in contras t to the existence of the substance, which is called "subsistence." Yet many misinte rpretati ons arise from this, and it is more A 187

this? Not from exInserted in Kant's copy of the first edition: "Whence does he know 23:47) 34; p. LXXXV, (E perience." Nothing comes out of nothing, and nothing can revert into nothing.


Section III. Systematic representation of all synthetic principles Drecise and correct if one characterizes the accident " way in which the existence of a substance is positively determinedJo Nevertheless, thanks to the conditions of the logical use of our understanding, it is still unavoidable for us to abstract out, as it were, that which can change in the existence of a substance substance remains, and to consider it in relation to what is really persistent and damental;a thus this category also stands under the of re1:at1ClllS, more as their condition than as itself containing a relatl,on. Now on this persistence there is also grounded a correction of the concept of alteration. Arising and perishing are not alterations of that which arises or perishes. Alteration is a way of existing that succeeds anis other way of existing of the very same object. Hence everything altered is lasting, and only its state changes. 7I Thus since this concerns only the determinations that can cease or begin, we can say, in an expression that seems somewhat paradoxical, only persists (the substance) is altered, while that which is changeableb does not suffer any alteration but rather a change, since some cease and others begin. Alteration can therefore be perceived only in substances, and arising or perishing per se cannot be a possible perception unless it concerns merely a determination of that which persists, for it is this very that persists that makes possible the representation of the transition from one state into another, and from non-being into being, which can therefore be empirically cognized only as changing determinations of that which lasts. 72 If you assume that something simply began to be, then you would have to have a point of time in which it did not exist. But what would you attach this to, if not to that which already exists? For an empty time that would precede is not an object of perception; but if you connect this origination to things that existed antecedently and which endure until that which arises, then the latter would be only same a determination of the former, as that which persists. It is just with perishing: for this presupposes the empirical representation of a time at which there is no longer an appearance. Substances (in appearance) are the substrata of all tinle-del:enmillations. The arising of some of them and the perishing of others would itself remove the sole condition of the empirical unity of and the appearances would then be related to two different times, in istence flowed side by side, which is absurd. For there is only one in which all different times must not be placed simultaneously but one after another. Persistence is accordingly a necessary condition under

B 231


B 232


das Beban"Zicbe und Radikafe das Wm:'de!17arc


Doctrine of Elements. Pt.

n. Div. 1. Bk. n. Ch. n

appearances, as things or objects, are determinable in a possible experience. As to the empirical criterion of this necessary persistence and with it of the substantiality of appearances, however, what follows wi11 give us the opportunity to note what is necessary.7 3

Second Analogy. <Principle of temporal sequence according to the law of causality.>a,74

the first edition:]

Everything that happens (begins to be) presupposes something which it follows in accordance with a rule.

[In the second edition:]

<All alterations occur in accordance with the law of the connection of cause and effect.> Proof <b(That all appearances of the temporal sequence are collectively only alterations, i.e., a successive being and not-being of the determinations of the substance that persists there, consequently that the being of the substance itself, which succeeds its not-being, or its not-being, which succeeds its being, in other words, that the arising or perishing of the substance does not occur, the previous principle has shown. This could also have been expressed thus: All change (succession) of appearances is only alteration; for the arising or perishing of substance are not alterations of it, since the concept of alteration presupposes one and the same subject as existing with two opposed determinations, and thus as persisting. - Mter this preliminary reminder the proof follows.) I perceive that appearances succeed one another, i.e., that a state of things exists at one time the opposite of which existed in the previous state. Thus I really connect two perceptions in time. Now connection is not the work of mere sense and intuition, but is here rather the product of a synthetic faculty of the imagination, which determines inner sense with regard to temporal relations. This, however, can combine the two states in question in two different ways, so that either one or the other precedes in time; for time cannot be perceived in itself, nor can what precedes and what follows in objects C be as it were empirically determined in relation d to it. I am therefore only conscious that my
In the first edition: "Principle of Generation." The following two paragraphs were added in the second edition. c Objecte d Bez:ich1l71Z



Section HI. Systematic representation of all synthetic pnnclpl'Es

imagination places one state before and the after, not one state precedes the other in the object;a or, in other words, thx'ough mere perception the objective relation of the appearances that are succeeding one another remains undetermined. Now in order for this to be cognized as determined, the relation between the two states must be thought in such a way that it is thereby necessarily determined vice versa. them must be placed before and which after rather concept, however, that carries a necessity of synthetic with it can only be a pure concept of understanding, which does not lie in the perception, and that is here the concept of the rdation of cause and effect, the former of which determines the latter in time, as its consequence/ and not as something that could merely precede in nation c (or not even be perceived at all). Therefore it is only because we subject the sequence of the appearances and thus all alteration to the law of causality that experience itself, i.e., empirical cognition of is possible; consequently they themselves, as objects of experience, are possible only in accordance with this law.> dThe apprehension of the manifold of appearance is always successive. The representations of the parts succeed one another. they also succeed in the object is a second point for reflection, not contained in the first. Now one can, to be sure, call everything, even every representation, insofar as one is conscious of it, an object;e only what this word is to mean in the case of appearances, not insofar as they are (as representations) objects/ but rather only insofar as designate an object,g requires a deeper investigation. Insofar as are, merely as representations, at the same time objects of consciousness, they do not differ from their apprehension, i.e., from their being taken up into the synthesis of the imagination, and one must h1.erefore say that the manifold of appearances is always successivfly generated in the mind. If appearances were things in themselves, then no human being would be able to assess from the succession of representations how the manifold is combined in the object. h For we have to do our representations; how things in themselves may be (without regard to
, Objecte Folge , in der HiT/hiLi/I7i'!!: d Although the text common to the two editions resumes here, in his copy of the first edition Kant crossed out the next fourteen paragraphs, through A20rlB 246, suggesting that at one point he had contemplated an extensive revision of the second analogy that he did not in the end undertake (E, p. 34). , Obieet f Oijecte Object h Object





Doctrine of Elements. Pt. n. Div. 1. Bk. n. Ch. n affect us) is entirely beyond our represe ntation s throug h in cognitive sphere. Now althoug h the appearances are not things us to given be can themselves, and nevertheless are the only thing that for cognition, I still have to show what sort of combin ation in time pertains to the manifold in the appearances itself even though the reprethe sentatio n of it in appreh ension is always successive. Thus, e.g., stands that house a of appreh ension of the manifold in the appearance of before me is successive. Now the questio n is whethe r the manifold . concede will one no ly this house itself is also successive, which certain transcen to object an of t Now, however, as soon as I raise my concep an dental significance, the house is not a thing in itself at all but only is which of object appearance, i.e., a represe ntation , the transce ndental the how n, questio the unknown; therefo re what do I unders tand by manifold may be combin ed in the appearance itself (which is yet nothis ing in itself)? Here that which lies in the successive apprehension in me, to given is that considered as represe ntation , but the appearance taspite of the fact that it is nothing more than a sum of these represen I which t, concep tions, is conside red as their object, with which my quickly One draw from the represe ntation s of apprehension, is to agree. sees that, since the agreem ent of cogniti on with the object" is truth, only the formal conditions of empirical truth can be inquire d after here, and appearance, in contrad istincti on to the represe ntation s of apprehension, can thereby only be represe nted as the object" that is distinct from them if it stands under a rule that distinguishes it from every other y. appreh ension, and makes one way of combin ing the manifold necessar y necessar this of on That in the appearance which contains the conditi rule of appreh ension is the object. ' Now let us proceed to our problem . That someth ing happens, i.e., that someth ing or a state comes to be that previously was not, cannot be empirically perceived except where an appearance precedes that does an not contain this state in itself; for a reality that would follow on be can things, of empty time, thus an arising not precede d by any state an of nsion apprehe appreh ended just as little as empty time itself. Every Since one. r anothe occurre nce is therefo re a percep tion that follows this is the case in all synthesis of appreh ension, however, as I have shown above in the case of the appearance of a house, the apprehension I of an occurre nce is not yet thereby distinguished from any other. Yet g happenin a s also note that, if in the case of an appearance that contain I call the precedi ng state of percep tion A and the following one E, then B can only follow A in appreh ension, but the percep tion A cannot fola

Object Object , Object


Section HI. Systema tic represen tation of all syntheti c prIl'lClj)les

low but only precede B. E.g., I see a ship driven perception of its positio n downst ream follows the percep tion upstream, and it is impossible that in the of this aDPe:u ance the ship should first be perceived downst ream and upstream. The order in the sequence of the percept ions in ap1pn~he:nsion is therefore here determ ined, and the appreh ension is the previous example of a house my perceptions rooftop and ended at the ground , but could also have ended above; likewise I could have appreh ended the manifold of empirical intuitio n from the right or from the left:. In the series of these perceptions there was therefo re no determ inate it necessary when I had to begin in the appreh ension in order to cornbme manifold empirically. But this rule is always to be found in the perception of that which happens, and it makes the order of that follow one anothe r (in the appreh ension of this necess ary. In our case I must therefo re derive the sequen ce apprehension from the objecti ve sequen ce of appearances, for otherw ise the former would be entirely undete rmined and no appearance be distinguished from any other. The former proves notblng about the connec tion of the manifold in the object, a because it is entirely arbitrary. This connec tion must therefo re consist in the order of the manifold of appearance in accordance with the ap]pn~henSJlOn of one thing (that which happens) follows that of the other cedes) in accord ance with a rule. Only thereby can I be lust]t1E~(1 ing of the appearance itself, and not merely of my ap]pn~hE~nslOI1, sequence is to be encoun tered in it, which is to say as much as not arrange the appreh ension otherwise than in exactly this sefmenc,~. In accordance with such a rule there must therefo re lie in in general precedes ::m occurre nce the conditi on{or a rule, in accordance with which this occurre nce always and necessarily follows; conversely, however, I cannot go back from the occurre nce (through apprehension) what precedes. For no appearance goes [Tom the following point of time to tlle precedi ng one, but it is merely to some preced ing point or other; on the the progress from a given time to the determ inately following one is necessary. Hence, since there is still sometll ing that follows, I must necessa rily relate it to someth ing else in general that precedes, and on it follows in accordance with a rule, i.e., necessarily, so that occurrence, as the conditi oned, yields a secure indicat ion of some cOJl1d.!tl()n, but it is the latter that determ ines the occurrence. If one were to suppose that nothin g precede d an occurre nce that it must follow in accordance with a rule, then all sequence of peJrception
, Object


A 193

B 239



Doctrine of Elements. Pt. H. Div. 1. Bk. H. Ch. H

B 240


B 24I


be determined solely in apprehension, i.e., merely subjectively, but it not thereby be objectively determined which of the perceptions must really be the preceding one and which the succeeding one. In this way we would have only a play of representations that would not be related to any object a at all, i.e., by means of our perception no appearance would be distinguished from any other as far as the temporal relation is concerned, since the succession in the apprehending is always the same, there is therefore nothing in the appearance that determines it so that a certain sequence is thereby made necessary as objective. I would therefore not say that in appearance two states follow one another, rather only that one apprehension follows the other, which is something merely subjective, and determines no object,b and thus cannot count as the cognition of any object (not even in the appearance). If, therefore, we experience that something happens, then we always presuppose that something else precedes it, which it follows in accordance with a rule. For without this I would not say of the object" that it follows, since the mere sequence in my apprehension, if it is not, by means of a rule, determined in relation to something preceding, does not justify any sequence in the object. d Therefore I always make my subjective synthesis (of apprehension) objective with respect to a rule in accordance with which the appearances in their sequence, i.e., as they occur, are determined through the preceding state, and only under this presupposition alone is the experience of something that happens even possible. To be sure, it seems as if this contradicts everything that has always been said about the course of the use of our understanding, according to which it is only through the perception and comparison of sequences of Huny occurrences on preceding appearances that we are led to discover a rule, in accordance with which certain occurrences always follow certain appearances, and are thereby first prompted to form the concept of cause. On such a footing this concept would be merely empirical, and the rule that it supplies, that everything that happens has a cause, would be just as contingent as the experience itself: its universality and necessity would then be merely feigned, and would have no true universal validity, since they would not be grounded a priori but only on induction. But the case is the same here as with other pure a priori representations (e.g., space and time) that we can extract as clear concepts from experience only because we have put them into experience, and


Object Object Object Objecte


Section HI. Systematic representation of all syJltbLetl.c prmclp1les

experience is hence first brought about through them. course series logical clarity of this representation of a rule determining possible if we have occurrences, as that of a concept of cause, is made use of it in experience, but a consideration it, as the condition of the synthetic unity of the appearances in time, was nevertheless ground of experience itself, and therefore preceded it a It is therefore important to show by an example even in experience we never ascribe sequence (of an occurrence, in which something happens that previously did not exist) to the a and distinguish it from the subjective sequence of our except when a rule is the ground that necessitates us to "hcp,-up order of the perceptions rather than another, indeed that it is this necessitation that first makes possible the representation of a succession in the object. b We have representations in us, of which we can also conscious. But let this consciousness reach as far and be as exact precise as one wants, there always remain only representations, i.e., inner determinations of our mind in this or that temporal relation. Now how do we come to posit an object C for these representations, or ascribe to their subjective reality, as modifications, some sort of objective reality? Objective significance cannot consist in the relation d to anotrler representation (of that which one would call the object), for would simply raise anew the question: How does this representation in turn go beyond itself and acquire objective significance in addition to the subjective significance that is proper to it as a determination of the state of mind? If we investigate what new characteristic is given to our sentations by the relatione to an object, and what is the dignity they thereby receive, we find that it does nothing beyond making combination of representations necessary in a certain'i\1ay, and sU!JJecting them to a rule; and conversely that objective significance is conferred on our representations only insofar as a certain order in temporal relation is necessary. In the synthesis of the appearances the manifold representations always follow one another. Now by this means no objectf at all is represented; since through this sequence, which is common to all apprehensions, nothing is distinguished from anything else. But as soon as I perceive or anticipate that there is in this sequence a relationg to the
., Object b Object Object






f Object


Doctrin e of Element s. Pt.

n. Div. 1. Ek. n. Ch. n

nce preced ing state, from which the represe ntation follows in accorda ing with a rule, I represe nt someth ing as an occurre nce, or as someth dea in time in place must I that object an cognize I i.e., that happens, otherbe cannot state, ng precedi the after which, , position termina te the wise assigned to it. Thus if! perceive that someth ing happen s, then s, first thing contain ed in this represe ntation is that someth ing precede a for it is just in relation to this that the appeara nce acquires its tempoit ral relation, that, namely, of existing after a precedi ng time in which this in n positio al tempor inate determ its acquire can did not. But it state relation throug h someth ing being presupp osed in the precedi ng from i.e., follows in accordance with a rule: on which it always I cannot reverse the series and place that that first, which it results, which it follows; and, second, that if the that to prior s which happen instate that precede s is posited, then this determ inate occurrence B 244 evitably and necessarily follows. Thereb y does it come about that there is an order among our represe ntation s, in which the present one (insoe, far as it has come to be) points to some precedi ng state as a correlat A 199 howis, which given, is that event this of , rmined to be sure still undete ever, determ inately related to the latter, as its consequence, and necessarily connec ted with it in the tempor al series. Now if it is a necessary law of our sensibility, thus a fonnal condines tion of an percept ions, that the preced ing time necessarily determi except time ng followi the at arrive cannot I that (in the following time by passing throug h the precedi ng one), then it is also an indispensable the law of the empiri cal repres entatio n of the tempor al series that g appearances of the past time determ ine every existence in the followin as time, and that these, as occurrences, do not take place except insofar the former determ ine their existence in time, i.e., establish it in accorally dance with a rule. For only in the appear ances can we empiric b of times. cogniz e this contin uity in the connec tion Unders tanding belongs to all experience and its possibility, and the the first thing that it does for L~is is not to make the represe ntation of posobject an of ntation represe the make to rather objects distinct, but on sible at all. Now this happens throug h its conferr ing tempor al order B 245 a as these, of each to ng assigni by ce existen their and the appearances preconseq uence, a place in time determ ined a priori in regard to the ceding appearances, withou t which it would not agree with time itself, which determ ines the positio n of all its parts a priori. Now this deterA200 apminatio n of positio n Cannot be borrow ed from the relation of the but, pearances to absolut e time (for that is not an object of perception), positions their ine determ must lves themse nces appeara the conversely,


Section In. Systematic representation of all synthetic principles

in time for each other, and make this determination in order necessary, i.e., that which follows or happens must succeed which was contained in the previous state in accordance with a general means of the rule, from which arises a series of appearances, in which understanding the very same order and constant connectionu in ries of possible perceptions is produced and made necessary as encountered a priori in the form of inner experience 111 perceptions would have to have their place. That something happens, therefore, is a perception that to a possible experience, which becomes actual ifI regard the PO:SltlOn appearance as determined in time, thus if I regard it as an can always be found in the connectionC of perceptions in accordance with a rule. This rule for determining something with respect to its temporal sequence, however, is that in what precedes, is to be encountered under which the occurrence always (i.e., nece~;sanl'y) follows. Thus the principle of sufficient reasond is the ground of ble experience, namely the objective cognition of appearances gard to their relation in the successive seriese of time. The ground of proof of this proposition, however, rests solely on following moments. To all empirical cognition there belongs the synthesis of the manifold through the imaginatiOli, is successive; i.e., the representations always follow each other in it. But the order of the sequence (what must precede and what must is not determined in the imagination at all, and the series of successive! resentations can be taken backwards just as well as forwards. But if synthesis is a synthesis of apprehension (of the manifold of a given appearance), then the order in the objectg is determined, or, to speak more precisely, there is therein an order of the successive synthesis that determines an object/] in accordance with which something would necessarily have to precede and, if this is posited, the other necessarily have to follow. If, therefore, my perception is to contain the cognition of an occurrence, namely that something actually happens, then it must be an empirical judgment in which one thinks that the sequence is determined, i.e., that it presupposes another appearance in time it follows necessarily or in accordance with a rule. Contrariwise, if I were to posit that which precedes and the occurrence not it nec-




del'Satz vom zureichcndcn Gnmdc

Erdmann, reading der

instead of der


Object Object


Doctrine of Elements. Pt. H. Div. T. Bk. H. Ch. n





essarily, then I have to it to a subjective of my imaginings, and if! represented something objective it I would have to call it a mere dream. Thus the relation of appearances (as possible perceptions) in accordance with which the existence of that which succeeds (what happens) is determined in time necessarily and in accordance with a rule something that precedes consequently the relation of cause to effect, is the condition of objective validity of our empirical judgments with regard to the series of perceptions, thus of their empirical truth, and therefore of experience. Hence the principle of the causal relation in the sequence of appearances is valid for all objects of experience the conditions of succession), since it is itself the ground of the possibility of such an experience. Here, however, there is a reservation that must be raised. The principle of causal connection among appearances is, in our formula, limited to the successiona of them, although in the use of this principle it turns out that it also applies to their accompaniment,b and cause and effect can be simultaneous. E.g., there is warmth in a room that is not to be encountered in the outside air. I look around for the cause, and find a heated stove. Now this, as the cause, is simultaneous with its effect, the warmth of the chamber; thus here there is no successionc in time between cause and effect, rather they are simultaneous, yet the law still holds. The majority of efficient causesd in nature are simultaneous with their effects, and the temporal sequence of the latter is occasioned only by the fact that the cause cannot achieve its entire effect in one instant. But in the instant in which the effect first arises, it is always simultaneous with tlLe causality of its cause, since if the cause had ceased to be an instant before then the effect would never have arisen. Here one must note that it is the order of time and not its lapse that is taken account of; the relation remains even if no time has elapsed. The time between the causality of the cause and its immediate effect can be vanishing (they can therefore be simultaneous), but the temporal relation of the one to the other still remains determinable. If I consider a ball that lies on a stuffed pillow and makes a dent in it as a cause, it is simultaneous with its effect. Yet I still distinguish the two by means of the temporal relation of the dynamical connection. For if I lay the ball on the pillow the dent follows its previously smooth shape; but if (for whatever reason) the pillow has a dent, a leaden ball does not follow it. The temporal sequence is accordingly the only empirical criterion of the effect in relatione to the causality of the cause that precedes it. The
Bcgleitl'I71j!;, here meaning simultaneous occurrence, as earlier at A I 83 lE 226.

Following the fourth edition, reading "Ursachen" instead of "Ursacbe."


Section HI. Systematic representation of all synthetic princllp!tos glass is the cause of rising of the water above its though both appearances are simultaneous. For as soon as I water into the glass from a larger vessel, something follows, the alteration of the horizontal state which the water had there into a concave state that it assumes in the glass. This causality leads to the concept of action, this to the of force, and thereby to the concept of substance. 75 Since I not crowd my critical project, which concerns solely the sources of synthetic ori cognition, with analyses that address merely the eluclClatlon amplification) of concepts, I leave the detailed discussion of these concepts to a future system of pure reason - especially since one can '1 I r,p'1 rh, find such an analysis in rich measure even in the familiar textbooks of this sort. Yet I cannot leave untouched the empirical criterion of a substance, insofar as it seems to manifest itself better and more through action than through the persistence of the appearance. Where there is action, consequently activity and force, there is also substance, and in this alone must the seat of this fruitful source of appearances be sought. That is quite well said; but if one what one understands by substance, and in so doing avoid a vicious circle, then the question is not so easily answered. How one infer directly from the action to the persistence of that acts, is yet such an essential and singular characteristic of the substance (phaenomenon)? Yet given what we have already said, the solution of the question is not subject to such a difficulty, though after the usual fashion ceeding merely analytically with its concepts) it would be pnt'irph, uble. Action already signifes the relation of the subject of causality to the effect. Now since all effect consists in that which happens, consequently in the changeable, which indicates succession in time, the ultimate subject of the changeable is tl1erefore that which persists,e.;,js the substratum of everything that changes, i.e., the substance. For according to the principle of causality actions are always the primary ground of all change of appearances, and therefore cannot lie in a subject that changes, since otherwise further actions and another subject, determines this change, would be required. Now on this account action, as a sufficient empirical criterion, proves substantiality without it necessary for me first to seek out its persistence through compared perceptions, a way in which the completeness that is requisite for the quantitya and strict universality of the concept could not be attained. For that the primary subject of the causality of all arising and perishing cannot itself arise and perish (in the field of appearances) is a certain inference, which leads to empirical necessity and persistence in existence, consequently to the concept of a substance as appearance.
, GrOfle
A 204

B 250


B 25 I



Doctrine of Elements. Pt. II. Div. I. Bk. II. Ch. n If something happens, the mere arising, regard to that comes to is already in itself an object investigation. It is already necessary to investigate the transition from the non-being of a state to this state, assuming that this state contained no in the appearance. This arising concerns, as was shown in section A,a not the substance that does not arise), but its state. It is therefore merely not an origination out of nothing. If this origination is alteration, regarded as the effect of a foreign cause, then it is called creation, which cannot be admitted as an occurrence among the appearances, for its possibility alone would already undermine the unity of experience, though if I consider things not as phenomena but rather as things in themselves as objects of mere understanding, then, though they are substances, they can be regarded as dependent for their existence on a foreign cause; which, however, would introduce entirely new meanings for the words and would not to appearances as possible objects of experience. Now in general anything can be altered, how it is possible that upon a state in one point of time an opposite one could follow in the next - of these we have a priori not the least concept. For this acquaintance with actual forces is required, which can only be given empirically, e.g., acquaintance with moving forces, or, what comes to the same thing, with certain successive appearances (as motions) which indicate such forces. But the form of such an alteration, the condition under which alone it, as the arising of another state, can occur (whatever the content, i.e., state, that is altered might be), consequently the succession of the states itself (that which has happened), can still be conof causality and the conditions of sidered a priori according to the time.* If a substance passes out of a state a into another state b, then the in time of the latter is different from the point in time of the first the apstate follows it. Likewise the second state as a reality pearance) is also distinguished from the first, in which it did not yet exist, as b is distinguished from zero; i.e., if the state b differs from the state a even in magnitude, then the alteration would be an arising of b-a, which did not exist in the prior state, and with regard to which the latter = 0.
* Note well that I am not talking about the alteration of certain relations b in

B 252


B 253


A 207 / B 252

general, but rather of the alteration of the state. Hence if a body is moved uniformly, then it does not alter its state (of motion) at all, although it does if its motion increases or diminishes.

That is,

in the "First Analogy."

b Rclatio77C71


Section IH. Systematic representation of all synthetic prinC"ipl{~s The question therefore arises, how a thing passes into another one = b. Between two instants there is between two states in those instances there is always a dlIel:elllce has a magnitude (for all parts of appearances are in turn magnimdes). Thus every transition from one state into in a time that is contained between two instants, determines the state from which the thing proceeds the state at which it arrives. Both are therefore boundaries of the time of an teration, consequently of the intermediate state between two states, and as such they belong to the whole alteration. Now every alteration has a cause, which manifests its causality in the entire time alteration proceeds. Thus this cause does not produce its alt,eratton denly (all at once or in an instant), but rather in a time increases from the initial instant a to its comrlletton nitude of the reality (b-a) is also generated through grees that are contained between the first and the last. alteration is therefore possible only through a continuous action of causality, insofar as it is uniform, is called a moment. The alteration does not conthem as effect. sist of these moments, but it is generated That is, now, the law of the continuity of an tlle of which is this: That neither time nor appearance in time consists passes est parts, and that nevertheless in its alteration the state through all these parts, as elements, to its second state. No difference of the real in appearance is the smallest, just as no difference in the magnitude of times is, and thus the new state of reality grows out of the first, in which it did not exist, through all the degrees of reality, the differences between which are an smaller than that between 0 and a. What utility this proposition may have in research into nature does not concern us here. But how such a proposition, which seems to amplify our cognition of nature so much, is possible a priori, very much requires our scmtiny, even though it is obvious it is real and correct, and one might therefore believe oneself to relieved of the question how it is possible. For there are so many presumptions of the amplification of our cognition tl1rough reason of them that it must be adopted as a general principle to be all and not to believe and accept even the clearest dOf,rmatic proof of this sort of proposition without documents that could provide a wellgrounded deduction. All growth of empirical cognitions and every advance in perception is nothing but an amplification of the determination of inner sense, i.e., a progress in time, whatever the objects may be, either appearances or pure intuitions. This progress in time determines everything, and is not itself determined by anything further: i.e., its parts are only in and given through the synthesis of it, but they are not given before it. For


A 209

B 255

A 2 IQ

Doctrine of Elements. Pt. II. Div. I. Bk. II. Ch. II


A2 I I

this reason every transition in perception to something that follows in time is a determination of time through the generation of this perception and, since that is always and in all its parts a magnitude, the generation of a perception as a magnitude through all degrees, of which none is the smallest, from zero to its determinate degree. It is from this that the possibility of cognizing a a law concerning the form of alterations becomes obvious. We anticipate only our own apprehension, the formal condition of which, since it is present in us prior to all given appearance, must surely be able to be cognized a priori. In the same way, then, that time is the sensible condition of the possibility of a continuous progress of that which exists to that which follows it, the understanding, by means of the unity of apperception, is the a priori condition of the possibility of a continuous determination of all positions for the appearances in this time, through the series of causes and effects, the former of which inevitably draw the existence of the latter after them and thereby make the empirical cognition of temporal relations (universally) valid for time, thus objectively valid.

C. Third Analogy. <Principle of simultaneity, according to the law of interaction, or community.>a the first edition:] All substances, insofar as they are simultaneous, stand in thoroughgoing community (i.e., interaction with one another). [In the second edition:]

<All substances, insofar as they can be perceived in space as

simultaneous, are in thoroughgoing interaction.>76 Proof


b<Things are simultaneous if in empirical intuition the perception of one can follow the perception of the other reciprocally (which in the temporal sequence of appearances, as has been shown in the case of the second principle, cannot happen). Thus I can direct my perception first to the moon and subsequently to the earth, or, conversely, first to the earth and then subsequently to the moon, and on this account, since the perceptions of these objects can follow each other reciprocally, I say that they exist simultaneously. Now simultaneity is the existence of the manifold at the same time. But one cannot perceive time itself and thereby
In the first edition: "Principle of community." This paragraph added in the second edition.


Section HI. Systematic representation of all sy:Jltrletlc pnnclj:>1es

derive from the fact things are positioned at same their perceptions can follow each other reciprocally. The ~"T.thl'~'~ the imagination in apprehension would therefore the other is these perceptions as one that is present in the subject not, and conversely, but not the objects a are simultaneous, i.e., if the one is then the other also is in the same time, that is necreessary in order for the perceptions to be able to succeed ciprocally. Consequently, a concept of the simultanereciprocal sequence of the determinations of these ously existing externally to each other is required in order to say that the reciprocal sequence of perceptions is grounded in b thereby to represent the simultaneity as objective. relation of substances in which the one contains deter'm:lIlatlcms ground of which is contained in the other is and, if the latter reciprocally contains the ground of the deter'mllnatlCil1S of the former, it is the relation of community or interaction. tlle simultaneity of substances in space cannot be cognized in experience otherwise than under the presupposition of an interaction among this is therefore also the condition of the possibility of the things themselves as objects of experience.> 'Things are simultaneous insofar as they exist at one same time. But how does one cognize that they exist at one and the same time? If the order in the synthesis of the apprehension of this manifold is indifferent, i.e., if it can proceed from A through E, C, and D to but also conversely from E to A. For if they existedd in time one the other (in the order that begins with A and ends at E), it would be impossible to begin the apprehension at the perception of E and proceed backwards to A, since A would belong to past and thus can no longer be an object of apprehension/,n Now if you assume that in a manifold of substances as appearances


A 2 II


Objecte Objecte , The text common to the two editions resumes here. dReading sie 71'ar'en instead of sie ware, so that the antecedent can be plural; even so, it remains unclear whether Kant intends the antecedent to be the "things" referred to at the beginning of the paragraph, or the representations A through E constituting the manifold. , In his copy of the first edition, Kant struck out the preceding paragraph and inserted the following note: "Space makes community possible. Now since the thinking being with all its faculties, whose effects belong merely to inner sense, is not a relation [Relation] of space, the co771771cniu771 of the soul with the body is therefore not comprehensible. The community of things in themselves must either have a third substance, in which they exist as accidentia and are in relation to one another - Spinozism - or, since this won't do, of possible community. If I it remains incomprehensible. Space is that are in me, the cognitive faculty of inner sense consider bodies merely may well stand in community with those of outer sense." (E LXXXVI, p. 34; 23:31-2)


Doctrin e of Elements. Pt. H. Div. T. Bk. II. Ch.

affect any be comple tely isolated, i.e., none each of them siother nor receive a reciprocal influence from it, then I say that their that and ion, multan eity would not be the object of a possible percept by the existence of the one could not lead to the existence of the other B 259 sepwere they any path of empirical synthesis. For if you though t that a comple tely empty space, then the percept ion that proceeds arated certain ly determ ine the existence from one to the other in time able of the latter by means of a succeeding percept ion, but would not be former the to distinguish whethe r that appearance objectively follows or is rather simultaneous with it. In additio n to the mere existence there must therefo re be something ely throug h which A determ ines the positio n of B in time, and convers this under which B does the same for A, since only also someth ing substances be empirically represe nted as existing those can on conditi determ ines the positio n of another in simulta neousl y. Now only ce time which is the cause of it or its determ ination s. Thus each substan a only with regard to its determinations) (since it can be a conseq uence s in must simultaneously contain Lh.e causality of certain determination must they the other and the effects of the causality of the other, i.e., sistand in dynamical commu nity (immediately or mediately) if their A2 I 3 everymultan eity is to be cognize d in any possible experience. But now the thing in regard to objects of experience is necessary withou t which necesis it experience of these objects itself would be impossible. Thus B 260 sary for all substances in appearance, insofar as they are simultaneous, to stand in thoroug hgoing commu nity interac tion with each other. The word "community"b is ambiguous in our language, and can C mean either communio or commcrcium. We use it here in the latter sense, as a dynamical community, withou t which even the local community (com7:1Zunio spatiz)d could never be empirically cognized. Fram our expein riences it is easy to notice that only continu ous influence in all places that light the one object to another, that space can lead our sense heavenly bodies effects a mediate comthe and eyes our n plays betwee munity betwee n us and the latter and thereby proves the simultaneity e of the latter, and that we cannot empirically alter any place (perceiv of ion percept this alteration) withou t matter everywhere making the can our positio n possible; and only by means of its reciprocal influence the even of it establish their simultaneity and thereby the coexistence ity commun t most distant objects (though only mediately). Withou

hip in a common , I.e., "commu nity" or "commer ce," thc former connotin g members g interaction connotin latter the parts, the among n interactio ly necessari not whole but order or relations hip among d "Commu nity of spaces," that is, a single spatial objects.


Section Ill. Systematic representation of all synthetic principles every perception (of appearance in space) is broken off from the and the chain of empirical representations, i.e., experience, to start entirely over with every new object a without the one being in the least connected or being able to stand in a relation with it. I do not in the least hereby mean to refute that may well exist where perceptions do not reach, and empirical cognition of simultaneity takes place; it is an object b for our possible experience at all. The following can serve as an elucidation. In our mind appearances, as contained in a possible experience, must stand in a comrnunity (co7lZ71zunio) of apperception, and insofar as the objects are to be represented as being connected by existing simultaneously, they must reciprocally determine their position in one time and thereby constitute a whole. If this subjective community is to rest on an objective ground, or is to be related to appearances as substances, then perception as ground, must make possible the perception of the other, and conversely, so that the succession that always exists in the perceptions, as apprehensions, will not be ascribed to the objects, c these can be represented as existing simultaneously. But this is a re<:ipro,:al ence, i.e., a real community (co7lZmercium) of substances, wlthcmt the empirical relation of simultaneity could not in experience. Through this commerce d the appearances, insofar as they stand outside one another and yet in connection, constitute a composite reale), and composites!of this sort are possible in many ways. Hence the three dynamical relations, from which others arise, are those of inherence, of consequence, and of composition.g
B 26r

A 215

B 262

These, then, are the three analogies of experience. are nothIng other than principles of the determination of the existence ances in time, in accordance with all three of its modi: of tion to time itself, as a magnitude (the magnitude of existence, i.e., duration); that of the relation in time, as a series (one after another); and finally that in time as a sum of all existence (simultaneous). This of time-determination is through and through dynamical, i.e., time is not regarded as that within which experience immediately determines

Object Object , Objecten

d C077lmcrciu771, printed as a German rather than Latin word.

printed as a German rather than Latin word.


Doctrine of Elements. Pt. H. Div. I. Bk. IT. Ch. II is impossible, since absolute time existence, the positio n of means of which appearances could be is not an object of percep tion held togethe r; rather the rule of the underst anding , throug h which temalone the existence of appearances can acquire synthet ic unity in a thus time, poral relations, determ ines the positio n of each of them in for each and every time. priori and empirical sense) we unders tand the combin ation of the (in By nature ry appeara nces as regards their existence, in accordance with necessa laws, certain re rules, i.e., in accorda nce with laws. There are therefo which first make a nature possible; the empirical and indeed a laws can only obtain and be found by means of experience, and indeed itin accord with its origina l laws, in accorda nce with which experience the exhibit self first become s possible. Our analogies therefo re really exunity of nature in the combin ation of all appearances under certain (insofar ponent s, which express nothing other than the relation of time apperception, as it compre hends all existence in itself) to the unity towhich can only obtain in synthesis in accorda nce with rules. Thus gether they say: All appearances lie in one nature, and must lie therein, desince withou t this a priori unity no unity of experience, thus also no termina tion of the objects in it, would be possible. About the method of proof, however, which we have employ ed in the ity, case of these transce ndental laws of nature, and about its singular precept a one remark is to be made, which must be very import ant as synfor every other attemp t to prove intellec tual and at the same time s analogie thetic a priori proposi tions. If we had wanted to prove these exists dogmatically, i.e., from concep ts - namely, that everyth ing that nce will only be encoun tered in that which persists; that every occurre accorin presupposes someth ing in the previous state, which it follows the dance with a rule; finally, that in the manifo ld that is simultaneous a to each other in accorda nce with a states are simulta neous in relation in rule (stand in commu nity) - then all effort would have been entirely existhe vain. For one cannot get from one object and its existence to tence of anothe r or its way of existing throug h mere concep ts of these us? things, no matter how much one analyzes them. So what is left for oball end The possibility of experience, as a cogniti on in which in the objects must be able to be given to us if their represe ntation is to have of form l jective reality for us. In this third thing, now, the essentia appearan which consists in the synthet ic unity of the apperce ption of y ances, we found a priori conditi ons of the thoroug hgoing and necessar even which t time-de termina tion of all existence in appearance, withou empiric al time-de termina tion would be impossible, and we found rules of synthet ic a priori unity by means of which we could anticipate expe-


Section Ill. Systematic representation of all synthetic principles

rience. In the absence of this method, and in the delusion to prove dogmatically synthetic propositions that understanding recommends as its principles,a a proof the of sufficient reason was often sought, but always in vain. No one ever even thought of the other two analogies, though one always tacitly emlP!()Ve:Q them,* since the due of the categories was missing, alone can uncover and make noticeable every gap of the understanding, in concepts as well as in principles.





The postulates of empirical thinking in general.7 8 with the formal conditions of experience accordance with intuition and concepts) is possible. 2. That which is connected b with the material of ence (of sensation) is actual. 3. That whose connectionC with the actual is determined in acc:oniaIlCe with general conditions of experience is (exists) necessarily.d

~Whatever agrees



* The unity of the world-whole, in which all appearances are to be connected, is obviously a mere conclusion from the tacitly assumed principle of the community of all substances that are simultaneous: for, were they isolated, tl1ey would not as parts constitute a whole, and were their connection (interaction of the manifold) not already necessary on account of simultaneity, then one could not infer from the latter, as a merely ideal relation, to the former, as a real one. Nevertheless we have shown, in its proper place, that community is really the ground of the possibility of an empirical cognition of coexistence, and that one therefore really only infers from the latter back to the former, as its condition.


I8/l3 265

The following notes are entered in Kant's copy of the first edition following A2 18: "The contingency of the alterable is only inferred from the fact that in accordance with the second analogy every state of its existence always requires a ground, and not vice versa, that it always requires a ground becanse it is contingent. We call absolutely contingent that which has no sufficient ground; never here, since it is never complete." (E LXXXVII, p. 35; 2303 2) "On possibility: That the concept of which can be given in a corresponding intuition is possible." (E LXXXVIII, p. 35; 23032) "'-\That can be thought indeterminately in any time [is possible]." (E LXXXIX, p. 35;
2n 2 )

"That which is determined in time [is actual]." (E XC, p. 36; 23:32) "That which is determined through the concept of time itself [is (exists) necessarily]." (E XCI, p. 36; 23032)


Doctrine of Elements. Pt. n. Div. I. Bk. n. Ch. H


Elucid ation

of The categories of modali ty have this peculiarity: as a determ ination ascribed are the object a they do not augme nt the concep t to which they b in the least, but rather express only the relation to the faculty of cogstill nition. If the concep t of a thing is already entirely comple te, I can if or, ask about this object whethe r it is merely possible, or also actual, s ination it is the latter, whethe r it is also necessary? No further determ is how asked: in the object' itself are hereby though t; rather, it is only unthe object itself (togeth er with all its determ ination s) related to the nt, judgme derstan ding and its empirical use, to the empirical power of and to reason (in its application to experience)? For this very reason the principles of modality are also nothing further than definitions of the concepts of possibility, actuality, and necesall sity in their empirical use, and thus at the same time restrictions of aland categories to merely empirical use, withou t any permis sion to lowance for their transce ndental use. For if the categories are not of have a merely logical significance and analytically express the form B 267 and y, thinkin g, but are to concer n things and their possibility, actualit necessity, then they must pertain to possible experience and its synthetic unity, in which alone objects of cogniti on are given. The postulate of the possibility of things thus requires that their conA220

" (E XCII, "That which is determin ed in time and space is actual. Against idealism. p. 36; 2303 2) That, however, "Everyth ing actual is necessary, either absolutely or hypothetically. for absolute continge ncy of things in themselves cannot be holds only of thought. " (E XCIII, p. 36; 23:32) ly deter"That which exists, thus in other things outside our thoughts , is thorough rCi7/i"imuJ ens an of concept the of principle the is on propositi mined. This [concept of the origin]. Whence the concept of [most real being] as conceptus this? of the absolute necessity s. This is the "Therein also belongs the propositi on that all negations are limitation 23:32-3) 36; p. XCIV, (E reason." of synthetic method accidents. "We do not attribute continge ncy to substances, but only to the alterable 23033) 36; p. XCv, Causes." (E e. The possi"The three criteria of hypotheses, always only in relation to experienc the hypothof behalf in up thought is which that of reality the is, hypothes bility of the esis. Its necessity must be certain." (E XCVI, p. 36; 23:33) a Objects s the frequent use of Verhiilmis rather b In this section, as in the precedin g, Kant continue a relation between the cognitive about speaking is he where here even than the usage of the faculty and its object rather than among objects, and thus by . Unless otherwise expected been have might term latter the " Aesthetic ndental "Transce . noted, our "relation " translates VerhiiltnZ:,_ , Objecte


Section Ill. Systematic representation of all synthetic principles

cept agree the formal conditions of an experience in gt:.llt:Jl'H. however, namely the objective fonn of experience in general, contains all synthesis that is requisite for the cognition of objects." A that includes a synthesis in it is to be held as empty, and does not to object, if this synthesis does not belong to experience, either as borro'we:d from it, in which case it is an empirical concept, or as one on a condition, experience in general (its fonn) rests, pure concept, which nevertheless belongs to experience, since its can be encountered only in the latter. For whence one derive the character of the possibility of an object that is means thetic a priori concept, if not from the synthesis constitutes of the empirical cognition of objects?C That in such a concept no contradiction must be contained is, to be sure, a necessary logical but it is far from sufficient for the objective reality of the concept, i.e., for the possibility of such an object as is thought through concept,79 in the concept of a figure that is enclosed between two straight lines is no contradiction, for the concepts of two straight lines and interrests not section contain no negation of a figure; rather the on the concept in itself, but on its construction in space, i.e., on the conthese in turn obditions of space and its detenninations; jective reality, i.e., they pertain to possible things, because they contain in themselves a priori the form of experience in general. \Ve shall now make obvious the extensive utility and influence of this postulate of possibility. If I represent to myself a thing that persists, so that everything that changes merely belongs to its states, I can never if cognize from such a concept alone that such a thing is possible. I represent something to myself that is so constituted if it is posited something else always and inevitably succeeds this may well be able to be so thought without contradiction; but such a canproperty (as causality) will be encountered in any possible not thereby be judged. Finally, I can represent various things stances) to myself that are so constituted that the state of one is followed by a consequence in the state of the other, and conversely; but whether such a relation can pertain to any things cannot derived Tom these concepts, which contain a merely arbitrary synthesis. Thus only from the fact that these concepts express a priori the relations of the perceptions in every experience does one cognize reality, i.e., their transcendental truth, and, to be sure, of experience, but yet not independently of all relationd to the form of
, Objecte Object , Objecte






Doctrine of Elements. Pt. H. Div. 1. Bk. H. Ch. H synthet ic unity in which alone objects an experience in general and can be empirically cognized. But if one wanted to make entirely new concepts of substances, of us, forces, and of interac tions from the materia l that percep tion offers itnce experie from tion connec withou t borrow ing the example of their for brain, the of ts figmen but end up with nothing self, then one be no indications at all, since in the possibility of which there their case one did not accept experience as instructress nor borrow the these concepts from it. Invente d concepts of this sort cannot acquire ns conditio as ies, categor the like charact er of their possibility a priori, given ones as ri, posterio a experience depends, but only on which d throug h experience itself, and their possibility must either be cognize was that ce substan A all. at d a posteriori and empirically or not cognize dipersistently present in space yet withou t filling it (like that interme inwould some which ate thing betwee n matter and thinkin g beings, the intuit to mind our of troduce),80 or a special fundam ental power to future (not merely, say, to deduce it), or, finally, a faculty of our mind dishow matter (no stand in a commu nity of though ts with other men entant they may be)81 - these are concep ts the possibility of which is its and nce experie in tirely groundless, because it cannot be ground ed s thought of ation known laws, and withou t this it is an arbitrar y combin to claim no make can that, althoug h it contains no contrad iction, still one objective reality, thus to the possibility of the sort of object that intrinsily evident is it ed, would here think. As far as reality is concern cally forbidd en to think it in concreto withou t getting help from experiof ence, because it can only pertain to sensation, as the matter can one that relation the experience, and does not concer n the form of always play with in fictions. a But I leave aside everyth ing the possibility of which can only be derived from actuality in experience, and conside r here only the possito bility of things throug h concep ts a priori, about which I proceed but s, concept such assert that it can never occur by itself solely from in always only as formal and objective conditi ons of an experience general. cogIt may look, to be sure, as if the possibility of a triangle could be experiof ndent indepe ly nized from its concep t in itself (it is certain t ence); for in fact we can give it an object entirely a priori, i.e., construc realways still would it it. But since this is only the form of an object, main only a produc t of the imagination, the possibility of whose object would still remain doubtful, as requiri ng someth ing more, namely that obsuch a figure be though t solely under those conditi ons on which all of n conditio priori a formal a is jects of experience rest. Now that space

B 270


B 271





Systematic representation of all sy:lltllLetlc princip'les

outer experiences, this very same formative synthesis means which we construct a figure in imagination is entirely that which we exercise in the apprehension of an appearance in order to make a concept of experience of it - it is this alone that connects with this concept the representation of the of such a thus the possibility of continuous magnitudes, even of maf!nlare tudes in general, since the concepts of conclear from the concepts themselves, but only from them as ditions of the determination of objects in experience in general; and where should one want to seek objects that correspond to the concepts, if not in the experience through which alone objects are given to us? - although without anticipating experience itself we can cognize and characterize the possibility of things solely in relation to the formal conditions under which something can be determined as an object in experience at all, thus fully a priori but only in relationb to these dons and within their boundaries. 82 The postulate for cognizing the actuality of things requires perception, thus sensation of which one is conscious - not immediate DerCE~D tion of the object itself the existence of which is to be cognized, its connection with some actual perception in accordance with the analogies of experience, which exhibit all real connection in an ence in general. In the mere concept of a thing no characteristic of its existence can be encountered at all. For even if this concept is so complete that it lacks nothing required for thinking of a thing with all its inner deltenmUlabut dons, still existence has nothing in the least to do with all of only with the question of whether such a thing is given to us in such a way that the perception of it could in any case precede For that the concept precedes the perception signifies its mere perception, which yields the material for the concept, is sole cn:lra,cteristic of actuality. However, one can also cognize the existence the thing prior to the perception of it, and therefore cognize it comparatively a priori, if only it is connected C with some perceptions in accordance with the principles of their empirical connection d analogies). For in that case the existence of the thing is still connected" with our perceptions in a possible experience, and with the guidance of the analogies we can get from our actual perceptions to the thing in series of possible perceptions. Thus we cognize the existence of a magnetic mat, bildrndc

B 272





hiingt . ..



Doctrine of Elements. Pt. H. Div. I. Ek. II. Ch.

bodies from the percept ion of attracte d iron filings, ter penetra ting this matter is impossible for us althoug h an immed iate percept ion in accordance with the laws of For given the constitu tion of our organs. we could also happen upon ions sensibility and the context of our percept nce if our senses, the experie an in it of the immediate empirical intuitio n e experience in possibl of form crudeness of which does not affect the er is appended whatev and ion general, were finer. Thus wherever percept reaches our too there , reaches to it in accordance with empirical laws experience, with hegin not do we If cogniti on of the existence of uJ,.ings. a of aption connec al empiric the of or proceed in accordance with laws B 274 pearances, then we are only making a vain display of wantin g to discover n or research the existence of any thing. b<However, a powerful objectio , idealism by made is ly mediate e against these rules for proving existenc the refutation of which belongs here.

Refuta tion of Idealis m83 Idealism (1 mean materi al idealism) is the theory that declares the exisintence of objects in space outside us to be either merely doubtful and the is fonner the ible; demon strable , or else false and imposs asal empiric one only s proble matic idealism of Descartes, who declare dogthe is latter the able; sertion (assertio), namely I aID, to be indubit the matic idealism of Berkeley, who declares space, togethe r with all somebe to on, conditi able things to which it is attache d as an insepar thing that is impossible in itself, and who therefore also declares things if in space to be merely imaginary.84 Dogma tic idealism is unavoidable themin things the to pertain to one regards space as a propert y that is selves; for then it, along with everything for which it serves as a condition, is a non-entity. The ground for this idealism, however, has been , underc ut by us in the Transcendental Aesthetic. Problem atic idealism our only es profess rather which does not assert anything about this, but B 275 of incapacity for proving an existence outside us from our own by means philoh thoroug a for riate immediate experience, is rational and approp nt sophical manne r of thought, allowing, namely, no decisive judgme must ed demand is that proof until a sufficient proof has been found. The therefo re establish that we have experie nce and not merely imagina prove can one unless lished tion of outer things, which cannot be accomp that even our inner experie nce, undoub ted by Descartes, is possible only under the presupposition of outer experience.

its proof and the The following sentence, the ensuing "Refutat ion of Idealism," and subseque nt remarks are all added in the second edition (B 274-9).


Section Ill. Systematic representation of all synthetic principl,es

The mere, but empirically determined, consciousness of my own existence proves the existence of objects in space outside me.

I am conscious of my existence as determined in time. All time-deterThis persismination presupposes something persistent in tent thing, however, cannot be something in me, since my own existence in time can first be determined only thr'ough is possible thing. a Thus the perception of this persistent through a thing outside me and not through the mere representation of a thing outside me. Consequently, the determination of my existence in time is possible only by means of the existence b of actual I perceive outside myself. Now consciousness in time is necessarily combined with the consciousness of the possibility of mination: Therefore it is also necessarily combined with the existence of the things outside me, as the condition of i.e., the consciousness of my own existence is at the same time an urlmedlate consciousness of the existence of other things outside me. Note 1. One will realize that in the preceding the game that idealism plays has with greater justice been turned against it. Idealism assumed that the only immediate experience is inner experience, and that from that outer things could only be inferred, as in case in which one infers from given effects to determinate causes, reliably, since the cause of the representations we perhaps ascribe to outer things can also lie in us. Yet here it is proved that outer experience is really immediate, * that only by means is possible not,
* The immediate consciousness of the existence of outer things is not presupposed but proved in the preceding theorem, whether we have insight into the possibility of this consciousness or not. The question about the latter would be whether we have only an inner sense but no outer one, rather merely outer imaginMion. Rut it is dear that in order for us even to imagine something as external, i.e., to exhibit it to sense in intuition, we must already have an outer sense, and by this means immediately distinguish the mere receptivity of an , According to the revised preface (Bxxxix), this sentence is to be replaced by the following: "This persistent thing, however, cannot be an intuition in me. For all grounds of determination of my existence that can be encountered in me are representations, and as such require something persistent that is distinct even from them, in relation to which their change, thus my existence in the time in which they change, can be determined." b Exiuenz , Existenz



B 276

B 2 77


I. Bk. H. Ch. H Doc trine of Elem ents . Pt. H. Div.

its determinaour own existence, consciousness sure, to Of course, the repr esen tatio n I am, tion in time, i.e., inne r experience. is that can accompany all thinking, whi ch expresses the consciousness a of a sub ject in itself, but existence that whi ch imm edia tely includes the empirical cognition, i.e., experinot thus not yet any cog niti on tdes the thou ght of som ethi ng exis ence; for to that ther e belongs, besi to rd rega in i.e., time, r ing, intu itio n, and in this case inne oute r objects are abch whi for , ned rmi whi ch the subject mus t be dete erience itself is con sequ entl y only solutely requisite, so that inne r exp 8s oute r experience. med iate and possible only thro ugh cog niti on in experience for the deNo te 2. All use of our faculty of completely. No t only can we perterm inat ion of time agrees with this thro ugh the change in oute r relations ceiveb an tim e-de term inat ion only of the persists in space (e.g., the mot ion (motion) relative to that anyhave earth);86 we do not even sun with regard to the objects on the B 278 as e, tanc subs base tt""le con cep t of a thin g persistent on which we could n draw not is even this persistence intu itio n, except mer ely mat ter, and ssary nece presupposed a priori as the from oute r experience, but rath er thus also as the determination of ion, con diti on of all time -det erm inat existence thro ugh the existence' of inne r sense in regard to our own myself in the repr esen tatio n I is no oute r things. The consciousness of llec tual repr esen tatio n of the selfintu itio n at all, but a 11lerely inte hen ce this I does not have the least activity of a thin king subject. And ent, could serve as the correlate for predicate of intu itio n that, as per sist as as, say, imp ene trab ility in matter, tim e-de term inat ion in inne r sense, 87 emp iric al intuition, does. tenced of oute r objects is required No te 3. Fro m the fact that the exis consciousness of our self it does not for the possibility of a dete rmi nate tatio n of oute r thin gs includes at the follow that every intuitive repr esen may well be the mer e effect of the same time thei r existence, for that as in delusions); but this is possible drea ms as ima gina tion h, of previous oute r perceptions, whic mer ely thro ugh the repr odu ctio n obr oute of thro ugh the actuality as has bee n shown, are possible only inne r experience in general is that only jects. Her e it had to be prov ed nce in general. liVhether this or that possible only thro ugh oute r experie


. For ity that characterizes every imagining oute r intu ition from the spon tane ty oi facul the te sense wou ld itsel f anni hila even mer ely to ima gine an oute r ion. inat imag ed thro ugh the intu ition , whic h is to be dete rmin



71.elm tf11" inste ad of "1'01'1 Following Erch-nann, reading ",pab7 Existenz r of this sentence. Existenz here and in the rema inde



Section Ill. Systematic representation of all synthetic principles putative experience is not mere imagination must be ascertained according to its particular determinations and through its coherence the criteria of all actual experience.

aFinally, as far as the third postulate is concerned, it pertains to necessity in existence, not the merely formal and logical necessity in the connection of concepts. 88 Now since no existenceb of objects of the senses can be cognized fully a priori, but always only a ori relative to another already given existence, but since even then we can only arrive at an existenceC must be cont:uuled somewhere in the nexus of experience of which the given perception is a part, the necessity of existenced can thus never be cognized from concepts but rather always only from the connection with that which is perceived, in accordance with general laws of experience. Now there is no existence that could be cognized as necessary under the of other given appearances except the existence of effects from given causes in accordance with laws of causality. Thus it is not the existence of things (substances) but of their state of which alone we can cognize the necessity, and moreover only from other states, which are given in perception, in accordance with empirical laws of causality. From it follows that the criterion of necessity lies solely in the law of possible experience that everything that happens is determined a priori through its cause in appearance. Hence we cognize only the necessity of effects in nature, the causes of which are given to us, and the mark of necessity in existence does not reach beyond the field of possible experience, even in this it does not hold of the existence' of things, as substances, since these can never be regarded as empirical effects, or as something that happens and arises. Necessity therefore concerns only the relations of appearances in accordance with the dynamical law of causality, and the possibility grounded upon it of inferring a priori from some given existence (a cause) to another existence (the effect). Everything that happens is hypothetically necessary; that is a principle that subjects alteration in the world to a law, i.e., a rule of necessary existence, with(mt which not even nature itself would obtain. Hence the "Nothing happens through a mere accident" (in mundo non datur casus)!


A 227

B 280


" The text common to the two editions resumes here. b EYistenz

Existenz , Existenz f In the world there is no chance.


Doctrine of Elements. Pt. H. Div. 1. Bk. H. Ch. n of nature; likewise the propos ition "No necessity in nais an a priori e ture is blind, but is rather conditioned, consequently comprehensibl which h throug sort the of laws are Both necessity" t.1}e play of alterations is subjected to a nature of things (as appearsame thing, to the unity of the understanding, in ances), or, what is of which alone they can belong to an experience, as the synthetic unity first The les. princip cal dynami the to belong appearances. Both of these is properl y a consequence of the principle of causality (under the analoy, gies of experience). The second belongs to the principles of modalit which, ty, necessi of t concep the ination determ causal which adds to the however, stands under a rule of underst anding . The principle of contiseries of appearances (alterations) (in in nuity forbade any any gap or cleft betwee n two appearalso but mundo non datur intuitio ns in space (non datur hiatus);' al empiric ances in the sum of all thus: "Nothi ng can enter experience ition for one can express the propos it as a part of empirical synthepermits or even that proves a on~ might think of outside of that void sis." For as far as concern s the this does not belong to the world), (the the field of possible experience only decides about queswhich , anding jurisdiction of the mere underst for empirical cognition, nces appeara tions concern ing the use of given beyond the sphere of a goes which and it is a problem for ideal reason, surroun ds and bounds what about judge possible experience and would ndental dialectic. transce the in red conside this, and must therefo re be itions (in 77lzmdo propos four these of order the We could easily represe nt fatum), in acdatur non casus, datur non saltus, non datur non datur les of tranprincip all like just ies, categor the the order of cordance practiced already the but , position its each show scendental origin, r, Howeve it. to due the r discove easily or this for himself reader win in g anythin permit not do they that this, in they are all united simply and empirical synthesis that could violate or infringe the understanding conits of unity the i.e., nces, appeara all of the continu ous connection!' all cepts. For it is in this alone that the unity of experience, in which e. possibl is place, perceptions must have their Wheth er the field of possibility is greater than the field that contains setg everyth ing actual, and whethe r the latter is in turn greater than the be of that which is necessary, are proper questions, and can, to be sure,

B 281


B 282


There is no fate. In the world there is no leap. , There is no hiatus. is different from the d Inserted in Kant's copy of the first edition: "The 36; 23:33) 77tctapbys;Cll117, in which there is no effect at all." (E XCVII, p. is no fate. there chance, no is there leap, no is there is no hiatus,



Section Ill. Systematic representation of all synthetic principles solved synthetically, though they also under reathings, as appearson alone; for they mean, roughly, to ask ances, belong together in the sum total and the context of a experience, of which each given perception is a which th(~ref01'e could not be combined with any other appearances, or whether my ceptions could belong to more than one possible experience to eXi)erlerlce general connection).a The understanding gives a general only the rule, in accordance with the subjective and conditions of sensibility as well as of apperception, which alone make it possible. Even were they possible, we could still not conceive of and make comprehensible other forms intuition (than space or other forms of understanding (than the discursive form that of cognition through concepts); and even if we could, are still not belong to experience, as the sole cognition in which given to us. Whether other perceptions those in general belong to our entire possible experience and therefore an entifiely the undelrst:lnciing, ent field of matter can obtain cannot be decided is given. which has to do only with the synthesis of that Otherwise the poverty of our usual inferences through which we (every forth a great realm of possibility, of which everything is very obvious. acject of experience) is only a small part, tual is possible" - from this there follows naturally, in accordance with the logical rules of conversion, the merely "Something possible is actual," which then seems to mean as much as "Much is possible that is not actuaL" It certainly looks as if one increase the number of that which is possible beyond that of the since something must be added to the former to constitute the But I do not acknowledge this addition to the possible. For that would have to be added to the possible would be All that can be added to my understanding is something beyond agreement with the formal conditions of experience, namely connection with some perception or other; but whatever is connected with this in accordance with empirical laws is actual, even if it is not im'mledjat(~ly De:rcE~iv(ed. However, that another series of appearances in thoroughgoing connection with that which is given to me in perception, thus more than a single all-encompassing experience, is possible, cannot be mf"en'ed that which is given, and even less without anything being given at for without matter b nothing at all can be thought. which is possible only under conditions that are themselves merely possible is not possible in all respects. But this is the way the question is taken when


A 2 3I



Stoff, i.e., matter as contrasted to form, rather than matter in a specifically physical


Doctrine of Elements. Pt. n. Div. 1. Bk. n. Ch. n one wants to possibility things extends further than experience can reach. 89 I have only mentioned these questions in order not to leave a gap in what according to common opinion belongs among the concepts of the understanding. In fact, however, absolute possibility (which is valid in every respect) is no mere concept of the understanding, and can in no way be of empirical use, rather it belongs solely to reason, which goes beyond all possible empirical use of the understanding. Hence we have had to satisfy ourselves here with a merely critical remark, but otherwise left the matter in obscurity pending further treatment later on. Since I now conclude this section, and with it at the same time the system of principles of the pure understanding, I must still provide the reason a why I have called the principles b of modality "postulates." I will not here take this expression in the significance that, contrary to the usage C of mathematics, to whom it nevertheless properly belongs, some recent philosophical writers 9 have uscd it, namely that postulation means the same as putting a proposition forth as immediately certain without justification or proof; for if we were to allow that synthetic propositions, no matter how evident they might be, could claim unconditional acceptance without any deduction, merely on their own claim, then critique of the understanding would be lost, and, since there is no Jack of audacious pretensions that common belief does not refuse (which is, however, no credential),d our understanding would therefore be open to every delusion, without being able to deny its approval to those claims that, though unjustifable, demand to be admitted as actual axioms in the very same confident tone. When, therefore, a determination is added a priori to the concept of a thing, then for such a proposition if not a proof then at least a deduction of the legitimacy of its assertion must unfailingly be supplied. The principles of modality are not, however, objective-synthetic, since the predicates of possibility, actuality, and necessity do not in the least augment the concept of which they are asserted in such a way as to add something to the representation of the object. But since they are nevertheless always synthetic, they are so only subjectively, i.e., they add to the concept of a thing (the real), ahout which they do not otherwise say anything, the cognitive power whence it arises and has its seat, so that, if it is merely connected in the understanding with the formal conditions of experience, its object is called possible; if it is in connectione with per-


A 2 33




, Sinn d Kreditiv


Section HI. Systematic representation of all synthetic prin(.:lples ception (sensation, as matter of the senses), mined means of the understanding, then the is if it is determined through the connectionb of perceptions in accordance with concepts, then the object is called necessary. The of modality therefore do not assert of a concept anything other than accion of the cognitive faculty through which it is Now in mathematics a postulate is the practical proposition contains nothing except the synthesis through which we first give ourselves an and generate its concept, e.g., to describe a circle with a given line a given point on a plane; and a proposition of this sort cannot be since the procedure that it demands is precisely first generate the concept of such a figure. Accordingly we can pm;tulate the principles of modality with the very same right, since they do not augment* their concept of things in general, but rather indicate way in which in general it is combined with the cognitive power. c


* Through

the actuality of a thing I certainly posit more than possibility, but not in the thing; for that can never contain more in actnality than what was contained in its complete possibility. But while possibility was merely a positing d of a thing in relatione to the understanding (to its empirical usc), actnality is at the same time its connection with perception.

Object b ZIIS'77117iIC77hfl77j!; e The following series of notes is inserted in Kant's copy of the first edition at A234-S, presumably constituting notes made for the "General Remark" that he adds at this in the second edition: "Now comes the proposition.: how are synthetic a pri07'i propositions possible." (E XCVIII, p. 37; 23:33) "Finally: How are synthetic a priori propositions possible through concepts, how are they possible through the construction of concepts?" (E XCIX, p. 37; 23:33) "On the possibility of an ars characteristica vel c0771binatoritf." (E C, p. 37; 23:33) "It is remarkable that for these postulates we must always have a mechanical medium[:] either a model as a string that lies, or the motion of this string around a point." (E Cl, p. 37; 23'33) "That all principles and synthetic a priori propositions in general do not go further than objects of experience, and that if we would still go beyond them then no intuition can correspond to them." (E cn, p. 38; 23:33-4) "That the pure laws of understanding also teach nothing further than the laws under which alone experience in general is possible, not the particular laws of the objects of experience. But that the laws of appearances (which are merely in us) thus have their seat and origin in the understanding, therefore also in us, is not to be marveled at. Indeed it is not possible to cognize a law with its necessity in such a way that we could have cognized it otherwise than in our own understanding. The chemical laws are not laws so much as rules of nature." (E CHI, p. 38; 23:34)



Doctrine of Elements. Pt. H. Div. 1. Bk. H. Ch. H


a<General Note on

System of Principles

of It is very remark able that we cannot have insight into thc possibility always must any thing in accorda nce with the mere categories, but we have available an intuitio n in order for it to display the objective reality of of the pure concep t of the underst anding . Take, e.g., the categories derelation. b How I) someth ing can exist only as subjec t, not as mere termina tion of other things, i.e., can be substan ce; or how 2) because be a someth ing is, someth ing else must be, thus how someth ing can one of ce if several things exist, from the existen cause at all; or 3) this in and the others follows and vice versa, of them someth ing cannot - insight into these substances can way a commu nity thing also holds of the same The all. at ts concep mere be had from or-her categories, e.g., how a thing can be one with a numbe r of others n is taken togethe r, i.e., be a magnit ude, etc. Thus as long as intuitio the lacking, one does not know whethe r one thinks an object' through fits categories, and whethe r there can ever be any objecrd that even ves themsel by not are ies categor the that ed confirm is them; and so it of cogniti ons, but mere forms of though t for making cogniti ons out proposiic synthet no that follows it way same the In given intuitio ns. B 289 existence there is tion can be made out mere categor ies - e.g., in and never as subject as only exist can that ing someth substance, i.e., mere predicate; or, everyth ing is a quantu m, etc. - if there is nothing conthat we can use in order to go beyond a given concep t and thereby nect it with another . Hence also no one has ever succeeded in proving a synthet ic propos ition merely from pure concepts of the understanda ing, e.g., the propos ition "Every conting ently existing thing has recause." One could never get further than to prove that withou t this at latione we could not compr ehend the existence! of the contingent unthe all, i.e., cognize the existence of such a thing a priori throug h the derstanding; from which, however, it does not follow that this is also look conditi on of the possibility of things themselves. Hence if one will aware become will one y, causalit of le princip the of back on our proof g ce: that we could prove it only of objects of possible experien and "Every tlling that happen s (every occurre nce) presupp oses a cause"; h indeed we could prove it only as a princip le of the possibility of expea

This note was added in its entirety in the second edition.

Relation , Object d Object

f Existenz,

and in the next clause.



Section Ill. Systematic representation of all



rience, hence of cognition of an object a given in enlpiiril:al proposition ition, and not from mere concepts. That contingent must have a cause" may be evident to everyone from mere concepts is not to be denied; but then the concept of the contingent is already taken in such a way that it contains, not the category of modality (as something, the non-existence of which can be thought), of relation b (as something that can only exist as the consequence of something else), and then it is, of course, an "VVhat can only exist as a consequence has its cause." In are to give examples of contingent existence, we ations and not merely to the possibility of the th()U~~ht site.*,91 Alteration, however, is an occurrence is possible as such only through a cause, the non-being of which is thus possible in and thus one cognizes contingency from the fact that can exist only as the effect of a cause; thus if a thing is to be contingent, it's an analytic proposition to say that it has a cause. It is even more remarkable, however, that in order to understand possibility of things in accordance with the categories, thus to esintutablish the objective reality of the latter, we do not merely itions, but always outer intuitions. If we take, e.g., the pure of relarion,c we find that I) in order to give something persists in in-. tuition, corresponding to the concept of substance to establish the objective reality of this concept), we need an intuition in space (of matter), since space alone persistently determines, while flows. 2) however, and thus everything that is in inner sense, In order to exhibit alteration as the intuition corresponding to the concept of causality, we must take motion, as alteration in space, as our example, indeed only by that means can we make alterations, the possibility of which cannot be comprehended any pure understand* One can easily think of the not-being of matter, but the ancients did not infer its contingency from that. And even the change from the being to the nonbeing of a given state of a thing, in which all alteration consists, does not prove the contingency of this state at all, as it were, from the actuality of its opposite; e.g., the rest of the body that follows its motion still does not prove the contingency of its motion just because the former is the opposite of the latter. For this opposite is here opposed to the other only logically, not realiter. In order to prove the contingency of the motion of the body, one would have to prove that instead of the motion in the preceding point of time, the could have been at rest then, not that it rests later; for in the later case the two opposites are perfectly consistent.
" Objects & RcI,1tiOll , Relation

B 290

B 291

B 290


n. Ch. n Doctrine of Elements. Pt. n. Div. I. Bk. con trad icto rily opposed is ing, intuitable. e thin g. Now how it sam and dete rmi nati ons in the existence e of the same thing stat an opp osed is possible that from a given state B 292 reason withble ensi preh e com sho uld follow not only can not be mad with out intucan not even be mad e und erst and able out an example, in space, the existence ion of a ition, and this intu itio n is the mot ns) sequence of opp osed determinatio of which in diff eren t places (as a e mak to y entl sequ sub er us; for in ord first makes alte rati on intu itab le to the as , time p gras to able be t we mus even inne r alterations ugh a line, and grasp the inner althro ely rativ figu e, sens r of inne thus grasp the drawing of this line (motion), and tera tion thro ugh oute r intuugh thro es stat t a diff eren successive existence of ourse1f in oses someupp pres on rati alte that all ition; the real gro und of whi ch is eived as perc be to ely mer er in ord thin g that persists in intu itio n, even inner in d foun be to n nt intu itio alteration, but ther e is no pers iste to not is nity mu com of the category sense. - Finally, the possibility of iposs not is it thus and e reason, be com preh end ed at all thro ugh mer out with t cep con this ctive reality ble to have insi ght into the obje n in space. For how would one conitio intu r oute ed intu itio n, and inde substances exist, the existence' of the ceiveb the possibility that if several the existence of the othe r (as an efone can follow reciprocally from som ethi ng in the former, there must fect), and thus that because ther e is in the othe r that can not be underon that acc oun t also be som ethi ng B 293 er alone? For this is requisite for stoo d from the existence of the latt ng things each of is not even com preh ensi ble amo community, its subsistence. ugh thro rs the othe whi ch is enti rely isolated from tances of the subs the to ity com mun Hen ce Leibniz, who ascribed a a divinity ded nee e, alon ing the und erst and as conceived wor ld rightly ity mun com tllis e alon tence for mediation; for from thei r exis 92 But we can read ily grasp the possiseemed to him incomprehensible. as appearances) if we represent them bility of com mun ity (of substances this already con tain s in itself apriin space, thus in oute r intu itio n. For (in diti ons of the possibility of the real ori formal oute r relations as con be y easil as just can It . ity) mun effect and countereffect, thus in com the thus and , des nitu mag as things established that the possibility of mag nitu de, can also be exhibited of gory cate the of objective reality by means of that alone can it subse only in oute r intu itio n, and that longg bein id avo to er ord in e. But que ntly also be applied to inne r sens this to the reader's furt her thought. of ples exam the e leav win ded I mus t orta nce , not only in ord er to conThi s enti re rem ark is of grea t imp

EX/:<tenz b df11I'C11 nce. , Existenz, used thro ugho ut this sente


Section 1II. Systematic representation of all syndletic pnnclp!les

firm our preceding refutation of idealism, even more, we come to talk of self-cognition from mere inner consciousness the determination of our nature without the assistance of outer empirical intuitions, to indicate to us the limits of the of such a cognition. 93 The final conclusion of this entire section is pure understanding are nothing further a possibility of experience, and all synthetic a priori propositions are reon lated to the latter alone, indeed their possibility itself rests this relation.>b

B 294


Doctrine ofthe Power ofJu dgme nt

phen omen a


traveled throug h the land of pure understandWe have now not d it, ing, and carefully inspect ed each part of it, but we have also surveye

additions and some As in thc first edition. For the second edition, Kant made extensive "Amphiboly of the the on appendix the to prior deletions in the body of this chapter up to the appenConcepts of Reflection." We will present each version of the chapter marginal pagiThe changed. not were that passages dix in its entirety, repeating those nation and notes will mark where the changes were made. chapter in Kant's copy of the first edition: b The following notes appear at the start of this a priori "Here is the question: How far does the possibility of synthetic cognition through merely ed determin is that s categorie through extend? If there is talk of a thing and yield no reason, hence also through categories, then such propositi ons are analytic, cognition." (E CIV, p. 38; 23:34) "1. On appearan ce and illusion. They consist of pure relations "2. How can one say that bodies are appearances. of these repre[lauter Rclatio71cn]; soul consists of pure [lauter] synthesis and analysis 23:34) 38; p. Cv, (E ce." intelligen sentations. The I is i70U771cn071; I as "(E CVI, p. 38; 23'34) "Being of sense - being of understanding; .'ensibilia 38; 23:34) p. (E them." cognize not "We can only think 170117710170, l being, "One must think things in themselves through the concept of a most-rea 23:34) 39; p. CVIII, (E since this excludes all experience." thought, but "i\fwuhr' pbOC71()771,0170'71 or a whole of substances in space may readily be 23:35) 39; p. CIX, (E not as 71OllJ71cn017, since they are isolated." the only thing "The same things as beings of sense or understa nding. I myself am 23:35) 39; p. that does not intuit itself." (E cx, to order intu"Categor ies do not serve to cognize things for themselves, but only 23:35) 39; p. CXI, (E ces." itions in space and time, i.e., appearan cognized "Until now one believed that through categories one actually already manifold the bringing for thought of forms only are they that something; now we see of intuition s to synthetic unity of appercep tion." (E CXlI, p. 39; 23:35)


Chapter HI. On the Ground of the Distinction <A>

and determined the place for each thing in it.9 4 This IS an island, and enclosed in unalterable boundaries nature It is the land of truth Ca charming name), surrounded a broad and stormy ocean, the true seat of illusion, where many a fog and melting iceberg pretend to be new lands and, ceaselessly deceiving empty hopes the voyager looking around for new discoveries, entwine him in adventures from which he can never escape and also never bring to an end. But before we venture out on this sea, to search through all its breadth and become certain of whether there is anytllin.gto hope for in it, it will be useful first to cast yet another glance at map of the land that we would now leave, and to ask, we could not be satisfied with what it contains, or even must be satlst:ted with it out of necessity, if there is no other ground on which we could build; and, second, what title we occupy even can it securely against all hostile claims. Although we adequately answered these questions in the course mary overview of their solutions can still strengthen unifying their various moments in one point. We have seen, namely, that everything that the understanding out of itself, without borrowing it from experience, it nevertheless solely for the sake of use in experience. The principles of pure standing, whether they are a priori constitutive (like the m~lthemlatlcal principles) or merely regulative (like the dynamical principles), contain nothing but only the pure schema, as it were, for possible experience; for the un,,1er'stand:mgthis has its unity only from tlle synthetic unity originally and from itself imparts to the synthesis of the imagination in relation to apperception, and in relationa to and agreement with the appearances, as data for possible cognition, must already stand a priori,b But now even if these rules of tlle understanding are not only true a priori but are rather even the source of all truth, i.e.,