New Anti-Kant by František Přihonský

1

New Anti-Kant

ACKNOWLEDGEMENTS

The translation of Přihonský’s New Anti-Kant is a project directed by

Professor Kevin Mulligan, Department of Philosophy at the University of

Geneva, as a part of the IRIS-Centre Lémanique project on formal con-

cepts.

I am very grateful to Professors Jan Sebestik and Edgar Morscher for

their generous scientific input. Their comments provided incentive as

well as information and any residual mistakes are mine alone. This book

owes much to Margaret Sletteland’s editorial skills and I thank her for

shaping the manuscript with style and precision.

František Přihonský

New Anti-Kant

edited & translated by Anita Kasabova

4

5

6

7

CONTENTS

Technical Remarks................................................................................. 8Editor’s Preface ................................................................................... 11Lexicological Issues ............................................................................. 14Edgar Morscher’s Introduction to the New Edition of the New Anti-Kant

(excerpt) .............................................................................................. 241. Who was Franz Přihonský? ........................................................... 26

1.1. A short biography .................................................................. 261.2. Přihonský’s relations with Bolzano ........................................ 271.3. Přihonský’s contributions to Bolzano’s complete scientific work

292. Přihonský’s New Anti-Kant .......................................................... 31

2.1. History .................................................................................. 313. I. Glossary – Přihonský-Bolzano ................................................... 35

3.1. I. A. German-English ............................................................. 353.2. I. B. English-German ............................................................. 39

4. II. Glossary – Kant (based on Guyer & Wood’s, Pluhar’s, and KempSmith’s translations of the Critique of Pure Reason)............................. 44

4.1. II. A. German-English............................................................ 444.2. II. B. English-German............................................................ 51

5. New Anti-Kant or examination of the Critique of Pure Reason

according to the concepts set down in Bolzano’s Theory of Science byDr. František Přihonský Bautzen, 1850................................................ 59

5.1. New Anti-Kant PREFACE .................................................... 675.2. New Anti-Kant INTRODUCTION ........................................ 715.3. New Anti-Kant TREATISE................................................... 865.4. New Anti-Kant APPENDIX ................................................ 246

REFERENCES................................................................................... 255INDEX .............................................................................................. 262

TECHNICAL REMARKS

A. The translation

(1) Brackets are used for translator’s additions to text and notes.To avoid confusion with translator’s additions to text, parenthesesappearing within parentheses in Přihonský’s text have not beenchanged to brackets.Quotations in Přihonský’s text are generally run in to the text.Přihonský’s page references may precede or follow a quotation.Spaced in dashes (related to paragraphing) have been preservedfrom the original text. The number sequence in bold and brackets indicates original pagi-nation.

(2) Use and mention: Přihonský uses spaced type for mentioning lin-guistic expressions (words and sentences) and logical notions (rep-resentations and propositions, as well as representations andpropositions as such), and two points for announcing them. Un-fortunately his procedure is rather inconsistent, since he frequentlyuses a linguistic expression for referring to either a word/sentenceor a representation/proposition without indicating its particular us-age. In the translation I use italics instead of spaced type for indi-cating the mention. In consulting Morscher’s edition of New Anti-

Kant, I have implemented the following notation for distinguishingbetween use and mention when it does not appear in the original.For example, on pp. 000–000 the mention of a term is indicated by‘ ’ (the word ‘truth’, the representation ‘truth’).

(3) Special colon use for mention: a colon may introduce words thatfit naturally into the grammar of the sentence (i.e., where a colonis not needed and would ordinarily be incorrect).

(4) I have replaced Přihonský’s abbreviations “K.” and “B.” by theproper names Kant and Bolzano.

(5) Přihonský uses ( ) to indicate his interpolations. Following Mor-scher, I add < > to indicate Přihonský’s interpolations in quota-tions, for example, “This function <(activity)>”, p. 00.

(6) For Přihonský’s notes, an asterisk precedes the note number.

(7) The translation employs two glossaries: one for Přihonský-Bolza-nian terminology and the other for Kantian terminology.

Editor’s Preface 9

(8) Přihonský refers to Hartenstein’s edition of the Critique of Pure

Reason. To facilitate the identification of Přihonský’s Kantianquotations, I have annotated the references to the first Critique asB (second edition, published in 1787), e.g., B39, and in one caseas A (first edition, published in 1781), using the Academy Edition

(vol. IV) of Kant’s writings, Kant's gesammelte Schriften, ed. Ber-lin Academy of Sciences (Berlin: Georg Reimer, 23 vols. 1902–1955, 9 vols. 1966–1979).

N.B. The translations of quotations from Kant’s Critique of Pure Reason

are my own, though I have consulted and contrasted the following Eng-lish translations: P. Guyer & A. Wood (Cambridge: Cambridge Univ.Press, 1998); W. Pluhar (Indianapolis: Hackett, 1996); N. Kemp Smith(London: Macmillan, 1929, 2nd impression 1933, 1990). In addition, Ihave consulted the following English translations of other works by Kant:Critique of Practical Reason, trans. L. White Beck (New York: Macmil-lan, 1956); Grounding for the Metaphysics of Morals, trans. J. Ellington(Indianapolis: Hackett, 1980); Prolegomena to any Future Metaphysics

that will be able to come forward as science, trans. G. Hatfield (Cam-bridge Univ. Press: Cambridge, 2002); The Jäsche Logic and Other Lec-

tures on Logic, trans. J. M. Young (Cambridge: Cambridge Univ. Press,1992).

B. The notes

1. ABBREVIATIONS

G – Grounding for the Metaphysics of Morals (1785)JL – Jäsche Logic (1800)KpV – Critique of Practical Reason (1788)KrV – Critique of Pure Reason (A, 1781; B, 1787)KU – Critique of the Faculty of Judging (1790)ML – Von der mathematischen Lehrart (1833–1841)MM – Miscellanea Mathematica, vols. 1, 3, 4, 6 (1803–1815)NAK – New Anti-Kant (1850)P – Prolegomena to any Future Metaphysics (1783)PU – Paradoxes of the Infinite (1851)RW – Lehrbuch der Religionswissenschaft (1834)WL – Theory of Science (1837)

2. According to contemporary usage, I use [ ] for indicating Přihonskýand Bolzano’s names for logical notions (representations andpropositions, as well as representations and propositions as such).

Editor’s Preface10

3. I refer to the following editions of Bolzano’s works: Bernard Bolzano Gesamtausgabe, hrsg. E. Winter, J. Berg, F.Kambartel, J. Louzil, B. van Rootselaar; F. Frommann-HolzboogVerlag, Stuttgart Bad-Cannstatt, comprising an Einleitung and 4series, I. Schriften, II. Nachlass, III. Briefwechsel, IV. Dokumente,1969 (BBGA). : Anti-Euklid (manuscript, ~ 1830, Kazimir Vecerka ed., Sbornik

1967, Prague, pp. 204–215). : Athanasia oder Gründe für die Unsterblichkeit der Seele. Ein Buch

für jeden Gebildeten, der hierüber zur Beruhigung gelangen will.Sulzbach 1827, 2nd edition, Sulzbach, 1838, reprint 1970. : Beyträge zu einer begründeteren Darstellung der Mathematik

(1810) in Bolzano’s Early Mathematical Works, CzechoslovakStudies in the History of Science, Prague, 1981. : Bolzano’s Wissenschaftslehre in einer Selbstanzeige, first pub-lished in 1949, in E. Winter: Leben und geistige Entwicklung desSozialethikers und Mathematikers Bernard Bolzano (1781–1848),Hallische Monographien, Halle. : Briefwechsel Bolzano-Exner, ed. E. Winter, B. Bolzano'sSchriften, vol. 4, published by the Königl.-Böhmische Gesellschaftder Wissenschaften, Prague, 1935. : Wissenschaftslehre, Versuch einer ausführlichen und grösstenteilsneuen Darstellung der Logik mit steter Rücksicht auf derenbisherige Bearbeiter. Hrsg. von mehreren seiner Freunde mit einerVorrede des Dr. J. Ch. A. Heinroth. 4 Bd. Sulzbach, 1837, reeditedin 4 vols. by W. Schultz, 2. Neudruck der 2. Auflage, Leipzig,1929; Lizenzausgabe Scientia Verlag, Aalen, 1981 (BBGA I, vols.11–14). : Paradoxien des Unendlichen. 1st edition F. Prihonsky ed., Leipzig1851; reedited, Meiner, Hamburg, 1921, new ed. B. van Root-selaar, 1975. : Was ist Philosophie (posthumous manuscript), ed. Rodopi (1969).

I have also consulted available English translations of Bolzano’s works;namely, Theory of Science, partial translations: R. George (London:Blackwell, 1972); B. Terrell (Dordrecht: Reidel, 1973); Paradoxes of the

Infinite, trans. D. A. Steele (London: Routledge, 1950); Appendix on the

Kantian Theory of the Construction of Concepts through Intuitions; andother (partial) translations of Bolzano’s early mathematical works by S.Russ in W. Ewald, From Kant to Hilbert, vol. 1 (Oxford: Clarendon Press,1996).

EDITOR’S PREFACE

This is the translation of an almost unknown text that appeared in 1850and has not been reprinted in the last 150 years.1 It was written by Fran-tišek Přihonský, a little-known nineteenth century Czech author living inthe Danube Monarchy, who wrote in German on a topic that has againgained in relevance after a century and a half: the critique of Kantianphilosophy. Přihonský was a pupil and friend of Bernard Bolzano, the“Bohemian Leibniz,” who taught religion at the German university inPrague until he was removed from office in 1820 and who is well-knowntoday for his extensive writings on mathematics, ethics, logic, and phi-losophy, with constant attention to earlier authors.2 In the New Anti-Kant

or examination of the Critique of Pure Reason according to the concepts

set down in Bolzano’s Theory of Science, Přihonský systematizes Bol-zano’s critiques of Kant. But he is not merely Bolzano’s spokesman, foralthough he examines Kantian philosophy from a Bolzanian point of viewand bases his objections on Bolzano’s texts, Přihonský evaluates Kant’sdoctrines by indicating the differences between Kant and Bolzano, and heoccasionally presents his own arguments against Kant.

Why read a New Anti-Kant today? If we consider this question from acontemporary angle, we may ask whether the problem is Kant’s Critique

or the critique of Kant’s Critique. In many universities Kant has nowbeen so successfully eradicated from the syllabus that his name merelyevokes the shadow of an unexamined abuse. Nevertheless, whilst nearlyeveryone has heard of Kant, almost no one has heard of an Anti-Kant. Tounderstand the purpose of a New Anti-Kant, however, one has to have atleast some idea of what the anti refers to. And since Kantian philosophywas dominant not only in Germany but nearly everywhere, philosophystudents should not be unacquainted with it. Otherwise the young gen-eration would be limited to examining its own opinion with regard toscience and be deprived of the benefit of studying the different points ofview pertaining to a domain of study. Interestingly, this point was madenot today but in 1798 by a member of the commission for educational

1 A new edition was prepared by H. Scholz and W. Dubislav in 1931, but it never

appeared. Edgar Morscher published the first new edition in 2003 (Academia Verlag, St.

Augustin).2 Bolzano’s main contribution to philosophy is entitled Theory of Science, attempt at a

detailed and in the main novel exposition of logic with constant attention to earlierauthors (1837).


reform in Austria, who argued that Kantian philosophy should be taughtin universities as an optional course.3 His appeal was rejected.

Why write an Anti-Kant? In the Danube Monarchy an anti-Kantiancampaign was launched less than a decade after the publication of theCritique of Pure Reason, a campaign which presupposed philosophicaland theological differences between Kant’s doctrines and the official lineof thought.4 Apparently, Kantian philosophy was considered a menace tothe political and religious state of affairs during the reign of Franz II, andthe commission for educational reform had the task of revising programsof study by limiting the development of ways of thinking that they saw aspotentially harmful to a regime which had restored the combined powerof church and state. Two such dangers were idealist and early liberalistthoughts, and they were epitomized by Kantian philosophy.5 So it washardly by chance that a series of Austrian Anti-Kants appeared between1788 and 1895, most of them written by Catholic priests, namely, Stat-tler, Horváth, Bolzano, Přihonský, and Brentano.6 For over a century

3 Cf. Eduard Winter (1968), who quotes Zippe, a spiritual counselor at the court of Franz

II (and a former rector of the Prague seminary) who objected to the suppression of

Kantian philosophy; cf. Romantismus, Restauration und Frühliberalismus imösterreichischen Vormärz, Wien, Europa Verlag, p. 23. Winter points out that this reform

caused many disputes between the friends of the Restoration and the (more enlightened)

friends of scientific research.4 Cf. Winter (op. cit.). Cf. also R. Bauer (1966), Der Idealismus und seine Gegner in

Oesterreich, Beihefte zum Euphorion, Zeitschrift für Literaturgeschichte, 3. Heft, Carl

Winter Universitätsverlag, Heidelberg. Bauer argues that there is a purely Austrian and

Catholic (Jesuit) tradition of anti-Kantism which, in his view, is a basic feature of

Catholic Austrian writings of the Vormärz (the period preceding the Bohemian

Enlightenment), which was set against the doctrines that appeared in Protestant Germany.

Kant’s critical theory, in particular, was not considered compatible with the metaphysical

bases of “Catholic orthodoxy”. Franz Přihonský’s New Anti-Kant is grounded in the same

Catholic position as the original Anti-Kant written by the Bavarian former Jesuit Benedikt

Stattler (1788). Another Catholic ancestor of Austrian Kant-criticism is the Abbott J. B.

Horváth, whose Declaratio Infirmitatis Fundamentorum Operis Kantiani « Critik derreinen Vernunft » appeared in Buda, in 1797. More than half a century later, Franz

Brentano pursued this line of Kant-criticism in Psychologie vom empirischen Standpunkt(1874) and in Die vier Phasen der Philosophie und ihr augenblicklicher Standpunkt(1895). Cf. Fr. W. v. Hoppenstedt (1933), Franz Brentanos Kantkritik, doctoral

dissertation, Kiel.5 Cf. Winter (1968), Romantismus, Restauration und Frühliberalismus im

österreichischen Vormärz, op. cit., pp. 19-27, where he discusses the impacts of the

Kaiser’s Catholic Restoration.6 As is often the case with those who intend to serve church and state with their best

intentions, Bolzano became a victim of this regime because in his sermons he promoted a

tolerant, liberal, and progressive way of thinking that was considered intolerable by the


these Catholic thinkers maintained the same line of objections to both thetheoretical and practical aspects of Kantian philosophy. It is not surpris-ing that Přihonský pays tribute to the first Anti-Kant, Stattler’s, by chris-tening his book New Anti-Kant.

The objections to Kant’s theoretical philosophy are directed againsthis view that epistemology underwrites metaphysics and logic. They con-cern the following points: (1) Kant distinguishes between phaenomena

and noumena and not only questions the reality of the existing andcreated world but even claims that we have no access to a nonsensoryreality. In addition, the Kantian assumption that there are things-in-them-selves about which we know nothing is contradictory. (2) Kant presup-poses the concepts of experience and necessity without defining them.7

(3) Kant claims that synthetic a priori judgments are grounded on pureintuitions and that these pure intuitions are space and time. (4) Kant in-troduces a division between philosophy and mathematics and further di-vides the doctrine of definition into the construction of mathematicalconcepts through intuitions and the exposition or analysis of philosophi-cal concepts. (5) Kant claims that the metaphysical deduction of the cate-gories shows their relation to the forms of judgment and theirindependence of experience. He then claims that the transcendental de-duction of the categories shows a) that the latter are a priori applicable toobjects of experience and b) that there are certain principles of the under-standing which are the conditions of possibility for experience itself.

The objections to Kant’s practical philosophy usually concern thefollowing points: (6) Kant proclaims the autonomy of the human mindand in this way puts man on an equal footing with God. (7) Kant ques-tions the reality of the world created by God. (8) Kant claims that thehuman mind can be its own legislator and that the categorical imperativeis the supreme moral law.8 (9) Kant claims that this moral law requires a

conservative political and ecclesiastic powers; and he not only lost his position at the

university but was almost excommunicated for criticizing the official theological manual.

Cf. Winter (1933), Bernard Bolzano und sein Kreis, (1949): Leben und geistigeEntwicklung des Sozialethikers und Mathematikers Bernard Bolzano, (1956) Derböhmische Vormärz in Briefen B. Bolzanos an F. Přihonský. Cf. also Winter 1968, op. cit.7 Cf. also Winter’s (1949) discussion of Bolzano’s questioning of Kant in the Miscellaniae

philosophico-theologicae vol.2 1806/12. Bolzano asks: how many judgments do we hold

with conviction without knowing their grounds and how do such convictions arise, viz

combinations of concepts we hold to be necessary?, op. cit., Halle, Niemeyer, p. 23.8 Cf. Bauer (1966, op. cit., pp. 42), Winter (1949, op. cit., pp. 21). N.B. Orthodox

religious thinkers (Solovyov, Losev) objected to Kant’s doctrines on the same grounds as

their Roman Catholic counterparts, but in Russia and other orthodox countries, there was

no campaign against Kantian philosophy, and Kant’s works were taught at the university.


free will that is unconditioned by the experiential world and that humanwill can only use its freedom insofar as it is determined by the moral law.(10) Kant claims that God’s existence must be proved in metaphysics andthat theological doctrines should not be part of metaphysics.

These are reasons for writing the critique of a Critique, but whyshould anyone read it? There is some irony in the fact that Přihonský’sAnti-Kant may turn out to be an Ante-Kant in more than one sense: First,Přihonský and Bolzano literally intended to be ante-Kant, in the sensethat they wanted to return German philosophy to its pre-Kantian, orLeibnizian, state. Second, this renewed interest in an active criticism ofKant inevitably returns us to reading (and thinking) about the philosophi-cal problems he posed. Third, Přihonský must have assumed that hisreaders were sufficiently well-acquainted with Kant’s texts to understandthe way he doctored his quotes from the Critique of Pure Reason. ThusPřihonský’s Anti-Kant takes us back to Kant. But in this critical reviewof a philosophy which is itself a critical review, we can discern a previewof future critics of philosophy, though they must wait their turn: namely,those scholars for whom the history of philosophy is philosophy itself;the present work is not written for them.9

LEXICOLOGICAL ISSUES

The first lexicological problem in translating a scientific text such as Při-honský’s New Anti-Kant is specifying the terminology. To avoid confu-sion between Kantian terminology on the one hand and Bolzanian termi-nology on the other, I have set up two separate glossaries. A second issueis to render the sense and the basic nature of the text adequately. Third,the index and my notes provide a lexicological key for showing the basicphilosophical problem posed by this text. As its title announces, two dif-ferent theoretical paradigms are confronted in the New Anti-Kant: Při-honský wants to replace the Kantian paradigm by the Bolzanian para- Cf. A. Losev (1923), “Философия имени” (first published in 1927), reprinted in Бытие,имя, космос, Mысль, Москва, 1993, pp. 858-860. Cf. also V. Solovyov who, although he

accepted Kant’s view that ethics is a discipline which is independent of metaphysics,

rejected Kant’s claim that God and the immortality of the soul are postulates, by arguing

that God's existence is the immediate content of a religious experience (i.e., it is not

deduced from a religious experience but is that which is experienced) and that God and

the soul are forces which directly create our moral reality; cf. Собрание Сочинения,

published posthumously (1911-1914), Просвещение, Петербург, reprinted in Брокхаус и

Ефрон, Т.2, Москва (1988), pp. 441-479.9 This is an adaptation of the second sentence of Kant’s preface to the Prolegomena

(1783), academy edition, p. 255.


digm, and he applies the latter to the former. This conflict, or the para-digm Přihonský is confronting, is traced through the index as well as re-constructed in my notes, and the examples below show the lexicologicalissues it raises:

1. The confrontation between the two theoretical paradigms is shownby the three authors’ use of two very different philosophical terminolo-gies. Přihonský and Bolzano base their theory, as well as their attack onKant, on the notion of truths as such, whereas Kant’s theoretical philoso-phy turns on the notion of synthetic a priori cognitions. Their respectiveterminological choices indicate their diverging strategies for explainingthe possibility of cognition (Erkenntnis), which includes determining therelation between thinking and cognizing and answering their respectiveepistemological questions: “what can I cognize?”10 and “what can I know(wissen)?”11 Přihonský and Bolzano claim that we can cognize that whichis true and that we can use this predicate for describing certain entities,such as judgments or beliefs, as having the property of truth.12 Kant, onthe other hand, holds that to obtain an objective perception or cognition,our concept of an object must be determined by an intuition. Or, to put itanother way: a cognition is produced when an intuition is combined witha concept in a judgment, and what makes this judgment true is our cogni-tion of an object, that is, the conformity of our concept to the object.13

10

What matters for Bolzano is that man has cognitive access to truths, including truths as

such, and in this sense he classifies the locutions knowing, cognizing, or being acquainted

with something as cognitions; cf. WLI, § 36.11

I follow Kant’s English/American translators (Guyer & Wood and Pluhar) in using ‘to

cognize’ and ‘to know’ for expressing the distinction between erkennen and wissen; and in

using ‘acquaintance’ for translating Kenntnis. N.B. Kant also uses the Latin term cognitiofor Erkenntnis; cf. KrV, A320/B377.12

Cf. WLI, § 19.13

“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 priori through reason.)”; cf. KrV,

preface, BXXVII. But the determination or cognition of an object we think of requires

intuition; cf. B166; cf. also B197, B351, B671. N.B. The Kantian schema for explaining

the possibility of our a priori cognition is a precursor of Husserl’s claim that we obtain a

cognition when an intentional act is confirmed or fulfilled by a perceptual act; cf. Edmund

Husserl (1901), Logical Investigations (Logische Untersuchungen, or LU), Tübingen: Max

Niemeyer, 1980; trans. J. N. Findlay (London: Routledge, 1970, reprinted 2002), I, §§ 3,

9, 19. For both Kant and Husserl, intuitive representations are either schematic or

symbolic; cf. KU, § 59; LUI, §§ 3, 18-20; LUVI, § 45. In addition, Husserl comes close to

Kant when he argues that linguistic acts have primacy over perceptual acts, because it is

the former that constitute objects: the inkpot or pen I see is constituted as an object and

becomes perceptually apparent to me when I name it. LUV, § 6; LUVI, § 4.


Kant’s account of how cognitions are produced is appropriately ex-pressed by Bolzano’s statement that “the possibility of cognizing(Erkennbarkeit) an object is the possibility of pronouncing a true judg-ment on it.”14 One might object that Přihonský-Bolzano and Kant havevery different accounts of cognition, since for Bolzano cognitions aretrue judgments because a cognition is a mental act which necessarily hasa true content, whereas for Kant all cognitions are judgments because acognition is necessarily a true mental act. I think, however, that this dif-ference is a superficial one, for Bolzano claims that all truths can be thecontents of possible cognitions and, consequently, that everything true isthinkable but not that everything thinkable is true. Kant would certainlyagree that we can think something without being able to cognize it.15 Thisis the backbone of his much-criticized view on things-in-themselveswhich are possible thoughts but not cognitions. Kant argues that I canthink whatever I like, as long as I don’t contradict myself, but to be ac-quainted with an object, I must be able to prove its possibility.16 Bolzanoin turn would accept Kant’s claim that the possibility of thinking (Denk-

barkeit) an object is grounded on the principle of contradiction. So itseems that there are intersections between the two conflicting paradigms.The real difference between these two accounts of the possibility of cog-nition lies in their authors’ diverging understanding of metaphysics: Při-honský and Bolzano postulate a possible discovery of truths: in order toexplain how a cognition is possible, we have to show that truths are ac-cessible to us. Kant examines the possibility of cognition itself: to ex-plain how a cognition is possible we have to show that it can have anobject.

2. A second lexicological problem concerns the theoretical and lin-guistic discrepancies in Přihonský’s, Bolzano’s, and Kant’s use of oneand the same term. Some of these discrepancies should be taken in alarger theoretical sense, since they are due to the conflict between twodifferent paradigms, such as the authors’ different uses of definition, ex-

perience, intuition, and logic.17

14

WLI, § 26.4; cf. also WLI, §§ 36-37. For Kant, on the other hand, a judgment is a

mediate cognition of an object, or an action by means of which given representations

become cognitions of an object; cf. KrV, B93.15

Perhaps the relation between what is thinkable and what is cognizable is a question of

degrees, for surely we must have some grasp of an object that we claim to be able to think.16

Cf. KrV, B XXVI (quoted in note 13 above).17

Cf. my note 127. The references to experience are listed in the index. Cf. also note 495

on Bolzano’s and Kant’s different ways of distinguishing between perception and

experience.


A definition, for Přihonský, is either an explication of concepts (Be-

griffsbestimmung) or, following Bolzano’s semiotics, an explication

(Verständigung) where a word is used to define something in such a waythat it modifies the reader’s understanding of the word.18 Bolzano’s ownnotion of definition as explication, however, is not an explication of con-cepts but an explication of the essential attributes of a thing.19 Kant criti-cizes this view, traditionally attributed to Aristotle, according to which adefinition is an analysis by genus and difference. He actually considers adefinition as an explication of concepts, and in his view such an explica-tion is either an analysis of concepts or a construction of concepts, and hedivides the doctrine of method in accord with these two kinds of defini-tions.20 Although Přihonský also uses the word definition in Kant’ssense,21 he successfully applies Bolzano’s notion of explication in hiscritique of the Kantian account of definition.22

Intuition is a problematic term for several reasons:

(i) Following Wolff, the Latin words intuitio and intuitivus translate An-

schauung and anschauende Erkenntnis. In contemporary English,however, the Latin derivative intuition is mainly used in a nonsci-entific sense: “to have a hunch.” It is used in a more scientificsense in “for you to have the intuition, that A, is just for it to seem

18

Cf. my notes 463-465 and 472-473, New Anti-Kant, p. 170, and WLVII, §§ 644, 653,

662, 668.2, 672. Bolzano’s notion of definition as explication is an interesting

contribution to current discussions in semiotics and communication theory. The references

to definition and explication are listed in the index.19

Cf. WLII, § 111, 209; WLIV, § 502. Cf. Dubislav (1931), op. cit., pp. 8, 134. Bolzano

says that we define an object that falls under a representation as such by naming the

essential attributes of this object and that a definition is a true assertion: “A has the

property b”. In addition, he claims (WLI, § 64.2) that the property-concept b refers to the

objects falling under the representation as such merely by means of the idea under which

we grasp them. But this claim is problematic, for how can the property-concept

[equiangular] be said to refer to an essential attribute of certain objects merely by means

of the representation [triangle] under which we grasp them?20

According to Kant, the philosophical method is analytic, for it cannot make use of a

concept until it is clarified or analyzed into its components, whereas the mathematical

method is synthetic, for it constructs or exhibits concepts by means of intuitions. Cf. my

notes 204, 463-473; and KrV, B758-760. Cf. also Dubislav (1931), Die Definition, pp. 2,

7-10, 114-115, 133-134.21

Cf. New Anti-Kant, p. 14, where Přihonský says that we define a representation by

stating its components.22

Cf. New Anti-Kant, pp. 167-171.


to you, that A.”23 In German, Anschauung means either to look atsomething or to have a view about something.24 So on the onehand, we have the German word Anschauung and its traditionalscientific translation, and on the other, we have completelydifferent uses of both words in German and English. This is notthe place to challenge the tradition, but as the translator of ascientific text, my duty to the reader is to point out this termino-logical problem.

(ii) Přihonský and Bolzano use only Anschauung, but Kant sometimesuses intuitus.25

(iii) Přihonský and Bolzano consider intuitions as data of experience andas immediate or noninferential cognitions. Unlike Kant they do notdistinguish between intuition, perception, and sensation.26

(iv) Kant considers intuitions as the data of experience and as the data ofreason. In the latter sense intuitions are a priori and a source ofevidence.27

23

George Bealer, “A Theory of the A Priori”, vol. 13, Philosophical Perspectives

(Oxford: Blackwell, 1999), pp. 29-55.24

An+schauen means “to look at” or “to contemplate”. (Intueri means: “to look into”.)

For this reason, the Bulgarian translator of the Critique of Pure Reason uses “наглед”; cf.

Критика на Чистия Разум, превод Цеко Торбов, София, Българска Академия на

Науките, 1967. But: “er hat sonderbare Anschauungen”; cf. Hermann Paul (1935),

Deutsches Wörterbuch, Niemeyer. The references to intuition are listed in the index.25

Following Wolff, Kant puts forward this Latin word in the Dohna-Wundlacken Logic

but modifies it for his own use-that is, for naming that which is not yet a cognition and for

referring to an intuitus purus when explaining how synthetic propositions are

constructions of concepts through intuitions. Cf. D-W Logic in Lectures on Logic, TheCambridge Edition of the Works of Immanuel Kant (Cambridge: Cambridge Univ. Press,

1992), pp. 752, 767; cf. also KrV, A320/B377.26

These are immediate cognitions which are intuitively self-evident or immediately

obvious; cf. WLIII, § 300.12. In this view (for which Bolzano and Přihonský criticize

Kant in regard to axioms), intuitions provide evidence, or the feeling of certainty and

necessity which accompanies a judgment; cf. WLIII, §§ 300.9, 313, 315.6, 316, note 1. Cf.

New Anti-Kant, pp. 174, and my notes 116, 162, 478, 480. N.B. But if intuitions provide

evidence, then they must be rational, since they present themselves as necessary, in the

sense that “it seems to me that if P, then ¬¬P ”; cf. Bealer (1999), op. cit, p. 3. This issue

visibly poses a problem for Bolzano, for in the epistemology (WLIII) he mentions a

“feeling for truth” through which we can come to know a truth without coming to know its

ground. In other words, although he admits that we can have a direct insight into

perceptual as well as conceptual truths, he does not admit that this insight or evidence is

the product of a (rational) intuition; cf. WLIII, § 316.3, 316, note 1.


Logic, for Přihonský and Bolzano, has a large and a narrow sense: takenin a large sense, logic is a theory of science and taken in a narrow sense,logic is a system of inference. But for Kant logic is the method for ob-taining a priori cognitions and synthetic a priori judgments which areobjectively valid. This is why he criticizes general or formal logic andintroduces a transcendental logic.

Other discrepancies in the use of one and the same word should betaken in a narrow linguistic sense. For example, Abfolge is used by Při-honský and Bolzano for designating the ground-consequence relation,whereas Kant uses Abfolge in the sense of succession. Also, Přihonský’sand Bolzano’s use of the expression πρωτον ψευδοζ is different. As onemight expect, both authors use it to criticize Kant, but Přihonský saysthat the πρωτον ψευδοζ of the Critique of Pure Reason is that the ob-jective validity of synthetic a priori judgments is based on pure intui-tions. Bolzano says that the πρωτον ψευδοζ of Kantian philosophy is theconfusion between a mental act, a concept, and the object of that con-cept.28

3. A third lexicological issue is Přihonský’s gradual replacement ofterms when quoting from the Critique which results in misinterpretations.For example, Ding, Gegenstand, and Objekt are distinct terms and dis-tinct concepts for Kant but not for Přihonský.29 Besides, Kant only usesDing in relation to things-in-themselves, so Přihonský falsifies his quo-tation of the Critique when referring to Kant’s Erscheinungsdinge.30

27

Here Kant can be brought into the contemporary discussion, since his account of a

priori intuitions comes close to Bealer’s theory of a priori knowledge, according to which

intuition or intellectual seeming is a primitive propositional attitude or concept

possession. We possess (or understand) a concept if and only if we can apply it to cases

presented either to pure thought or to phaenomenal experience. Our intuition is that which

intellectually seems to be so when we apply a concept to an object of experience: it seems

to me that it is possible to apply the concept “multigon” to a closed straight-sided plane

figure if and only if the property of being a multigon is the property of being a closed

straight-sided plane figure (Bealer, 1999, op. cit., p. 4). Bolzano’s Wahrheitsgefühl is such

a primitive propositional attitude, or an intellectual seeming-to-be-true, or an a prioriintuition.28

Cf. my note 202, New Anti-Kant, p. 69, and Bolzano in “Ueber den Begriff des

Schönen” (1843), in Mathematisch-physikalische und philosophische Schriften 1842-1843,

BBGA, vol. 18, p. 99.29

Cf. New Anti-Kant, pp. 87, 107, 139, 141, and my notes 251, 307, 391, 611. For Kant,

Gegenstände are objects of experience and they become objects for knowledge when the

manifold given in intuition is united with a concept. An object for knowledge through

concepts requires object of experience, so that it can be cognized; cf. KrV, B122, B136-

137, B197, B279.30

Cf. New Anti-Kant, p. 107, and my notes 307, 391, 611. See the index for appearance.


Similarly, Přihonský misinterprets the term schema when he calls it anintuition.31

There is a lexicological confusion between the concept-object rela-tions “contained under” (enthalten unter) and “falling under,” as well asbetween “falling under” and the property-concept relation “standing un-der,” since Přihonský uses stehen unter (or unterstehen) in regard to bothrelations.32 In addition, he slips into Kantian terminology by expressingthe concept-object relation as subsumieren unter (to fall under, to sub-sume)33 and by using the word construiren in paraphrasing Kant’s doc-trine of the schematism.34

The term Vorstellung is used by all three authors, but some confusionis caused by the differing English translations. Following Kant’s ownusage,35 his English translators use “representation,” whereas Bolzano’sEnglish translators use “idea.” These different translations reflect a se-mantic difficulty: strictly speaking, Vorstellung does not mean either“representation” or “idea”, despite the scientific tradition which usedthese terms synonymously.36 But to avoid this confusion and for lack of a

31

Cf. New Anti-Kant, p. 71, and my note 207 (the schema is a representation of the

method for providing a concept with its image, and it is a product of the imagination).32

Cf. New Anti-Kant, pp. 13, 29, 90, 100, and my note 110. N.B. Both authors also use

unterordnen for the relation of subordination between concepts.33

Cf. New Anti-Kant, pp. 81, 89.34

Ibid., cf. pp. 70-73, where Přihonský raises the polemical question of whether

geometrical concepts can be defined by constructing them through intuitions, although he

inaccurately quotes Kant as saying that an intuition is constructed by means of the

imagination (cf. KrV, B176-181, B746, and my notes 203 and 204). Kant uses the word

“construction” in B746 when he argues for the synthetic method of definition.35

Cf. KrV, A320/B376.36

Vorstellen literally means to put something in front of someone (vor etwas [jemand]

stellen), and in a figurative sense it is related to imagination (Vorstellungskraft).

Representation means either “to exhibit” or “to make present” or re+present: to “present

again”. Kant relates Vorstellung to repraesentatio for designating a mental (Gemüt) and

cognitive power which receives impressions and produces concepts and ideas (KrV,

A50/B74). So Kant uses the term representation as the generic term for ideas, concepts,

perceptions, intuitions, and sensations. The only similarity between Kant’s and Bolzano’s

use of the term Vorstellung is that neither of them limits it to imagination or images. For

Bolzano, a Vorstellung is distinct from ideas, concepts, images, sensations, and even

signs, although it comprises all of the above (cf. WLI, §§ 48.3, 50; WLIII, § 285, “Überden Begriff des Schönen”; BBGA, I.18, pp. 87-217), so “representation” seems suitable

for translating Vorstellung in Bolzano’s use of the word. He admits that it may sound

strange to speak of representations which no one represents (sich vorstellt), but he blames

this incongruity on the lack of an appropriate name for something that we inappropriately

associate with a mental change (WLI, § 50.3). Then again, he considers Vorstellungen as


better term, I use “representation” in translating Přihonský’s text and inmy Bolzanian quotations.

All three authors use the expression an sich, Kant with regard tothings and Přihonský and Bolzano with regard to representations andpropositions, but the English translations differ. I follow Kant’s transla-tors in translating Dinge an sich as “things-in-themselves” because thisexpression has been coined for so long in English that to change it nowwould confuse the reader. Bolzano’s translators use either “in itself”37 or“as such.”38 I use “as such” in my translation of Přihonský’s text andBolzano’s quotes because it more adequately renders the sense of an

sich.39

4. A problem which goes beyond the lexicological domain is how toconvey the style of the epoch and the author’s style in view of the con-temporary discussion. This problem is historical as well as linguistic:Přihonský’s nineteenth century German (with a Czech undertone) callsfor a contemporary translation so that it can be understood by twenty-first century readers. For example, I reduce an archaism such as mögen

doch diese Begriffe auf vielerlei Gegenstände sich beziehen to “theseconcepts may refer to various objects” (p. 104).

A more interesting example is provided by the table of contents whichcomprises the content of individual parts of the text in sentences ratherthan in abbreviations, since here the nineteenth century usage is accom-panied by a Czech undertone when Přihonský formulates rhetorical ques-tions in indirect discourse: “Wie falsch und schädlich die BehauptungKants sei, daß in philosophischen Untersuchungen genaue Definitionenund strenge Demonstrationen nicht stattfinden können”; or “Wie unrich-tig auch die Meinung, daß kein Begriff mit dem andern unmittelbar ver-

such as constituent parts of propositions (cf. WLI, § 50; ML, § 2.3) and as meanings of

signs, for a representation is that which is represented or signified by a sign (cf. WLIII, §

285).37

Cf. George’s (1972) and Terrell’s (1973) respective partial translations of the

Wissenschaftslehre and Rusnock’s (1998) translation of Von der mathematischen Lehrart.38

The early Jan Berg in Bolzano’s Logic (Stockholm: Almqvist, 1962) translates Satz an

sich as “proposition”, but later on he uses the expression “as such”; cf. “Bolzano the

Prescient Encyclopedist”, “Bolzano and Analytic Philosophy,” Grazer philosophischeStudien 53 (1997): pp. 13-33 (published by Rodopi).39

Taken literally, the combination of the preposition an with the reflexive pronoun sich is

closer to “in itself”, but “as such” comes closer to the Bolzanian notion of Satz an sichand Vorstellung an sich which refer to the meaning of the linguistic expressions Satz and

Vorstellung, respectively; cf. ML, §§ 1-2; WLI, §§ 19-20, 48-50.


knüpft werden könne.”40 In my translation I have maintained the strategy 40

The syntactic structure of these statements, in which the first assertion es wird gezeigt

is suppressed and in which wie falsch und schädlich or wie unrichtig are followed by a

subjunctive, is also characteristic of Slavonic languages. (In German, the subjunctive form

is expressed morphologically through the verb, whereas in Czech [and Bulgarian] it is

expressed syntactically, through a subordinate clause.) Přihonský did not write in Czech

because this art had all but disappeared under the Austrian regime. Nevertheless, his

German has Slavonic roots, which is why a literal translation of these statements renders

well-formed sentences in Czech or Bulgarian, but not in English or French. In Czech the

two statements would read: “Jak falešné i škodlivé je tvrzení Kantovo, podle kterého že

filosoficko zkoumáni nemajetný obsahovat přesné definice i přísné důkazy”; “Jak

nesprávné je také míněni, že žádný pojem nemůže být bezprostředně spojen s jiným”. For

historical and linguistic reasons (to put it briefly: Old Church Slavonic or Old Bulgarian is

the closest Slavonic form to Old Czech), Bulgarian syntax is also very close to

Přihonský’s, so the following translation into contemporary Bulgarian serves as a backup

test: “колко погрешно и вредно е твърдението на Кант, че във философските

изследвания не може да съществуват точни дефиниции и строги демонстрации”;

“Колко неправилно също така е мнението, че едно понятие не може непосредствено

да бъде свързано с друго”. Cf. S. E. Mann, Czech Historical Grammar (London: Athlone

Press, 1957), pp. 3-5 on the Czech language during the Austrian regime; pp. 19-23, on the


of transforming these locutions into sentences: “Kant’s claim that exactdefinitions and strict demonstrations cannot occur in philosophical in-vestigations is false and damaging”; “His opinion that no concept can beimmediately related to another is wrong”.

historical and linguistic relations between Slavonic languages. N.B. Two important

nineteenth century linguists of the modern Czech school had some connection with

Bolzano’s circle: Josef Dobrovský (who was Přihonský’s teacher at the Wendish Seminary

in Prague) and Josef Jungmann; cf. E. Winter (1932): Bernard Bolzano und sein Kreis,

Hegner, Leipzig; (1949) Leben und Entwicklung… (op. cit.), p. 11; (1956) Der böhmischeVormärz… (op. cit.), pp. 102, 113-115, pp. 22-28, 39-50, 93, 113-114; cf. also J. Loužil

(2000), “Bernard Bolzano, Josef Jungmann und die Anfänge der tschechischenNationalbewegung” in Bernard Bolzano und die Politik, Böhlau, Wien, pp. 181-200.

EDGAR MORSCHER’S INTRODUCTION TO THE NEW EDITION

OF THE NEW ANTI-KANT

(EXCERPT)

IN THE AREA OF CONFLICT BETWEEN KANT AND LEIBNIZ

A DEFINITION OF INTELLECTUAL POSITIONS

We define positions by relating them to other positions that are alreadyfamiliar to us or by introducing a coordinate system. Is this also valid forintellectual positions? Insofar as there is such a thing as a coordinatesystem for intellectual positions, if an agreement can be reached at all, itis only with much difficulty since intellectual positions tend to createtheir own coordinate systems and cancel existing ones. Hence intellec-tual quantities whose significance is beyond dispute, despite all contro-versies about the content of their doctrines, are still the most appropriateones for use as points of reference.

Bernard Bolzano’s position in the history of philosophy is defined byhis proximity to Leibniz and his distance from Kant. But greater or lesserdistance are not intellectual categories and can at best contribute figura-tively to the definition of intellectual positions. And even then it is stillnecessary to qualitatively fill intellectual relations with contents.

As far as the common ground of ideas and doctrinal contents is con-cerned, Bolzano’s distance from Kant was certainly just as great as hisproximity to Leibniz. In another sense, however, Bolzano was closer toKant since Bolzano had been preoccupied with Kant and Kant’s philoso-phy all his life, from his early youth. He had already studied the Critique

of Pure Reason when he was eighteen and subjected it to his first criticalexamination.41 This critical view of Kant is expressed not only in his dia-ries and notes but also in his first publications,42 as well as in many later

41

Bolzano, Lebensbeschreibung des Dr. B. Bolzano mit einigen seiner ungedruckten

Aufsätze und dem Bildnisse des Verfassers eingeleitet und erläutert von dem Herausgeber,

Sulzbach: J. E. v. Seidel, 1836, p. 23, Winter, Bernard Bolzano und sein Kreis. Dargestellt

mit erstmaliger Heranziehung der Nachlässe Bolzanos und seiner Freunde, Leipzig: Jakob

Hegner, 1933, p. 36.42

Bolzano, Betrachtungen über einige Gegenstände der Elementargeometrie, Prague: Karl

Barth, 1804, pp. V-II and 13; Bolzano, Beyträge zu einer begründeteren Darstellung der

Mathematik, Prague, Caspar Weidmann, 1810, pp. 135-152.

Edgar Morscher 25

ones.43 Bolzano quite obviously examined Kant’s philosophy more con-sistently and intensively than Leibniz’s. Kant gave him numerous sug-gestions for problems, but in solving them Bolzano usually took a differ-ent or an opposite way. For Bolzano, Kant was akin to an intellectualscratching post. This also constitutes a kind of intellectual proximity.

So no matter how much Bolzano was stimulated by Kant’s way ofputting problems, he did not agree with his way of solving them. Instead,he had much more in common with Leibniz as regards the solutions toproblems, especially in the domains of logic and metaphysics. HenceBolzano’s philosophy was occasionally apostrophized as “pre-Kantian.”44

This characterization was probably supposed to convey the impressionthat Bolzano’s philosophy was antiquated and that he was not concernedwith Kant or, anyway, that he did not understand him. This is certainlynot the case. Bolzano by no means “faked” his way past Kant but foughtto secure the “pre-Kantian” position in intensive debates with Kant’sdoctrines and in severely criticizing them. So Bolzano crossed thethreshold of Kant’s philosophy, to use Robert Zimmermann’s words,45

without stopping at it. Thus the epithet “anti-Kantian” is more appropri-ate for Bolzano’s philosophy and characterizes it more unequivocallythan “pre-Kantian.”

Bolzano agrees with Leibniz in his overall philosophical and scientificview. Like Leibniz, Bolzano is not only interested in scientific progressin the most diverse domains but also actively contributes to it through theresults of his own research—especially in mathematics, as well as inother domains, such as physics. In addition, as one of the last universalgenii with an extensive scientific education and varied scientificinterests, Bolzano is able to keep track, at least passively, of the scien-tific progress of his time in most fields.

Thus intellectual proximity and intellectual distance prove to be verycomplex categories. If they are not to remain empty or meaningless shellsof classification, they must be supplemented and replenished by qualita-

43

Cf., e.g., RW I, §§ 61-62 (pp. 151-165), and WLIII, § 305 (pp. 178-199), § 315 (pp.

238-258). “RW” is the usual abbreviation for Bolzano’s Lehrbuch der

Religionswissenschaft (1834) and “WL” for Bolzano’s Wissenschaftslehre (1837). In the

New Anti-Kant Přihonský makes full use of Bolzano’s Kant-criticism in RW and WL and

generally adopts much of it in a sometimes only slightly modified form.44

Anonymous (1828/1), Winter, Bernard Bolzano und sein Kreis, p. 223, Hirschgeber

(1a), p. 447ff. and p. 599. According to R. Zimmermann, “Philosophie und Philosophen in

Oesterreich” in: Oesterreichisch-Ungarische Revue. Neue Folge 6 (Vienna 1888-1889, pp.

177-198, 259-272), p. 198, Bolzano has “stopped a step before Kant.”45

However, R. Zimmermann did not mean Bolzano, op. cit., p. 198.

Introduction26

tive analyses, that is, by showing the common ground and the differencesas regards their content.

This is precisely the task that Franz Přihonský, Bolzano’s friend andstudent, sets himself in the two texts which appeared after Bolzano’sdeath. In the New Anti-Kant he examines Kant’s philosophy (particularlythe Critique of Pure Reason) from Bolzano’s point of view and accordingto the concepts Bolzano developed in the Theory of Science. In doing so,he shows the differences between Bolzano and Kant as regards content.In a text that is considerably shorter, although it appeared independently,“Doctrine of Atoms by the Blessed Bolzano,” Přihonský treats the prob-lem which most of all connects Bolzano with Leibniz’s philosophy, thedoctrine of monads or atoms. Both texts are published in this edition forthe first time since 1850 and 1857, respectively—and this time con-jointly. (This translation is a small excerpt of Morscher’s introduction,and it comprises only the first part on Přihonský and the history of theNew Anti-Kant).

In this introduction I will first briefly present Franz Přihonský, theauthor of the two texts. After that I will deal with the origin, content, andeditorial history of the New Anti-Kant.

1. Who was Franz Přihonský?

1.1. A short biography46

Franz Přihonský was born on 6 October 1788 in Prague, where he at-tended grammar school and completed the so-called philosophical stud-ies. This is where he met Bernard Bolzano, who was professor of relig-ious studies and preacher within the programme of philosophical studies.Then Přihonský turned to theological studies and joined the archbishop’sseminary. He was ordained in 1811. After that, he did spiritual welfarework for seven years (mainly as chaplain and one year as administrator)and stayed in touch with Bolzano. From 1818 to 1822 he was an assistantor—as it was called then—“adjunct at the chair of theoretical and practi-

46

Cf. Winter, Einführung in: Der böhmische Vormärz in Briefen B. Bolzano’s an F.

Přihonský (1828-1848). Beiträge zur deutsch-slawischen Wechselseitigkeit (Berlin:

Akademie-Verlag, 1956), pp. 1-106; Wurzbach, Biographisches Lexikons des Kaiserthums

Oesterrreich (Vienna: k.k. Hof- und Staatsdruckerei, 1872) (1), Schematismus (1) and

Handbuch (1); as well as personal notes of the bishopric’s counselor Dr. Siegfried Seifert,

former director of the cathedral’s archives at Bautzen. All quotes are taken from

Schematismus (1) for the years 1819-1823.

Edgar Morscher 27

cal philosophy.” In 1822 and 1823 he was also “supply teacher at thechair of aesthetics, the history of art, and the sciences and history ofphilosophy” at the University of Prague. From 1823 to 1824 he per-formed the duties of spiritual administrator and confessor for the Carme-lites at the Hradšin. During that time he completed his doctoral studiesand received his doctorate in philosophy on 1 June 1824. On a recom-mendation of the bishop of Bautzen, whose province it was, in September1824 Přihonský was appointed to the chair of the Wendish Seminary ofSt. Peter in Prague, where he was responsible for the ecclesiastic educa-tion of young Sorbs from Lusatia. From 1827 to 1839 he also held theoffice of co-director of the private bohemian educational and medicalsanatorium for the blind in Prague. Because of his commendableservices, he was named honorable canon of the chapter in Bautzen in1834. In the academic year of 1835–1836, he was dean of the philosophi-cal faculty at the University of Prague, without belonging to the faculty’steaching staff on that account (at the time the dean was elected from thecircle of “registered doctors and members of the philosophical faculty”for whom it was not corequisite to be professors). In 1839 he was ap-pointed canon of St. Peter’s monastery in Bautzen (“Budissin in the royalsaxon province of Upper Lusatia”) and moved there. From then on hewas an active member of the ecclesiastic chapter in Bautzen, at first ascanon and then from 1843 as capitular, later as scholastic, from 1854 ascantor, and from 1857 as senior of the ecclesiastic chapter. For hissixtieth birthday on 6 October 1848, some of his students founded theassociation “Dr. Přihonský’s Institute for Communicants.” The aim ofthis association was to educate boys who lived far from school.

On 12 January 1859 Přihonský died in Bautzen, at the age of 71. Hewas buried at the Nikolai cemetery at the church of St. Nikolai, of whichonly a ruin remains. A memorial stone for Přihonský can still be seen inthe wall of the cemetery.

1.2. Přihonský’s relations with Bolzano

Přihonský belonged to the inner circle of Bolzano’s students and friends.Although Bolzano had a very balanced and amicable nature, his relationswith his students and friends were subject to certain fluctuations. Thusoccasionally he not only apostrophized Prague as an “abderite,”47 in a hu-morous though definitely critical way, but referred to his students and

47

Bolzano, Wissenschaft und Religion im Vormärz. Der Briefwechsel Bernard Bolzanos

mit Michael Josef Fesl 1822-1848. Mit einer Einleitung von E. Winter. (Berlin:

Akademie-Verlag, 1965), pp. 299, 309, 310, 401.

Introduction28

friends who lived there (Franz Schneider, Johann August Zimmermann,occasionally Franz Přihonský, and even himself) as “abderites.”48 Thisterm was intended ironically and not affectionately, as is shown by hismany critical remarks against the three “Pragueians,” whom he alwaysaccused of idleness and lack of dedication to his cause.49 He included Při-honský in this rebuke and sometimes even referred to him explicitly.50

If the “Pragueians” achieved too little, according to Bolzano, his stu-dent Michael Fesl sometimes overdid things, rashly, imprudently, andinjudiciously. Many times a publication appeared too fast for Bolzano’staste or before it had received finishing touches; prefaces and notesturned out to be polemical; or a letter was sent off rashly—in the veryopposite of the “Pragueian” lethargy. For this reason, time and again Feslgot a lot of criticism from Bolzano.51 But Bolzano appreciated Fesl’sunlimited dedication and commitment,52 which is why he occasionallyregretted his criticism and affectionately called him “Michel” (or even“my dear Michel”).53

Such a cordial relationship never evolved between Bolzano and Při-honský54—in any case it never appeared so clearly in the correspondence(unlike with Fesl). But in the end, Přihonský’s merits could not be ig-nored. These included his intellectual abilities as well as his balancedjudgments and his increasing dedication to the common cause. Ulti-mately, these merits affected Bolzano’s judgment of him.

Still, overall, amongst his students and friends, Bolzano probably ap-preciated Přihonský the most. At least this is indicated in the passagesoccurring with increasing frequency in Bolzano’s correspondence with

48

Bolzano, Der böhmische Vormärz in Briefen B. Bolzano’s an F. Přihonský, p. 229;

Bolzano, Wissenschaft und Religion im Vormärz, pp. 310, 311, 391.49

Bolzano, Der böhmische Vormärz in Briefen B. Bolzano’s an F. Přihonský, pp. 150,

151; Bolzano, Wissenschaft und Religion im Vormärz, pp. 102, 168, 172, 188, 217, 227,

228, 300.50


Bolzano, Wissenschaft und Religion im Vormärz, pp. 229, 256f,m 278, 284, 295, 297,

322-323.52

Bolzano, Wissenschaft und Religion im Vormärz, pp. 121, 216, 226-227, 292, 378, 394.53


In his letters to Přihonský, however, Bolzano addressed him in an increasingly cordial

manner: he began with “Dearest friend!” or “Most esteemed friend!” which in 1839 he

changed to “My most beloved friend!” or “Beloved friend!”; in 1844 he turned this into

the constant address “Dearly beloved friend!” which he finally increased to “My dearly

beloved and most amiable friend!”; cf. Bolzano, Der böhmische Vormärz in Briefen B.

Bolzano’s an F. Přihonský.

Edgar Morscher 29

Fesl, in which he explicitly praises Přihonský.55 His appreciation appearsmost plainly in a letter Bolzano wrote to Přihonský on 10 March 1845:“for you are the only one of my friends who expresses himself quite can-didly and who at the same time is familiar with everything that concernsme and whose interest reaches a sufficient degree for penetrating the coreof the matter.”56

1.3. Přihonský’s contributions to Bolzano’s complete scientific work

Bolzano was not disappointed in either his judgment or his expectationsof Přihonský. During Bolzano’s lifetime, however, Přihonský did little tospread the latter’s ideas. Apart from the “Paper by a Catholic chaplain tothe author of two letters prompted by an article recently published inDresden: the pure Catholic doctrine” (Sulzbach, 1828),57 only one essay58

and eight reviews59 (three of them about Bolzano’s works), as well asseventy-seven entries in Pierer’s lexicon,60 made a significant difference.In addition, there is Přihonský’s preface to the “polemic” between Bol-

55

Bolzano, Wissenschaft und Religion im Vormärz, pp. 232, 284-285, 287, 354, 355, 371,

394, 492.56

Bolzano, Der böhmische Vormärz in Briefen B. Bolzano’s an F. Přihonský, p. 284.57

Přihonský, Schreiben eines katholischen Geistlichen an den Verfasser der zwey Briefe,

durch die jüngst zu Dresden erschienene Schrift: die reine katholische Lehre, veranlasst,

Sulzbach: Seidel, 1828.58

Přihonský, “Ueber den Begriff eines katholischen Dogma”, in: Freimüthige Blätter über

Theologie und Kirchenthum 17, Stuttgart, 1841, pp. 5-53.59

Přihonský, “Besprechung von Bolzano’s “Athanasia” (1827)” in: Jenaische Allgemeine

Literatur-Zeitung 28, no. 91-93, pp. 337-360; “Besprechung von Bolzano’s

“Religionswissenschaf”” in: Freimüthige Blätter über Theologie und Kirchenthum, 5,

Stuttgart, 1835, pp. 265-267; Besprechung von Bolzanos “Religjonsbekenntnisse zweier

Vernunftfreunde, nämlich eines protestantischen und eines katholischen Theologen

(1835)” in Freimüthige Blätter über Theologie und Kirchenthum 6, Stuttgart 1837, pp.

353-370; Besprechung von Johann Baptist v. Hirscher: Die christliche Moral, als Lehre

von der Verwirklichung des göttlichen Reiches in der Menschheit (Tübingen, 1836) in:

Jenaische allgemeine Literatur-Zeitung 34, vol. 4, no. 214-217, pp. 265-269; Die

Monadologie in: Leipziger Revue, Zeitschrift für Literatur, Kunst und Leben, 1847, no.

24, pp. 94-96; Besprechung von Fechners “Ueber das höchste Gut” (Leipzig, 1846), in:

Neue Jenaische allgemeine Literatur-Zeitung 6 (Leipzig, 1847), no. 228-229, pp. 912-915;

Besprechung von Zimmermann “Leibnitz’ Monadologie” (Vienna, Seidel, 1847), Neue

Jenaische allgemeine Literatur-Zeitung, 7, no. 61, pp. 241-243; and Besprechung von Karl

Ludolf Menzzer: Naturphilosophie (vol. 1, Halberstadt, 1847), in: Neue Jenaische

allgemeine Literatur-Zeitung, 7, no. 288-289, pp. 1149-1155.60

Přihonský’s entries in Pierer’s Encyclopädisches Wörterbuch der Wissenschaften,

Künste und Gewerbe, in Verbindung mit mehreren Gelehrten herausgegeben von Dr. A

.Binzer (Altenburg, Hahn, 1822, new edition 1824).

Introduction30

zano and Johann Anton Stoppani (1778–1836), which appeared in 1845under the title “On the Perfectibility of Catholicism.”61 The introductionto this book was originally written by Fesl,62 but Bolzano considered thisintroduction “quite unsuitable” in its content and its rambling and bom-bastic style.63 Therefore, he preferred Přihonský’s brief and factualpreface to Fesl’s well-meant but much too long-winded introduction.

After Bolzano’s death Přihonský fought unremittingly for the diffu-sion of Bolzano’s ideas: he not only wrote two obituaries for Bolzano64

but also posthumously published Bolzano’s Edification Speeches,65 theAbridged Manual of the Christian Catholic Religion,66 and the Three

Philosophical Treatises.67 But no doubt, his greatest editorial merit wasthe publication of the Paradoxes of the Infinite68 from Bolzano’s estate.Přihonský was personally involved in the development of the Para-

doxes,69 which for a long time remained Bolzano’s best-known and mostquoted work.70

61

Přihonský, Vorwort des Herausgebers. In: Bolzano’s Ueber die Perfectibilität des

Katholicismus. Streitschriften zweier katholischen Theologen; zugleich ein Beitrag zur

Aufhellung einiger wichtigen Begriffe aus Bolzano’s Religionswissenschaft (Leipzig,

Voss, 1845) (published anonymously).62

Fesl, Einleitung zu Bolzano’s: Ueber die Perfectibilität des Katholicismus, in Ganthaler

and Morscher: “Ueber Bolzano, Stoppani und die “Perfektibilität” des Katholizismus”, in:

Forschungsberichte und Mitteilungen des Forschungsintituts

Philosophie/Technik/Wirtschaft and der Universität Salzburg 15 (Salzburg, 1988), pp. 17-

75.63

Bolzano, Der böhmische Vormärz in Briefen B. Bolzano’s an F. Přihonský, p. 206.64

Přihonský, “Bolzano”, Bohemia 22 (Prague, 1849, semester 1, no. 40-41) and “Bernard

Bolzano, professor of the theory of religion at the University of Prague, born on 5 Oct.

1781, died on 18 Dec. 1848”. In: Neuer Nekrolog der Deutschen, 26 (Weimar, 1850), pp.

765-784.65

Přihonský wrote the preface of Bolzano’s Erbauungsreden (Prague, Wenzel Hess, 1849)

and was co-editor of the additional volumes 2, 3, and 4 of the Erbauungsreden.66

Bolzano, Kurzgefasstes Lehrbuch der katholisch-christlichen Religion, als der wahren

göttlichen Offenbarung (Bautzen, Weller, 1845).67

Bolzano, Drei philosophische Abhandlungen, welche auch von Nicht-philosophen sehr

wohl verstanden werden können, und vier akademische Reden von allgemein

menschlichem Interesse. Aus Dr. Bernard Bolzano’s schriftlichem Nachlasse (Leipzig,

Reclam, 1851) (BBGA, vol. 2A12/3), pp. 43-104.68

Dr. Bernard Bolzano’s Paradoxien des Unendlichen herausgegeben aus dem

schriftlichen Nachlasse des Verfassers von Dr. Fr. Přihonský (Leipzig, Reclam, 1851).69

Bolzano had already begun work on the Paradoxes in 1845 (cf. Bolzano, Der böhmische

Vormärz in Briefen B. Bolzano’s an F. Přihonský, p. 252. But the bigger part of this work

originated during Bolzano and Přihonský’s three-month stay at Liboch in the summer of

1847, with Přihonský’s active and even critical participation (Bolzano, Der böhmische

Edgar Morscher 31

But above all, after Bolzano’s death, Přihonský took the time to writetwo works on Bolzano's philosophy, and to complete and publish them.One of these had been planned in collaboration with Bolzano and wasalready in progress. It was here that Přihonský summarised Bolzano’sKant-criticism under the title New Anti-Kant. The other paper wasplanned as an essay for a journal and appeared as an independent publi-cation after the journal refused to accept it. This short paper deals withBolzano’s metaphysics under the title “Doctrine of Atoms by the BlessedBolzano.” These two works by Přihonský, which are not easily accessiblesince few copies have been preserved, are reprinted here for the firsttime. “. . .”

2. Přihonský’s New Anti-Kant

2.1. History

Bolzano repeatedly dedicated substantial passages and entire paragraphsto a debate with Kant. Thus, for instance, in the Contributions of 1810 hehad already added a separate “Appendix to the Kantian doctrine of theconstruction of concepts through intuitions.”71 And Bolzano dealt

Vormärz in Briefen B. Bolzano’s an F. Přihonský, p. 284). The Paradoxes were completed

in the summer of 1848 (Přihonský, preface to Paradoxien des Unendlichen [Leipzig,

Reclam, 1851], p. V; cf. also Bolzano, Wissenschaft und Religion im Vormärz, p. 405).70

Cf. Cantor, “Ueber unendliche, lineare Punktmannichfaltigkeiten”, 5. In:

Mathematische Annalen 21 (Leipzig, 1883), pp. 560-561 (cf. also p. 576); cf. also Cantor,

“Grundlagen einer allgemeinen Mannichfaltigkeitslehre. Ein mathematisch-

philosophischer Versuch in der Lehre des Unendlichen (Leipzig, Teubner, 1883), pp. 16-

17 (and p. 32), Gesammelte Abhandlungen mathematischen und philosophischen Inhalts.

(Berlin: Springer, 1932) pp. 179-180 (and p. 194); Dedekind, “Was ist und was sollen die

Zahlen?” (Braunschweig: Vieweg, 1888), p. 17, Dedekind (2nd ed., Braunschweig, 1893),

pp. XVI and 17; Bertrand Russell, The Principles of Mathematics (2nd ed., London: Allen

& Unwin, 1937), pp. 70, 201, 307, 357; Bertrand Russell, Introduction to Mathematical

Philosophy (London: Allen & Unwin, 1919), p. 138; Bertrand Russell, The Collected

Papers 1: Cambridge Essays 1888-1899 (London, Allen & Unwin, 1983), p. 363;

Brentano, Psychologie vom empirischen Standpunkt (Leipzig: Meiner, 1924 and 1925),

vol. 1, p. XLVII; Brentano, Auszug aus einem Brief an Oskar Kraus vom 31 Mai 1909, in:

Eduard Winter: Ausgewählte Schriften aus dem Nachlass, eingeleited und herausgegeben

von E. Morscher (Sankt Augustin: Akademia, 1993), p. 39.71

Bolzano, Beyträge zu einer begründeteren Darstellung der Mathematik (Prague:

Widtmann, 1810), pp. 135-152.

Introduction32

repeatedly with Kant’s doctrines72 in his main works—the Manual of the

Theory of Religion (1834) and the Theory of Science (1837)—in separateparagraphs. But, as in his debate with German idealism, which is scat-tered throughout many papers, Bolzano nowhere summarised his criti-cisms of Kant.

In the spring of 1837, Bolzano came across a French announcement ofa prize for a “presentation and evaluation of the Kantian system and allphilosophical systems that emerged from it.”73 In a letter to Přihonskýdated 18 June 1837, Bolzano first stated his interest in contributing a pa-per to this competition.74 The paper was to consist of two parts: a—preferably short—presentation and a detailed examination and criticismof the philosophical systems presented in the first part. It was this secondpart of the prize competition which really appealed to Bolzano, since itgave him an opportunity to introduce his own concepts and doctrines(especially on logic). Since the announcement left the participants tochoose between Latin and French, Bolzano suggested that the work bewritten in Latin (and that, in the first part, the Latin translation beaccompanied by the German original).75 Bolzano agreed to help inpreparing the extracts from the works to be presented and to draft anoutline of the critical part, but only in German. He wanted to persuadePřihonský to do the composition.76 In a letter to Fesl dated 7 August1837, Bolzano reported that he was going to work with Přihonský, whowas staying with him at the time. In this message to Fesl, Bolzano alsorequested assistance in preparing the bibliography.77

72

RWI, §§ 61 and 62 (pp. 151-165), WLIII, § 305 (pp. 178-199) and § 315 (pp. 238-258).

On the following presentation cf. also Winter, Leben und geistige Entwicklung des

Sozialethikers und Mathematikers Bernard Bolzano 1781-1848 (Halle: Niemeyer, 1949),

pp. 40-43.73

Bolzano Wissenschaft und Religion im Vormärz, p. 199 (the original says “Kantian

[systems]” instead of “the Kantian [system]”); for the following presentation cf. also

Winter, Leben und geistige Entwicklung des Sozialethikers und Mathematikers Bernard

Bolzano, op. cit., pp. 40-43.74

Bolzano, Der böhmische Vormärz in Briefen B. Bolzano’s an F. Přihonský, pp. 196-

197.75

Bolzano, Der böhmische Vormärz in Briefen B. Bolzano’s an F. Přihonský, pp. 197 and

198.76

Bolzano, Der böhmische Vormärz in Briefen B. Bolzano’s an F. Přihonský, pp. 196-197

and 198-199.77

Bolzano, Wissenschaft und Religion im Vormärz, p. 199.

Edgar Morscher 33

Although the “Parisian prize competition” or “prize essay” still cameup several times in the correspondence,78 by the end of 1837, Bolzanowas increasingly sceptical about whether it would be completed in time(by the end of 1838).79 Fesl mentioned the “Parisian prize question” in aletter written at the end of 1838,80 and in June 1839 he even reported thatit had been advertised again because the submitted papers were insuffi-cient.81 But meanwhile Bolzano had abandoned this plan. Instead Bolzanoand Přihonský agreed to deal critically with Kant’s philosophy and Ger-man idealism in particular papers, independently of the Parisian compe-tition. This had the advantage of sparing them the laborious and fruitlesstask of giving a detailed presentation of these philosophical systems andof translating relevant writings into Latin. Bolzano immediately sug-gested how to divide the work on the new project: Přihonský would bemainly responsible for the work on Kant,82 whilst he himself would takeon the debate with German Idealism.83 It was Bolzano’s explicit wish thatPřihonský publish the work on Kant under his own name84 and that hepreferably write it independently. Bolzano’s name, on the other hand,would be kept out as much as possible, although his concepts and viewsprovided the basis for the debate.85

Fesl frequently asked Bolzano about Přihonský’s Kant-criticism.86

The title “Anti-Kant”87 had already been mentioned by Fesl in 1840, and

78


203, 204-205.79


307; Bolzano, Wissenschaft und Religion im Vormärz, p. 220.80

Bolzano, Wissenschaft und Religion im Vormärz, p. 234.81



214-215, 219.83

Bolzano, Der böhmische Vormärz in Briefen B. Bolzano’s an F. Přihonský, p. 215;

Bolzano, Wissenschaft und Religion im Vormärz, pp. 228, 237, 260, 273, 274, 279, 281,

285. Bolzano never finished this work but only completed parts of it which were

published posthumously by Přihonský: Bolzano, Drei philosophische Abhandlungen

(1851), op. cit.84


240.85


214.86


Bolzano, Wissenschaft und Religion im Vormärz, p. 277.

Introduction34

later Bolzano adopted it in his letters to Přihonský.88 By February 1847,Přihonský’s work was ready for printing and—more importantly—obtained Bolzano’s “blessing” with the comment that it should appear theway it was because it was “good enough”.89 But Bolzano did not live tosee the publication of the book.

88

Bolzano, Der böhmische Vormärz in Briefen B. Bolzano’s an F. Přihonský, pp. 263-

269.89

Bolzano, Der böhmische Vormärz in Briefen B. Bolzano’s an F. Přihonský, p. 269. The

same judgment also applied to a second paper in which Přihonský criticized David

Friedrich Strauss; cf. Přihonský, “D. D. F. Strauss, in seiner christlichen Glaubenslehre

widerlegt ehe er geboren ward, durch die Ansichten des D. Bernard Bolzano”

(unpublished manuscript conserved in the archive of the cathedral chapter of St. Peter, in

Bautzen).

3. I. Glossary – Přihonský-Bolzano

3.1. I. A. German-English

A

Abfolge – ground-consequence relationAbleitbarkeit – deducibilityAbleitung – deductionallvollkommensten – most perfectan sich – as such (George [1972] and Terrell [1973] translate: in itself;Berg [1997] translates: as such)Annahme – assumptionannehmen – to assumeAnschaulichkeit – intuitivityAnschauung – intuitionauffassen – to graspauflösen – to dissolve, to decomposeAusdruck – expressionAussage – statement

B

Bedenken – doubts aboutBedeutung – meaningbedingt – conditionedBegriff – conceptBegriffsbestimmung – definition, explication of conceptsBegriffssatz – conceptual propositionBegriffswahrheit – conceptual truthbegründen – justifyingBegründung – grounding, justificationBehauptung – assertion, claimBeschaffenheit – attributeBestandteil – component, constituentBeweis – proofeinen Beweis führen – to establish a proof

D

Dasein – existence, beingDarstellung – presentation

Glossary36

dartun – to prove, to showDauer – durationDefinition – definitionDenkendes Wesen – thinking being

E

Eigenschaft – propertyEinwendung – objectionEinzelvorstellung – singular representationenthalten unter – contained under (relation: concept – object)Erfahrung – experienceerkennen – to cognizeErkenntnis – cognitionErkenntnistheorie – theory of knowledgeErklärung – definition, determination of conceptsErscheinung – phenomenonErscheinungsding – object of appearanceerschliessen – to infererweisen – to demonstrateerweislich – demonstrablees gibt, geben – there areEtwas – something

F

falsch – false (proposition)Folge – consequenceFolgerung – reasoningFürwahrhalten – taking to be true

G

Gegenstand – objectGegenständlichkeit – objectuality (George translates: reference)gegenstandslos – non-objectual (George translates: non-referring)Gegenstandslosigkeit – lack of objectualityGegenteil – contrarygegründet sein (auf etwas) – grounded on, based ongehören zu – belonging toGesetz – lawGewissmachung – confirmationGlückseligkeit – happinessGrund – ground

Anita Kasabova 37

Gründe – grounds, reasonsGrundsatz – principleGültigkeit – validity (George translates: satisfiability)

H

Herleitung – derivation (cf. Bolzano [1810]; I follow Russ [1996]. J. Laz[1993] uses deductio in the French translation)Hilfshypothese – supporting hypothesis

I

Idee – ideaInbegriff – collectionInhalt – content

K

Kenntnisse – cognitionsKraft – faculty, force

L

Lehrbuch – treatiseLehrsatz – theorem

M

Merkmal – markmittelbar – mediate

N

Nebenvorstellung – secondary representationNichtigkeit – vacuousness

P

Pflicht – dutyPflichtgebot – command of dutyPflichtsatz – deontic propositionPrüfung – test, examination

R

Raum – spaceRedensart – locutionrichtig – right (judgment)Richtigkeit – rightness

Glossary38

S

Satz – propositionSatz an sich – proposition as suchSchluss – inferenceSeelenkräfte – faculties of the soulseiendes – realSinn – senseSittengesetz – moral lawSprachgebrauch – usagestehen unter – standing under (relation: property-concept), falling under(relation:object-property or object-representation). N.B. Přihonský uses stehen

unter for both relations; cf. my note 110.subsumieren unter – falling under (relation: object-concept)

T

Teilvorstellung – part-representation (George translates: part-idea)Tugend – virtue

U

Übereinstimmung – accordUmfang – extensionunmittelbar – immediateunrichtig – wrong (judgment)Unterordnung – subordinationUnterscheidung – distinctionUnterschied – differenceunvermittelt – immediate, directUrsache – cause

V

veränderlich – variable, replaceableVerbindung – connectionVerhältnis – relationVerlässigkeit – reliabilityvermittelt – mediateVernunft – reasonVernunftwesen – rational beingVerständigung – explicationvoraussetzen – to presuppose

Anita Kasabova 39

Voraussetzung – presupposition, premissVorhandensein – presenceVorstellung – representation (George translates: idea)Vorstellung an sich – representation as such (George translates: objective

idea)

W

wahr – true (proposition)Wahrheit – truthWahrheit an sich – truth as such (George translates: objective truth)Wahrnehmung – perceptionWahrscheinlichkeit – probabilityWechselvorstellung – equivalent representationWeltweise – philosopherwirken – to actWirklichkeit, wirklich – actuality, reality; actual, real (George translates:reality, real)Wirkung – effectWissen – knowledgewissen – to knowWissenschaft – science

Z

Zeit – timeZeitdauer – temporal intervalZeitlänge – time spanzerlegen – to analysezukommen – belonging to, pertaining tozusammengesetzte Vorstellung – complex representationZuversicht – confidence

3.2. I. B. English-German

A

accord – Übereinstimmungto act – wirkenactual, actuality – Wirklichkeit, wirklichto analyse – zerlegenas such – an sich (George [1972] und Terrell [1973] übersetzen: in itself;Berg [1997] übersetzt: as such)assertion – Behauptung

Glossary40

to assume – annehmenassumption – Annahmeattribute – Beschaffenheit

B

belonging to – gehören zu

C

cause – Ursachecognition – Erkenntnisto cognize – erkennencollection – Inbegriffcommand of duty – Pflichtgebotcomplex representation – zusammengesetzte Vorstellungcomponent – Bestandteilconcept – Begriffconceptual proposition – Begriffssatzconceptual truth – Begriffswahrheitconditioned – bedingtconfidence – Zuversichtconfirmation – Gewissmachungconnection – Verbindungconsequence – Folgecontained under – enthalten unter (Verhältnis: Gegenstand-Begriff)content – Inhaltcontrary – Gegenteil

D

decompose – auflösendeducibility – Ableitbarkeitdeduction – Ableitungdefinition – Begriffsbestimmung, Definition, Erklärungdeontic proposition – Pflichtsatzderivation – Herleitung (cf. Bolzano [1810]; Russ [1996] übersetzt:derivation, aber J. Laz [1993] übersetzt: deductio)difference – Unterschieddirect – unmittelbar, unvermitteltdistinction – Unterscheidungdoubts about – Bedenkenduration – Dauerduty – Pflicht

Anita Kasabova 41

E

effect – Wirkungequivalent representation – Wechselvorstellungto establish a proof – einen Beweis führenexamination – Prüfungexistence – Daseinexperience – Erfahrungexplication – Verständigungexplication of concepts – Begriffsbestimmungexpression – Ausdruckextension – Umfang

F

faculty – Kraftfaculties of the soul – Seelenkräftefalling under – stehen unter (Verhältnis: Beschaffenheit–Begriff),subsumieren unter (Verhältnis: Begriff–Gegenstand)false – falsch (Satz)force – Kraft

G

ground – Grundground-consequence relation – Abfolgegrounded on – gegründet sein (auf etwas)grounding – Begründung

H

happiness – Glückseligkeit

I

idea – Ideeimmediate – unmittelbar, unvermitteltto infer – erschliesseninference – Schlussintuition – Anschauungintuitivity – Anschaulichkeit

J

justification – Begründung

Glossary42

justifying – begründen

K

to know – wissenknowledge – Wissen

L

lack of objectuality – Gegenstandslosigkeitlaw – Gesetzlocution – Redensart

M

mark – Merkmalmeaning – Bedeutungmediate – mittelbar, vermitteltmoral law – Sittengesetzmost perfect – allvollkommensten

N

non-objectual – gegenstandslos (George übersetzt: non-referring)

O

object – Gegenstandobject of appearance – Erscheinungsdingobjection – Einwendungobjectuality – Gegenständlichkeit (George übersetzt: reference)

P

part-representation – Teilvorstellung (George übersetzt: part-idea)perception – Wahrnehmungphenomenon – Erscheinungphilosopher – Weltweisepremiss – Voraussetzungpresence – Vorhandenseinpresentation – Darstellungpresuppose – voraussetzenpresupposition – Voraussetzungprinciple – Grundsatzprobability – Wahrscheinlichkeitproof – Beweisproperty – Eigenschaft

Anita Kasabova 43

proposition – Satzproposition as such – Satz an sichto prove – beweisen, dartun

R

rational being – Vernunftwesenthe real – seiendesreality, real – Wirklichkeit , wirklich (Georges Uebersetzung)reason – Vernunftreasoning – Folgerungreasons – Gründeto recognize – wiedererkennenrelation – Verhältnisreliability – Verlässigkeitreplaceable – veränderlichrepresentation – Vorstellung (George übersetzt: idea)representation as such – Vorstellung an sich (George übersetzt: objective

idea)right – richtig (Urteil)rightness – Richtigkeit

S

science – Wissenschaftsecondary representation – Nebenvorstellungsense – Sinnto show – dartunsingular representation – Einzelvorstellungsomething – Etwasspace – Raumstanding under – stehen unter (Verhältnis: Beschaffenheit–Begriff) N.B.Přihonskýverwendet stehen unter auch für das Verhältnis: Gegenstand-Beschaffenheit oder Gegenstand-Vorstellung, cf. meine Fussnote 14.statement – Aussagesubordination – Unterordnungsupporting hypothesis – Hilfshypothese

T

taking to be true – Fürwahrhaltentheorem – Lehrsatztheory of knowledge – Erkenntnistheorie

Glossary44

there are – es gibt, gebenthinking being – Denkendes Wesentime – Zeittime span – Zeitlängetreatise – Lehrbuchtrue – wahr (Satz)truth – Wahrheittruth as such – Wahrheit an sich (George übersetzt: objective truth)

U

usage – Sprachgebrauch

V

vacuousness – Nichtigkeitvalidity – Gültigkeit (George übersetzt: satisfiability)variable – veränderlichvirtue – Tugend

W

wrong – unrichtig (Urteil)

4. II. Glossary – Kant (based on Guyer & Wood’s, Pluhar’s, and

Kemp Smith’s translations of the Critique of Pure Reason)

4.1. II. A. German-English

A

Abfolge – successionableiten – to derive, to deduceAbsicht – aim, intention, regard, respectabsondern – to separateallgemein – universal, generalAllheit – totalityannehmen – to assumeanschauen – to intuitanschauend – intuitiveanschaulich – intuitiveAnschauung – intuitionan sich – in itselfAnspruch – claim, entitlement

Anita Kasabova 45

anzeigen – to indicate, to point outArt – species, kind, way, mannerauflösen – to dissolveAusdehnung – extensionAusdruck – expression

B

Bedeutung – meaningbedingt – conditionedBedingung – conditionbegrenzen – to boundBegriff – conceptbeilegen – to ascribeBenennung – name, designationBestandteil – componentbestätigen – to confirmbestehen – to consist, to holdbestimmen – to determineBewegung – motionBewegungsgrund – motiveBeweis – proofbewirken – to bring aboutBewusstsein – consciousness, awarenessBezeichnung – designationBeziehung – relationBild – image

D

Darstellung – presentation, exhibitiondartun – demonstrateDasein – existenceDefinition – definitionDenken – thinkingdeutlich – distinct, clearDing an sich – thing-in-itselfdunkel – obscure

E

Eigenschaft – propertyEinbildungs(kraft) – (faculty of) imaginationEindruck – impression

Glossary46

Einheit – unity, unitEinteilung – divisionEmpfänglichkeit – receptivityEmpfindung – sensationEndabsicht – final purposeendlich – finite, finallyendlos – endlessEndursache – final causeEndzweck – final purposeentstehen – to ariseErfahrung – experienceerkennen – to cognizeErkenntnis – cognitionErklärung – explanation, definitionErörterung – expositionErscheinung – appearanceEvidenz – (self)-evidence

F

Fähigkeit – capacityfalsch – falseFolge – sequencefolgern – to conclude, to inferFolgerung – consequenceFührwahrhalten – taking to be true

G

Ganze(s) – wholeGattung – genusGebiet – domainGebot – commandGegenstand – object (of experience)Geist – mind, spiritdes Geistes – intellectualGemüt – mindGemüts – mentalGesetz – lawGesichtspunkt – viewpointGestalt – shapeGlaube – belief, faithGlückseligkeit – happiness

Anita Kasabova 47

Grenze – boundaryGrösse – magnitudeGrund – ground, basisGrundsatz – principlegültig – valid

H

Handlung – act, actionherleiten – to derivehervorbringen – to produce

I

Ich (das) – the IIdee – ideaInbegriff – sum totalInhalt – content

K

kennen – to be acquainted withKenntnis – acquaintanceKraft – faculty, force, power

L

Lehre – doctrineLehrsatz – theorem

M

mannigfaltig(e) – manifold (adj.)Mannigfaltige – manifold (n.)meinen – to opine, to thinkMeinung – opinionMerkmal – markMittel – meansMoment – moment

N

nach – in accordance with, afterNaturbetrachtung – study of nature (following Kemp Smith [1929];Guyer & Wood [1998] translate: consideration of nature, and Pluhar[1996] translates: contemplation of nature)nichtig – vacuous, nugatory

Glossary48

Noogonie – noogony (Kant’s name for Locke’s system, for by its meanssensibility is produced (γον) by reason (νοΰς)Noologisten – noologists (Kant’s name for those who derive pure rationalcognitionfrom reason alone, i.e., Plato and Leibniz)Noumenon – noumenonNutzen – benefit, usefulness

O

Obersatz – major premisesObjekt – object (for knowledge)Ort – place, location

P

Perzeption – perceptionPflicht – dutyPhaenomenon – phenomenonPostulat – postulatePrinzip – principleprüfen – to examine, to test

R

Raum – spaceRegel – ruleReihe – seriesReihenfolge – sequencerein – pure (adj.)Relation – relationrichtig – correct

S

Satz – proposition, principleSchein – illusionSchema – schemaschliessen – to inferSchluss – inference, conclusionSchlussfolge(rung) – conclusion, inferenceSchranke – limitationSeele – soulSeelenlehre – psychology

Anita Kasabova 49

Sensualphilosoph – sensual philosophers (Kant calls Epicurus a philosopher of sensuality, as opposed to intellectual philosophers,such as Plato)Sinn – sensesinnlich – sensibleSinnlichkeit – sensibilitySitten – moralsStammbegriffe – genealogical core concepts (Guyer & Wood translate:ancestral

concepts; Kemp Smith translates: primary concepts; Pluhartranslates: root

concepts)Stoff – material, mattersubaltern – subordinatesubordinieren – to subordinatesubsumieren – to subsume

T

Tätigkeit – activityTeil – partTeilung – divisionTriebfeder – incentiveTrugschluss – fallacious inferenceTugend – virtue

U

Überlegung – reflectionÜberzeugung – convictionUmfang – range, domainunendlich – infiniteunerweislich – indemonstrableunmittelbar – immediate(ly)unrichtig – wrong (judgment)Unterordnung – subordinationUntersatz – minor premissUnterscheidung – distinctionUnterschied – differenceUrbild – archetypeUrsache – causeUrteil – judging (act), judgment (statement)

Glossary50

Urteilskraft – faculty of judging

V

Veränderung – changeVerbindung – combination, linkVerbot – prohibitionVerhältnis – relationVerknüpfung – connectionvermitteln – to mediateVermögen – faculty, capacityVernunft – reasonVernunftsschluss – inference of reasonVerstand – understandingverstehen – to understandverweisen – to referVielheit – plurality, multiplicityVoraussetzung – presuppositionvorstellen (sich) – to representVorstellung – representationVorstellungskraft, Vorstellungsvermögen – power, (faculty) ofrepresentation

W

Wahrnehmung – perceptionWahrscheinlichkeit – probabilitywechselseitig – reciprocalWechselwirkung – interactionWelt – worldWeltall – universeWelterkenntnis, Weltlehre, Weltwissenheit – cosmologyWeltreihe – world-seriesWeltweisheit – philosophyWesen – beingwiderlegen – to refuteWiderspruch – contradictionWiderstreit – conflict, oppositionwirken – to effect, to producewirkende Ursache – efficient causeWirklichkeit – actuality, realityWirkung – effectWissen – knowledge

Anita Kasabova 51

Wissenschaft – science

Z

Zeit – timeZeitfolge – temporal sequenceZergliederung – analysisZiel – aim, goalZugleichsein – simultaneityZusammenhang – connection, interconnection, context, linkzusammensetzen – to composeZustand – state, conditionzustande kommen – to come aboutzuverlässig – reliableZweck – end, purposezweckmässig – suitable, appropriate, purposive

4.2. II. B. English-German

A

in accordance with, after – nachacquaintance – Kenntnisto be acquainted with – kennenact, action – Handlungactivity – Tätigkeitactuality – Wirklichkeitaim – Absicht, Zielanalysis – Zergliederungappearance – Erscheinungarchetype – Urbildto arise – entstehento ascribe – beilegento assume – annehmenawareness – Bewusstsein

B

basis – Grundbeing – Wesenbelief – Glaubebenefit – Nutzen

Glossary52

to bound – begrenzenboundary – Grenzeto bring about – bewirken

C

capacity – Fähigkeit, Vermögencause – Ursachechange – Veränderungclaim – Anspruchclear – deutlich, klarcognition – Erkenntnisto cognize – erkennencombination – Verbindungto come about – zustande kommencommand – Gebotcomponent – Bestandteilto compose – zusammensetzenconcept – Begriffto conclude – folgernconclusion – Schluss, Schlussfolgecondition – Bedingung, Zustandconditioned – bedingtto confirm – bestätigenconflict – Widerstreitconnection – Verknüpfung, Zusammenhangconsciousness – Bewusstseinconsequence – Folgerungto consist – bestehencontent – Inhaltcontext – Zusammenhangcontradiction – Widerspruchconviction – Überzeugungcorrect – richtigcosmology – Welterkenntnis, Weltlehre, Weltwissenheit

D

to deduce – ableitendefinition – Definition, Erklärungto derive – herleiten, ableitendesignation – Bezeichnung, Benennungto determine – bestimmen

Anita Kasabova 53

difference – Unterschiedto dissolve – auflösendistinct – deutlichdistinction – Unterscheidungdivision – Einteilungdoctrine – Lehredomain – Gebiet, Umfangduty – Pflicht

E

effect – Wirkungto effect – wirkenefficient cause – wirkende Ursacheend – Zweckendless – endlosentitlement – Anspruchto examine – prüfenexhibition – Darstellungexistence – Daseinexperience – Erfahrungexplanation – Erklärungexposition – Erörterungexpression – Ausdruckextension – Ausdehnung

F

faculty – Vermögen, Kraft(faculty of) imagination – Einbildungs(kraft)faculty of judging – Urteilskraftfaculty of representation – Vorstellungskraft, Vorstellungsvermögenfaith – Glaubefallacious inference – Trugschlussfalse – falschfinal cause – Endursachefinal purpose – Endabsicht, Endzweckfinite, finally – endlichforce – Kraft

G

genealogical core concepts – Stammbegriffe (Guyer & Wood übersetzen:ancestral

Glossary54

concepts; Kemp Smith übersetzt: primary concepts; Pluhar übersetzt:root

concepts)general – allgemeingenus – Gattunggoal – Zielground – Grund

H

happiness – Glückseligkeit

I

the I – Ich (das)idea – Ideeillusion – Scheinimage – Bildimmediate(ly) – unmittelbarimpression – Eindruckincentive – Triebfederindemonstrable – unerweislichto indicate – anzeigento infer – schliesseninference – Schluss, Schlussfolgerunginference of reason – Vernunftsschlussinfinite – unendlichin itself – an sichintellectual – des Geistesintention – Absichtinteraction – Wechselwirkungto intuit – anschauenintuition – Anschauungintuitive – anschauend, anschaulich

J

judging (act) – Urteiljudgment (statement) – Urteil

K

kind – artto know – wissenknowledge – Kenntnis, Wissen

Anita Kasabova 55

L

law – Gesetzlimitation – Schrankelink – Verbindunglocation – Ort

M

magnitude – Grössemajor premiss – Obersatzmanifold (adj.) – mannigfaltig(e)manifold (n.) – Mannigfaltigemark – Merkmalmaterial – Stoffmatter – Stoffmeaning – Bedeutungmeans – Mittelto mediate – vermittelnmental – Gemütsmind – Geist, Gemütminor premiss – Untersatzmoment – Momentmorals – Sittenmotion – Bewegungmotive – Bewegungsgrundmultiplicity – Vielheit

N

name – Benennungnoogony – Noogonie (Kants Bezeichnung für Lockes system, weil es dieSinnlichkeit durch die Vernunft (νοΰς) sich erzeugen lässt (γον)noologists – Noologisten (Kant Bezeichnung für diejenigen, die die reine Vernunfterkenntnis nur aus der Vernunft herleiten, wie Plato andLeibniz)Noumenon – noumenon

O

object – Gegenstand (object of experience), Objekt (object forknowledge)obscure – dunkel

Glossary56

to opine – meinenopinion – Meinung

P

part – Teilperception – Perzeption, Wahrnehmungphenomenon – Phaenomenonphilosophy – Weltweisheitplace – Ortplurality – Vielheitpostulate – Postulatpower – Kraftpower of representation – Vorstellungskraftpresentation – Darstellungpresupposition – Voraussetzungprinciple – Grundsatz, Prinzip, Satzprobability – Wahrscheinlichkeitto produce – hervorbringen, wirkenprohibition – Verbotproof – Beweisproperty – Eigenschaftproposition – Satzpsychology – Seelenlehrepure (adj.) – reinpurpose – Zweckpurposive – zweckmässig

R

range – Umfangreality – Wirklichkeitreason – Vernunftreceptivity – Empfänglichkeitreciprocal – wechselseitigto refer – verweisenreflection – Überlegungto refute – widerlegenregard – Absichtrelation – Beziehung, Relation, Verhältnisreliable – zuverlässigto represent – vorstellen (sich)representation – Vorstellung

Anita Kasabova 57

respect – Absichtrule – Regel

S

schema – Schemascience – Wissenschaft(self)-evidence – Evidenzsensation – Empfindungsense – Sinnsensibility – Sinnlichkeitsensible – sinnlichsensual philosophers – Sensualphilosoph (Kant nennt Epikur auchPhilosoph der Sinnlichkeit, im Gegensatz zu Intellektualphilosophen wiePlato)to separate – absondernsequence – Folge, Reihenfolgeseries – Reiheshape – Gestaltsimultaneity – Zugleichseinsoul – Seelespace – Raumspecies – Artstate – Zustandstudy of nature – Naturbetrachtung (nach Kemp Smith; Guyer & Woodübersetzen: consideration of nature, und Pluhar übersetzt: contemplation of

nature)subordinate – subalternto subordinate – subordinierensubordination – Unterordnungto subsume – subsumierensuccession – Abfolgesuitable – zweckmässigsum total – Inbegriff

T

taking to be true – Führwahrhaltentemporal sequence – Zeitfolgeto test – prüfentheorem – Lehrsatzthing-in-itself – Ding an sich

Glossary58

thinking – Denkenthought – Gedanketime – Zeittotality – Allheit, Totalität

U

to understand – verstehenunderstanding – Verstandunit, unity – Einheituniversal – allgemeinuniverse – Weltallusage, use – Gebrauchusefulness – Nutzen

V

vacuous – nichtigvalid – gültigviewpoint – Gesichtspunktvirtue – Tugend

W

whole – Ganze(s)world – Weltworld-series – Weltreihewrong (judgment) – unrichtig

5. New Anti-Kant or examination of the Critique of Pure Reason

according to the concepts set down in Bolzano’s Theory of Science

by Dr. František Přihonský

Bautzen, 1850

on commission by A. Weller

“What is certain on grounds of reason cannot be refutedby a contrary experience.”

Bolzano’s Theory of Science, § 283, No. 5

František Přihonský60

CONTENTS

..................................................................................................................page

PREFACE

Justification of the entire investigation............................................................XIII/67

An emphasis on Kant’s merits in philosophical inquiry..................................XVII/68

INTRODUCTION

Some important concepts used, especially in regard to evaluating

the claims of critical philosophy..............................................................1/72

TREATISE

How Kant understands the distinction between a priori judgments

and empirical ..........................................................................................

judgments and how this distinction can be expressed more correctly

and accurately..........................................................................................21/87

Kant’s account of synthetic and analytic judgments. Is his definition

of these concepts correct and entirely adequate

for scientific ends? ..................................................................................34/95

Kant’s provisional explanation of the origin of synthetic a priorijudgments and the doubts it raises...........................................................37/98

The division of the Critique of Pure Reason into a “Transcendental

doctrine of elements” and a “Transcendental doctrine of method” .......43/101

How should we consider the claim Kant occasionally puts forward here,

that objects are given to us through sensibility and thought by the

understanding? ........................................................................................44/103

[IV] The “Transcendental aesthetic”, or the first part

of the “Transcendental doctrine of elements”.........................................45/103

The important division of representations into intuitions and concepts .........45/103

Kant’s notion of intuition in general and of pure intuition in particular

is not clear ...............................................................................................48/103

New Anti-Kant 61

Are what Kant calls forms of sensibility rightfully called

pure intuitions?........................................................................................ 53/111

Are these forms of sensibility space and time, as Kant would have it?

What actually are space and time? ................................................................. 55/112

Representations of time and space are not intuitions but pure concepts ........ 58/114

The objects of these representations, time and space themselves,

do not belong to the class of real things.................................................. 59/115

Some of Kant’s most outstanding teachings on space and time...................... 62/116

Their evaluation............................................................................................... 64/118

Should space and time, that is, the whole of infinite space and the whole

of infinite time, be considered as intuitions because they only have

a single object?……................................................................................. 65/118

Is it true that the parts of space are only thought in space itself and as its

limitations? Does this prove that space is singular? .............................. 65/118

Is the representation of space composed of an infinite number of parts?....... 67/119

Does each and every representation of a certain part of time contain

the representation of the whole of infinite time as a component?........... 68/121

Is it true that an object whose parts can only be represented by

limitations can never be represented by concepts?................................. 68/121

Does Kant rightly refer to the mathematical sciences [V] in order to

affirm beyond any doubt that time and space are pure intuitions?......... 69/122

Has Kant demonstrated anywhere and in any way that geometrical

propositions are based on pure intuitions? ............................................. 70/123

Is there any truth in Kant’s attempted distinction between an intuition

originating from a real drawing of an object and an intuition

supposedly brought about (constructed)

by the faculty of imagination? ................................................................. 70/124

Is the representation of time contained in arithmetical propositions

and does it even ground them, as Kant assumes? ................................... 73/127

The proposition that concepts without an intuition corresponding to them

yield no cognition is false ........................................................................ 75/128

The second part of the “Transcendental doctrine of elements”, the

“Transcendental logic”The latter divides into an “Analytic” and a

“Dialectic” .............................................................................................. 75/128

“Analytic of concepts ...................................................................................... 75/128

Kant’s doctrine of the pure concepts of the understanding or elementary

concepts, also called categories .............................................................. 75/128

Possible objections to this doctrine in general and to the deduction of the

categories from the forms of judgments, in particular ............................ 76/128

Important doubts about the fourfold division of judgments ............................ 79/129

It is doubtful whether Kant’s table of categories is complete ......................... 81/132


Kant’s deduction of the pure concepts of the understanding,

or the proof that the categories are only applicable to objects

of possible experience..............................................................................82/133

Objections ........................................................................................................83/134

The untenability of Kant’s explication of the necessary [VI] applicability

of all categories to intuitions ...................................................................84/135

Can intuitions contradict categories? .............................................................84/135

Are we justified in making a judgment based on certain intuitions we

have had, which contradicts a pure conceptual truth?............................85/137

The assertion that categories without intuitions are empty

(without an object) is false.......................................................................86/137

How can a better explanation than Kant’s be given of the fact

that experience agrees with the concepts we have of its objects? ...........86/137

The Critique contradicts itself by claiming that all categories

are only applicable to objects of possible experience .............................88/139

“ANALYTIC OF PRINCIPLES”

The vacuousness of the doctrine of the schematism ofthe pure concepts of the understanding ...................................................88 / 141

On which supreme principle do all analytic judgments rest, according to

Kant?........................................................................................................89 / 142

Is the origin of these judgments, if they are supposed to be true

cognitions,

sufficiently explained in this way?...........................................................91 / 142

How Kant expresses the supreme principle of all synthetic judgments.

Is their possibility really to be understood

by means of this principle? ......................................................................91 / 142

A systematic exposition of all synthetic principles

of the pure understanding ........................................................................92 / 143

A brief assessment of the latter ........................................................................93 / 144

Neither the magnitude of a temporal interval nor the magnitude

of a spatial distance belong to intuitions .................................................95 / 145

All causes, and not, as Kant thinks, many of them, are

simultaneous with their effects.........................................................................96 / 146

The postulates of empirical thought are wrong ...............................................97 / 148

The distinction of objects into phaenomena and noumena..............................99 / 149

Kant considers this distinction to be merely problematic .................................................100 / 150

[VII] Concepts of reflection in the Critique ....................................................101 / 151

Can every judgment be preceded by a reflection, as Kant assumes?..............103 / 153

Kant’s concepts of reflection do not form a whole, nor do they present

mere relations among concepts ...............................................................104 / 153

New Anti-Kant 63

It is false to think that the difference between categorical and other

(hypothetical and disjunctive) judgments consists in the fact that

the former express an internal relation between subject and

predicate and the latter an external relation between them.................... 105 / 154

It is also wrong to say that in problematic judgments the ground

of the connection is to be found in the form of judgments and in

assertoric and apodictic judgments in the matter ................................... 106 / 154

Kant’s view of Leibniz’s principle of the indiscernibles is wrong .................. 106 / 155

It is not true that logical affirmative propositionscan never conflict

with one another ...................................................................................... 106 / 155

It is also not true that determinations of all objects of appearance

are limited to mere relations ................................................................... 107 / 155

What could the question “does matter precede form or

does form precede matter?” mean and how should it be settled? .......... 107 / 156

“Transcendental dialectic” ............................................................................. 108 / 156

The origin of illusion and error, according to Kant........................................ 108 / 156

A better explanation of the origin of error ...................................................... 109 / 156

On pure reason as the seat of transcendental illusion...................................... 110 / 157

On the concepts of pure reason or transcendental ideas ................................. 111 / 158

Deduction of the three transcendental ideas (the soul,

the world, and God .................................................................................. 112 / 159

Evaluation of this deduction............................................................................ 113 / 161

[VIII] It is false to define God as a being which contains

the supreme condition for the possibility of everything which can be

thought ..................................................................................................... 114 / 162

Conceptual truths are completely independent of God’s will ......................... 114 / 162

On the dialectical inferences of pure reason.................................................... 115 / 163

On the three classes of such inferences: the paralogisms,

the antinomies, and the ideal of pure reason .......................................... 115 / 163

Do they really correspond to the three familiar forms of syllogism? ............. 115 / 163

Kant’s doctrine of inference is altogether defective........................................ 116 / 164

The division of syllogisms into categorical, hypothetical, and disjunctive

syllogisms is wrong.................................................................................. 118 / 166

On the four paralogisms of the transcendental doctrine of the soul ............... 119 / 167

Important doubts about these paralogisms ..................................................... 120 / 168

On the four antinomies in which reason supposedly gets entangled

if it dares to go beyond experience and make judgments about

suprasensory objects ............................................................................... 122 / 170

How Kant intends to resolve these antinomies................................................ 128 / 174

A critique of this strange doctrine ................................................................... 131 / 175


On the ideal and what is the only true ideal of which reason is capable,

according to Kant ....................................................................................142 / 185

Critical doubts about whether God is rightly defined as

the most real being...................................................................................143 / 186

Kant’s critique of the ontological proof of the existence of God.....................143 / 187

Where, in our view, the mistake in this proof lies............................................145 / 188

Kant’s mistakes in evaluating the cosmological proof ....................................151 / 192

The true relation between physicotheological and ontological inferences .....154 / 194

What does Kant teach on the determination of the ideas of pure reason?

And what may be said by way of reply ....................................................155 / 195 [IX]

Does constitutive use of ideas of pure reason necessarily obstruct our

investigation into nature? ........................................................................161 / 200

“Transcendental doctrine of method”.............................................................163 / 201

What does Kant understand by the discipline of pure reason? .......................163 / 201

Why does he reject the application

of a mathematical method in philosophy? ...............................................164 / 201

Kant’s claim that exact definitions and strict demonstrations cannot

occur in philosophical investigations is false and damaging..................164 / 202

His concepts concerning the nature of definition and some other logical

objectsare defective..................................................................................166 / 203

His opinion that no concept can be immediately related to another

is wron......................................................................................................169 / 205

Finally, his claim that philosophical proofs cannot reach such

a high degree of reliability as mathematical proofs is exaggerated .......174 / 210

What does the Critique teach about hypotheses? ............................................176 / 213

Must the presuppositions of a hypothesis be absolutely possible?..................178 / 214

There is some truth in the claim that the wildest hypothesis is more

acceptable for the explanation of a natural phaenomenon than is an

appeal to God...........................................................................................179 / 215

Is a hypothesis suspect merely because it requires certain auxiliary

hypotheses for the complete explanation of a phaenomenon? ................180 / 216

Is it true that the assumption that there is a totally perfect being and

that the soul is a simple substance leads to difficulties

which only new hypotheses can eliminate? .............................................181 / 216

Is it really absurd to want to lend mere probability to

a transcendental idea or a pure conceptual proposition?.......................182 / 217 [X]

Even mathematicians use probability proofs...................................................183 / 217

On the nature of the proofs of pure reason......................................................183 / 218

What should be thought of the rule put forward by Kant that we should

not start testing a proposition until we can anticipate the successful

completion of the test? .............................................................................185 / 219

New Anti-Kant 65

How should we consider the advice to counter every proof of a

transcendental proposition with a learned non liquet and to reject

such inferences en masse?....................................................................... 185 / 219

Does each transcendental proposition only allow for a single proof? ........... 186 / 219

What seems to us to be wrong in Kant’s view about

the apagogical form of proofs ................................................................. 187 / 220

On “The canon of pure reason”...................................................................... 188 / 221

In Kant’s view there is no canon for a speculative use of pure reason

but only for a practical use...................................................................... 189 / 222

What does Kant understand by the practical use of reason and in what

contradiction does this involve him?....................................................... 193 / 224

Is it true that moral commands are unconditioned, and if not,

what conditions are they subject to? ....................................................... 193 / 224

Which is the only unconditioned command of duty?....................................... 196 / 228

The deduction of the true supreme moral law................................................. 200 / 229

How does Kant express the supreme moral law?............................................ 200 / 230

Objections........................................................................................................ 203 / 231

On the type of inference that Kant calls the demand (the postulate)

of practical reason................................................................................... 205 / 234

Is it possible to establish a proof of God’s perfections

from the postulates of practical reason? ................................................. 206 / 235

On opining, knowing, and believing................................................................ 208 / 236

Kant’s way of determining these concepts does not comply with usage,

nor with the goals of science ........................................................................... 208 / 236 [XI]

The proposition that belief always occurs only in a practical relation

is erroneous ............................................................................................. 208 / 236

Pragmatic and doctrinal belief........................................................................ 209 / 237

The wish to reach a certain end cannot provide a ground for the belief

that some means or another is the most suitable for that end................. 209 / 238

Kant’s example of the opposite contradicts what he says ............................... 210 / 238

Does the so-called moral belief in God and immortality produce

such a firm and indestructible conviction as Kant credits it with? ......... 210 / 239

“The architectonic of pure reason” ................................................................ 212 / 240

What does Kant understand by a system or science?...................................... 213 / 241

This makes it possible to determine more precisely the idea according

to which a collection of cognitions must be ordered, so as to lay

claim

to the name of a science........................................................................... 215 / 242

Has Kant correctly grasped the difference between mathematics and

philosophy?.............................................................................................. 216 / 243


Should we agree with his claims: 1) that only mathematics can be taught

but not philosophy, 2) that mathematics excludes all illusion and

error, and 3) that we can obtain merely historical knowledge from

philosophical doctrines but not from mathematical ones?......................217 / 243

Why Kant cannot provide a satisfactory reply to the objection,

how can I expect an a priori cognition of objects, insofar as

they are given to our senses a posteriori?................................................218 / 244

“The history of pure reason” ..........................................................................219 / 245

Three points of view according to which we can grasp the changes

that occurred in metaphysics ...................................................................219 / 245 [XII]

Sensualist and intellectual philosophers .........................................................219 / 245

Empiricists and noologists...............................................................................220 / 245

Naturalists and scientific method ....................................................................220 / 245

Dogmatism, scepticism, and criticism .............................................................220 / 245

What is the only sense in which we may speak of intellectual intuitions?.......220 / 246

A closer examination teaches us that empirical or experiential judgments

are also based on pure conceptual propositions .....................................221 / 246

On the success of the critical method ..............................................................222 / 246

APPENDIX

A perspicuous presentation of the most distinguished doctrines

of the Critique of Pure Reason .................................................................223 / 247

5.1. New Anti-Kant

PREFACE

Even today it should not be too late to examine Kantian philosophy. Two

different editions of Kant’s works,*90 which appeared only a few yearsago, give sufficient evidence of the continuing interest in hisphilosophical investigations. Kant is not yet history; he is not an alreadyextinct celebrity but still counts a [IV] substantial number of discipleswho honour his points of view, at least to some extent. Severalsubsequent philosophical creations have been built on the basis of hisprinciples, and many of his eccentric theorems have even passed intomore recent systems and amalgamated with them, so that we would notbe mistaken in saying that they penetrate contemporary philosophy. Thisis why Hartenstein (in the preface to the new edition of Kant’s Works, p.VIII) writes that

“[h]e <(Kant)> is lucky . . ., in that subsequent systems are to a large extent

further developments of seeds we can trace back to his writings more or less

accurately, even if now and then their internal distortion is concealed by the

illusion of a sumptuous and lively abundance. Examples are Fichte’s idealism,

which is based exclusively on the pure concept of the self; Schelling’s older

doctrine of identity, where intellectual intuition becomes a source of knowledge;

Hegel’s Dialectic; and Herbart’s Monadology, based on the concept of being as an

absolute position and an acknowledgment of that which is given. Regardless of

the particulars of each of these thinkers’ works, all their systems recall thoughts,

[XV] questions, and problems which Kant has either shown more clearly in the

light of philosophical examination, or designated as possible sources of

knowledge for other intelligent beings without granting them the slightest

applicability to human thought, or validated as the essential and necessary

corrections of deeply rooted errors.

And Marbach could even discuss our philosopher in an admittedly

exaggerated way (in his Textbook on the History of Philosophy),91

asfollows: “During his long life, Kant applied his excellent insight to thedevelopment of philosophy—so it goes without saying that until now, all

*90 (1) Imm. Kant’s Works, complete edition in ten volumes, carefully revised. With a

preface by [Gustave] Hartenstein, Prof. of Philosophy at the Univ. of Leipzig; Modes &

Baumann, [Leipzig,] 1838. (2) Imm. Kant’s Complete Works, edited by Karl Rosenkranz

and Friedr. Wilh. Schubert; Leop. Voss, Leipzig, 1838.91 Gotthard Oswald Marbach, Lehrbuch der Geschichte der Philosophie, 3 vols. (1838-

1844) Leipzig: Otto Wiegand. Cf. Morscher (in the German edition of New Anti-Kant)who points out that this quote does not appear in vols. 1 or 2, and vol. 3 was not

published. Morscher suggests that Přihonský may have referred to a manuscript version

(cf. E. Morscher, Neuer Anti-Kant, ed. 2003, p. 270).


we, as his heirs, could do, was to expound his views. We should waitanother fifty years before writing a history of philosophy and see howclosely we will be gathered around him then, standing united under hisglorious spirit, whose light will eradicate most of our individualities.”Friedrich Francke, associate professor of philosophy at the University ofRostock, expressed himself in a similar way, only more mildly:

“I freely admit [XVI] that I consider neither Schelling nor Hegel, nor anyone with

a mind similar to theirs, as the greatest thinkers of the past and present centuries.

Instead, time and again any development of a healthy philosophical speculation

must orient itself with regard to Kant and Fries.”92

(On the Theory and Critique

of the Faculty of Judgment, Leipzig, 1838, preface, p. VI)93

Professor E. S. Mirbt, almost entirely Kantian except for a few objectionshe makes against his master, briefly declares that “the building materialof these (post-Kantian) systems was mainly borrowed from Kant’sworks.” (Kant and his Successors, Jena, 1841, p. 171) And likewiseProfessor G. E. Erdmann considered the Kantian system “as the seedwhich implicitly contains all subsequent conceptual constructions ofmodern times” (History of Modern Philosophy, Leipzig, 1843, vol. 3, sec.1, p. 24). Finally, the Friesian School headed by Dr. Apelt is known forits fidelity in following Kant’s principal doctrines. – Do we needanything else to justify the present investigation?If we then not infrequently contradict such an outstanding authority asthe Königsbergian scholar, [XVII] our diverging claims might seemimpudent to many and would indeed be so, if we were unable to supportthem by sufficient reasons. Yet we believe we can do this, admittedlywith the support of certain views that we will discuss later. We must letthe reader decide if, and to what extent, we actually succeed in thismatter, though we cannot but earnestly implore him not to conclude fromthe title of our book that we intend to open a passionate polemic. Wechose the name Anti-Kant merely to briefly designate our text, and theepithet new partly to point out the nature of our reasons for disagreeingwith Kant and partly to distinguish this book from another, alreadyexisting, Anti-Kant (Stattler’s well-known work).94 Thus we may hope toearn the readers’ confidence by assuring them that in our efforts weintend to promote only the truth. But in order to completely remove every 92 Jakob Friedrich Fries.93 Friedrich Francke, Zur Theorie and Kritik der Urtheilskraft, Leipzig: Julius Klinckhardt

(1838).94 Cf. Scholz (1931), who comments that Benedikt Stattler’s Anti-Kant (1788) is an

obsequious work in 3 vols. which is rightfully forgotten (see Morscher’s edition of

Přihonský, 2003, p. 203).

New Anti-Kant 69

thought [XVIII] to the contrary and to convince them that we do not havethe slightest intention of denigrating the achievements of this great manin the science of philosophy, we want to specially emphasise here someof the most important and unquestionable ones.

It is undeniable that the period immediately prior to Kant’s was a timeof vagueness and stagnation in philosophical research and that it was theKönigsbergian scholar who stirred up the spirit of reflection again, andwho both stimulated and revived the zeal for such investigations. Indeed,it would not even be a mistake to assert that what was achieved inphilosophy after Kant was mainly due to his involvement andencouragement. The new views he presented prompted philosophers tonumerous activities. Some approved of his views and tried to appropriatethem, whereas others were less favourably inclined and defied them. Theinstigator of this mental agitation never entered the battleground himselfon behalf of his cause. He was satisfied that his pupils had taken on[XIX] a vigorous defense of their master’s contested claims. Of coursethis latter did not always succeed, and astute opponents very soondiscovered weak points, which they exploited for even more successfulattacks. Then again, Kant’s friends considered themselves obliged tomend the exposed fissures and correct errors, as well as to discarduntenable claims and try their hand at building new systems. The newand rich life which arose as a necessary result certainly reflects noordinary merit won by Kant on behalf of the philosophical sciences. Inaddition, Kant has undeniably improved philosophy by not fightingineffectually against the extreme methods impairing a steadfast research,such as we have so far observed at philosophical lectures: an unrestraineddogmatism and a far too bold scepticism. Whilst trying to investigate thegrounds on which human knowledge is based, he also maintained that weshould neither make decisive assertions before having ascertained theirtruth nor boldly deny [XX] or doubt that which has no sufficient reason.This is how he attempted to introduce the critical method into philosophyby encouraging and recommending both examination and modestresearch, whilst adequately limiting exuberance in asserting, doubting,and denying. – Yet Kant achieved even more, not only for philosophy,but for all mankind by basing moral doctrine on purer foundations andliberating it from selfish motives. Earlier moralists for the most partrevered the principle of self-interest, a false and harmful principle, whichthey tried not only to endorse in science but to introduce in ordinary lifethrough popular writings. We can easily understand that mankind eagerlygrasped and maintained such a principle, since it so greatly flattered theirwishes, and it truly took all of Kant’s weighty authority to deprive them


of it and to convince them that it was false and harmful. The great manaccomplished this perfectly. The former slackness [XXI] eventuallydisappeared from the domain of ethics and was replaced by acommendable seriousness, which penetrated the entire moral doctrine. –We are truly convinced that these are Kant’s most highly regardedachievements in philosophical science as a whole. But they dopresuppose that he dealt with each of them specifically, with greataccomplishments. And did not Kant provoke a number of extremelyinteresting questions, for most of which he tried to find answers himself?Can we make synthetic judgments about nonsensory objects? Do timeand space actually exist? Do their representations belong to the class ofconcepts or intuitions? What can we possibly know of things? What kindof knowledge do we have? Is it merely subjective, and what limits can beset to it? How many concepts of explication or categories are there, andwhich of them are simple? And so on. – What important and fruitfultruths our philosopher has discovered and found in almost all domains ofphilosophy, including logic! [XXII] Even if he had done nothing exceptpoint out the distinctions between a priori propositions and empiricalpropositions, between intuitions and concepts, between analytic andsynthetic judgments (and though I would not say that he grasped themwith sufficient rigour, he undeniably drew attention to them), he has doneenough to keep a place in the history of philosophy for all time.

We know that Kant did not venture to establish a proper and completesystem of philosophy. He confined himself to exposing the mistakes andinadequacies of previous systems and to merely giving hints anddirections for building a new one. He mainly did this in those works hecalled Critiques, because of the method used, which we have praisedabove. The most outstanding and complete one is undeniably the Critique

of Pure Reason where he set down the most important results of hisphilosophical research, so that we may say that someone has examinedKantian philosophy if he has investigated [XXIII] the particular validityof the views and assertions in this work. This is why we think we are notmistaken in limiting our examination to the famous Critique. Anyonewho has thoughtfully followed our investigations and has, to some extent,acquired the concepts we use here, can easily judge particular features ofthe Critique of the Faculty of Judging, the Critique of Practical Reason,and other Kantian writings. We use notions taken in part from Bolzano’sremarkable writings, in particular his Theory of Science or Logic.Although the latter appeared more than a decade ago (Sulzbach, 1837),

New Anti-Kant 71

(strangely enough*95

) the most important concepts have hardly beendiffused, so that they may still be considered new, and we can presume

that only a few of our readers [XXIV] are acquainted with them.*96

Thisis why we have considered it necessary to begin our treatise by brieflyintroducing some of the exceptional notions we mainly intend to use, tomake sure that the following debates will be well understood.

*95 But not inexplicable: see An Evaluative Survey of Bolzano’s Theory of Science (Logic)and Theory of Religion, Sulzbach, Seidel, 1841, pp. 7-8.

*96 All the more so, since even professional philosophers do not seem to know anything

about Bolzano’s Logic. See, for example, The Importance of Logicians by Dr. Carl Prantl

(Munich, 1849), which deals with almost all relevant contemporary writings, except for

this most significant and extensive work. [Cf. Scholz (1931), who points out that the

Prantl Přihonský refers to is the famous author of the Geschichte der Logik im Abendlande(op. cit., p. 203).]

5.2. New Anti-Kant

INTRODUCTION

Every true philosopher should unquestionably strive towards thecognition of pure truth and, to be precise, a truth which is demonstrableby human beings. But then we must ask whether there is such a truth andwhat we understand by truth. This is why we believe we should start byexplicating this particular concept.

We frankly admit that we have not wanted to take the liberty ofconferring a meaning on a word as important as truth which could not bejustified by usage. We need to determine more precisely the sense inwhich we wish to take the word here, because a previous usage relatesseveral senses to it. It seems to us that our sense results with sufficientclarity from the following discourse:


“Amongst the three propositions: winged snakes have never existed on earth or

they are already extinct or they still exist, one is necessarily true or a truth.”97

[2] For if we talk in this way, it is obvious that we consider the true ortruth to be a certain proposition as such, no matter whether there issomeone who holds this proposition as true and expresses it or otherwiserepresents it, or whether there is no such being. So we take the word truthin the sense that there are truths as such and consequently that there arealso propositions as such which no one knows (except God) or even hasin mind.98 We say, for example: “without anyone having yet raised thequestion which is the thousandth decimal of the number π, one of tenpropositions we think of when explaining that the number referred to is a0, a 1, a 2, and so on, is still true, or a truth.” – So if we speak of truths inthis sense (Bolzano calls them objective; see his Theory of Science, § 24),it is obvious that we presuppose that there are truths and propositions as

such which should be distinguished from mental truths and mental

propositions. Although the latter appear in the mind of a thinking being,there is no way in which the former can be subordinated to them, for theyare no thoughts at all, neither real nor possible. So, although there aremental truths or mental propositions, we would not say this of truths assuch nor of propositions as such. There are (esse) no truths as such,especially not [3] eternal truths. Yet we sometimes say that religious andmoral truths are eternal, but this is only an inappropriate locution, and itssense is that such truths subsist in propositions which do not containtemporal determinations as the conditions of their validity. The oppositeis the case for certain other truths, for example, that three bushels of corncost three pounds, and so on, because for the last proposition to be true,its thought must be completed by a temporal determination such aspresently, or one like it. Precisely for this reason truths as such orpropositions as such and mental truths or mental propositions cannotreally be considered as two kinds of truths and propositions at all. It

97 Cf. Dubislav’s note that Přihonský refers to WLI, §§ 19, 25, 34, 48, 70, 72; WLII, §§

152, 182; and WLIII, § 307; and Bolzano’s respective distinctions between propositions as

such (these further divide into true and false propositions as such), judgments (which

divide into true judgments or cognitions and false judgments or errors), and

representations as such (which divide into simple and composed representations with

regard to content and into objectual and objectless representations with regard to

extension). (Cf. Morscher’s edition of Přihonský, 2003, p. 204.)98 Sich vorstellt. Přihonský summarises Bolzano’s explication of truths as such in WLI, §

19. Scholz (1931, in Morscher, op. cit., p. 197) points out that Bolzano has not dealt with

the awkward question of how a proposition can express something which is not formulated

or even thought by anyone; cf. also Palàgyi (1902), Kant und Bolzano, eine kritischeParallele, § 7; Halle, Niemeyer.

New Anti-Kant 73

would also be a mistake to define propositions and truths as such aspropositions and truths in abstracto. A proposition as such is not ajudgment abstracted from a judging subject, for even then it wouldremain something real, just as a watch, abstracted from its owner who iswearing it, still remains something real (consisting of metal, etc.).Propositions as such are to mental propositions as flowers are to paintedflowers. The only difference between them is that flowers are somethingthat exists, whereas propositions as such do not exist, as we have pointedout.*99 [4] If anyone should ask what, then, is a proposition as such or atruth as such, if they do not exist, we would reply that not every thing isnecessarily an existing thing. Similarly, we consider that somethingwhich is merely possible and does not yet have reality (or existence) issomething, and not nothing. Another objection is this: if truths as such donot exist, in what sense can we then assert that there are truths as such?

*

99 Here we compile propositions and truths because truths in our sense are also

propositions, that is, true propositions, so that what is valid for propositions in general

must also be valid for truths [Přihonský note].

(1) Scholz (1931) comments that for Bolzano, an eternal truth is a proposition expressing

a persistent (eternal) relation (cf. WLI, § 25), and it makes no sense for us to ask at which

point in time this proposition is true, as is the case of either mathematical propositions or

propositions that are true at all times; cf. Morscher, 2003, op. cit. p. 205. In addition, (2)

Přihonský’s example recalls Bolzano’s example of a painted fish which is not a kind of

fish but a kind of painting (WLI, § 29.4.a), but Přihonský does not expound this semantic

problem as Bolzano does, by analyzing the relation of attribution in the linguistic

expression of representations, where the adjective either determines or modifies the noun;

cf. also WLI, § 23 (“golden chandelier”), § 59.2, WLII, § 170 (where he comments on the

usage of literally incorrect expressions, such as calling a painting of a fish a “painted

fish,” although a painting is not a fish, and a painted fish is not a real fish), and his early

text entitled “Etwas aus der Logik” in Mathematische und Philosophische Schriften 1810-

1816, BBGA, 2, A, vol. 5, p. 143. Cf. also W. Künne ““Die Ernte wird erscheinen . . .”Die Geschichte der Bolzano-Rezeption 1849-1939” (1997), p. 35. On this kind of

“referential illusion”, cf. also Anton Marty, the Swiss philosopher who taught at the

German University of Prague (1880-1914). The linguistic form of expressions such as “a

painted fish,” “the late king,” or “a probable winner” differs from noun phrases with

premodifiers where literal attribution occurs, such as “small fish” or “Mediterranean fish,”

because this form only gives the impression of an attribution, since here the adjectives

more or less modify or change the meaning of the noun and hence the referent of the

modified meaning is a fiction which only bears a quasi-similarity to the proper meaning of

these names. According to Marty, such linguistic expressions show an analogous

linguistic structure which only appears to be grounded in a fictional analysis of thought,

for actually the structure of the thought is much more complex than what appears in the

linguistic structure; cf. Raum und Zeit (1916) p. 198; and “Von den logisch nichtbegründeten synsematischen Zeichen”, (1928), pp. 40-41, preceded by Funke’s

commentary “Von den semasiologischen Einheiten und ihren Untergruppen” (ibid., pp.

30-31).


What does there are refer to if it does not refer to existence, and whatentitles you to say that certain propositions are true and others not true ifthere are no truths as such? Do you not in this way predicate existence ofthem? – Our reply is that as the scholastics have already pointed out, thecopula est or the est tertii adjectis in propositions does not refer to a realbeing. For example, in the proposition: the possible as such is not real,the word is certainly does not indicate existence (Esse).

If someone should ask what this conjunction truly referred to, wewould reply that essentially it referred to having an attribute, for everyproposition of the [5] form: A is B really means nothing other than this: Ahas the attribute B; for instance: God is omnipotent, basically means:God has omnipotence; the possible is—not real only means: thepossible—has—no reality.100 Now if we specially assert that a certainproposition x is true, we assert nothing other that than this propositionpredicates an attribute of the object it deals with, an attribute which isappropriate to the object or which the object has. And when we generallysay that there are truths, basically this only has the sense that not allpropositions are false or that we assert the following: the proposition that

all propositions are false, is false (i.e., it has no truth).We consider the concept of propositions and truths as such to be so

important that we adamantly appeal to our readers to grasp itappropriately and to retain it. We will lose everything if they eithercannot accept this concept but consider truths as such only as certainthoughts or impressions in the mind of a thinking being or, in case weprevent this, if they then assert that they do not know what they shouldconsider as a truth as such, other than the very thing about which wemake a judgment. That is not the case at all: we believe we must stronglydistinguish between the thing and a proposition expressing somethingabout this thing, [6] either by predicating an attribute of it or by denyingthat it has such an attribute.101 The thing or object dealt with in a

100

Přihonský refers to Bolzano’s canonical form of propositions, “A has b”, in which the

concept of having replaces the copula “is.” Cf. WLI, § 60; WLII, §§ 126-127, 137; WLIII,

§§ 196.2a-3, 286.4.101

It seems that Přihonský follows Bolzano in distinguishing between two types of

negation (cf. also p. 7, note 6): 1) negating the predicate (verneinender Satz) and 2)

negating the sentence (Verneinungssatz or Berichtigungssatz). 1) is a de re negation with

a positive content: “A has lack of b” or “A has [the property of not having b]”, whereas 2)

is a de dicto negation with a negative content “(the sentence A has b) is not true”. Cf.

WLII, § 127, § 136, § 141, § 196.2a-3. Bolzano anticipates Frege’s claim that acts of

judging may have affirmative or negative contents and that we should not distinguish

between negative and affirmative judgments because there are no negative judgments but

only affirmative judgments, which can have a negative content, and that content can be

New Anti-Kant 75

proposition is often only a single one, but the true and false propositionsdealing with it may be countless. Then there are other cases where thenumber of objects dealt with in a proposition is infinite, while theproposition is only a single one: for example, every equilateral triangle isalso equiangular. But why should we worry about whether our readersunderstand us or agree with us, since the concept we wish to beacknowledged here has been recognized and established by so manyphilosophers? It even appears on a regular basis in daily life. That whichthe earlier logicians called veritatem objectivam, transcendentalem,metaphysicam and expounded as quae nemine cogitante dicie, uti res est,is nothing but a truth as such in the sense we have established here. –And how frequently do we not hear the following words on the lips ofordinary men: if only anyone knew the truth! Do we consider truth as anassertion or a judgment or even a mere representation of a proposition inthe mind of some thinking being (unless it be God’s omniscient mind, forall truths are known to Him)? Certainly not; instead, we presuppose that

there are truths as such which no one knows except for God. [7] Inmaking this assumption we do not intend to suggest that no one iscapable of cognizing something true, that is, any truth or several truths.No, the basic human intellect that uttered the wish: “if only anyone knewthe truth!” has not broken down so far as to deny all cognition. Ifsomeone actually held this proposition and thus asserted: we humanscannot know any truth or: all our judgments are false, he wouldcontradict himself. But if he considers himself forced to admit that he cancognize at least one truth,102 a simple reasoning shows him that he shouldknow103 several and even infinitely many truths or that truths are

generally cognizable.Once we have been granted that the concept of a truth as such and

consequently the concept of a proposition as such are both not emptyand, in addition, that we are capable of cognizing truths, we will advancethe claim that every mental proposition and every proposition as such iscomposed of certain parts which are not propositions themselves. Forevery proposition deals with a certain object, but how could this happenunless something in that proposition represented this object? Besides,

judged. Cf. Gottlob Frege (1897), Logic, 1997, pp. 227-250, and (1919), Negation (ibid,

pp. 346-361).102

Cf. Bolzano, WLI, § 40, on the proof that we can know at least one truth, the basic

truth-“I have ideas.” § 41 deals with the proof that we can know infinitely many truths.103

Přihonský writes erkennen müsse, which translates “should cognize”, though the text

should read “should be able to cognize,” since the point is that truths are cognizable

(erkennbar).


every proposition expresses something about the object it deals with or,to put it more precisely, [8] we predicate an attribute of it or deny that ithas such an attribute. (The latter surely means that for this object wepredicate the lack of an attribute, which is also a kind of property.)104

How could this be, unless the proposition contained something thatrepresented this property? – So we shall call all parts of a propositionwhich are not propositions themselves representations, regardless of anyother properties they may have. A proposition as such consists ofrepresentations as such, as a mental proposition consists of mental

representations. The etymology of the word representation admittedlyrecalls certain unrelated secondary concepts, not only in German but alsoin other languages. We should ignore them and retain the given definitiononly. The same goes for naming very general concepts. In arithmetic weuse the words: square, cube, root, power, exponent, factor, and a hundredothers without any difficulty, even though they bear unsuitable secondaryconcepts! –

If a proposition as such does not exist, then neither should itscomponent parts, the representations as such, be considered as existing,and our insistence that the concept should be taken in this way is not far-fetched. Other philosophers have also taken the word concept, if not theword representation, [9] in the sense105 that we are using here, forexample, when they give a negative reply to the question: “Are thereconcepts which are quite similar?” on the additional ground that thosetwo or several concepts we consider similar are nothing other than thesame concept either thought twice or several times or indicated by two orseveral words with a similar sense, for example: triangulum et trigonum.

We must not confuse a representation and the object it represents.106

So the representation: man is nothing less than something real, and still

104

See note 5 on negation, above.105

Here Bedeutung translates “sense”.106

Here Přihonský introduces the problem of the confusion between the properties of an

object and the components of its representation, which he discusses further on (pp. 36-37).

Bolzano deals with this confusion at length and traces it back to a false assumption by the

authors of the Logique de Port Royal that all attributes of an object must be components

of the representation of that object (WLI, §.65.10). As a result, the authors of the Logiqueestablished a rule of inverse relation between the content and extension of a

representation. According to the Port Royal rule, the content of a representation decreases

as its extension increases and vice versa. The larger its content, the smaller its extension

and vice versa. Bolzano claims that his rule is incorrect because it is possible to increase

the content of a concept not only without decreasing its extension but even by increasing

it (cf. WLII, § 120). For example, by increasing the content of the representation [watch],

we decrease its extension: [pocket watch] has a bigger content and a smaller extension

than [watch] and we can decrease the extension further by increasing the content to

New Anti-Kant 77

more so, nothing less than a living being endowed with a body and asoul, and so on, but all this is valid of the object represented by thisrepresentation. In addition, one and the same representation oftenrepresents several and even infinitely many objects. There is no doubtthat the representation man has several objects and the representationtriangle an infinite number of objects. Yet there are also representationswith only a single object which we thus call singular representations,such as the representations God, universe, lowest prime number.107 As theexamples of the representations man, God, and universe show, there arerepresentations which do not have a real (existing) object. But there arealso representations which have an inexistent object, such as therepresentation of a [10] representation as such. We have shown just nowthat its object does not exist and neither do the objects of therepresentations of a proposition as such, a truth as such, and so on. – Inaddition, we claim that there are representations without an object and wecall them objectless representations, as opposed to the rest which we callobjectual representations. One such objectless representation is:nothing,108 as well as the representations: round square, 0, √-1 , log (-1),and so on, although we may doubt whether objectless (imaginary)representations such as √-1 are entitled to the name of genuinerepresentations. Some philosophers have claimed that they are only

[golden pocket watch]; cf. WLI, § 72. Bolzano’s point is that the more attributes of an

object a concept refers to, the more marks the concept contains and vice versa. He then

accuses Kant of confusing the attributes of an object and the components of its concept,

whilst thanking him for elaborating the distinction between analytic and synthetic

judgments, owing to which he was able to avoid this mistake himself. “If I am so fortunate

as to have avoided a mistake here that has remained unnoticed by others, I will frankly

acknowledge that it was only thanks to Kant’s distinction between analytic and synthetic

judgments, a distinction which could not happen if all properties of an object were

components of its representation.” (WLI, § 120) Of course Kant does not explicitly

confuse the components of a representation and the attributes of its object, nor does

Bolzano claim this. Instead, Bolzano holds that the reason why Kant applies the PortRoyal rule (cf. Jäsche Logic, § 7) is that he confuses the components of a representation

with the attributes of its object. And Bolzano holds that the reason why Kant makes this

mistake is that he does not distinguish between representations as such and mental

representations (cf. WLI, § 65.11.a), the well-known distinction also emphasised by

Přihonský (cf. pp. 1 and 8, above); cf. also my note 161, below.107

Přihonský introduces Bolzano’s distinction between different types of representations;

cf. WLI, §§ 73, 78, 79.6. He also uses some of Bolzano’s examples.108

Přihonský takes up Bolzano’s notion of representations without any extension or

referent, such as [round square], [green virtue], [golden mountain] or [nothing], cf. WLI, §

67. Dubislav (1931) points out that for Bolzano, imaginary representations are a special

kind of objectless representation; cf. WLI, §§ 66, 70 (op. cit., p. 206). For Bolzano’s

concept of [nothing], cf. WLI, § 58.4.


words combined as a representation and that this is why they are oftencalled voces sine sensu. Yet the expressions √-1, log (-1), and so on, arenot meaningless words, such as abracadabra and so on, for we use themboth in teaching how to combine certain very clearly thought conceptsand in actually combining them. In addition, we successfully use them inmathematics for discovering the main truths. If the representation: a bodywhich is limited by twenty equal sides is a genuine representation, thenwhy should the representation: a body which is limited by twenty-fourequal sides not be a genuine representation? It is certainly true that thelatter does not have a corresponding object, [11] but do we necessarilyhave to think of an object represented by the word representation? Afterall, it is indisputable that such representations can occur as componentsin true and important propositions. Admittedly, the subject-representationof a proposition cannot be without an object (then the proposition wouldnot be dealing with an object, so how could it be true?). Neither can thepredicate-representation (the representation of the property which theproposition predicates of the subject) be without an object (then theproposition would not predicate an attribute of its subject). Still,imaginary representations can certainly occur in a proposition, althoughonly as components of its subject- or predicate-representation. So, forexample, the proposition: the representation of a round square isobjectless or has no object, is surely a truth. But the subject in thisproposition is: the representation of a round square, and therepresentation is: the representation of a representation. But the latter isan objectual representation, its object is the objectless representation, andthis lack of objectuality is asserted in the proposition.

Any proposition as such is necessarily composed of certain parts (asubject-representation, a predicate-representation, and a conjunction orthe concept of having). Likewise, many, but not all [12] representationsare themselves composed of certain other parts which we must also callrepresentations, according to our previous definition. Thus therepresentation of an equilateral triangle is unquestionably composed ofthe representation of a triangle and certain other representations, for therepresentation: equilateral triangle is exactly the same as the oneexpressed by the words: a triangle whose sides are all equal to each other.This is why we claim that besides the representation: triangle, thisrepresentation also contains the representations indicated by the word:sides, and so on, just as we believe that the representation nothing iscomposed of the representations something and nothing = not something.Let us therefore call representations composed of several others complex

New Anti-Kant 79

representations and those without further components simple

representations. It is indubitable that there must also be simplerepresentations,109 and surely we may accept without objection thatrepresentations indicated by the word nothing and the word something

are simple.If a certain object has the properties a, b, c, . . . by combining the

representations a, b, c, . . . (and several others), we can very easilycompose the representation: something which has the properties a, b, c. .

. . Even if that representation does not apply exclusively [13] to thisobject, it at least comprises this latter and can thus in any case be called arepresentation of this object (that is, both of this object and another one).But this should not induce us to believe either one of the followingassertions: 1) that all objectual representations arise in this way and evenless that every property of an object is included in its representation,even if the latter exclusively refers to this object, 2) that everycomponent of a complex objectual representation represents an attributethereof. Although the representation: equilateral triangle contains theconcept equilaterality which is an attribute that fits its object, there arealso various other properties that fit this object. For example,equiangularity, the equality of all perpendiculars drawn from the apexesto the opposite sides, and so on, is not mentioned in this representation. –It is also clear that every complex representation has various componentsof which we cannot say anything less than that they indicate theproperties of objects standing under it.110 The representation ofequilaterality also appears in the representation: non-equilateral triangle,in relation to the concept of negation. [14] Yet equilaterality is alreadydenied in this concept itself and is not a property of a non-equilateraltriangle at all. – Finally, we should point out that a complexrepresentation always has just a moderate number of components,

109

Cf. WLI, § 73. But unlike Přihonský, Bolzano also considers mixed representations,

such as [the rose which spreads this fragrance] or [this rose], which are obtained from

intuitions (WLI, §§ 73, 80).110

On Přihonský-Bolzano’s use of stehen unter (objects stand under representations), cf.

Wolfgang Künne (1997), who points out the difference between Bolzano and Frege: “I’d

rather live in Prague than in Jena, for in Prague I am allowed to stand under the concept

[human being], whereas in Jena the authorities want me to fall under it”. Künne,

“Propositions in Bolzano and Frege” in “Bolzano and Analytic Philosophy”, Grazer

philosophische Studien, vol. 53 (1997): p. 215 (pp. 203-240). Künne (1997, p. 215) claims

that for Bolzano objects can stand under representation and that the statements “x stands

under y”, “x is an object of y”, “y refers to x”, and “y represents x” are synonymous; he

cites WLI, § 66-67 (p. 304), and § 70 (p. 305), but in these passages Bolzano only uses

stehen unter once (§ 66, note 5), and then he refers to the relation of subordination

between concepts.


whereas its object can have an infinite number of properties. Forexample, the representation of a circular line has only a few components:all points on this line are equidistant from two points (the poles of thecircle), but a circular line has an infinite number of properties, such as:the line has a curvature and returns to itself, its parts are all on the sameplane, it has a centre; a diameter; a radius x. – In our view, by stating thecomponents of a complex representation, we define or explain therepresentation.

After these preliminary remarks, we can acquaint our readers with avery important objective difference between representations as such. Forit is clear that in respect to extension, the most peculiar representationsare those which only have a single object, but in respect to content, themost peculiar representations are simple ones. – If we take these twoproperties together and form the concept of a representation in which thetwo are united, that is, the concept of a representation which [15] has asimple content and whose extension comprises only a single object, wemay presume that if there really are such representations, they are of aneven more peculiar kind. If we add with Bolzano that this single object isreal, although in our opinion this addition is unnecessary, then this kindof representation becomes even more important. Bolzano gives them aparticular name, intuitions, and shows that the concept which Kant andother modern German philosophers have associated with this wordessentially fits these representations.111 However, he calls everyrepresentation that is not an intuition or does not contain an intuition as acomponent, a pure concept (see the Theory of Science, § 72ff.).112 Yet wedoubt that there are representations such as we have described here asintuitions. If a representation were to represent only a single and also realobject, then we might believe that it must be composed of a large numberof parts, because we can only compose a representation that fits this andno other object by stating a great number of its properties. So how can arepresentation which is quite simple and not composed of several partsnonetheless represent only a single object? – [16] Indeed this would be

111

Except that for Kant intuitions are single, but not simple, representations; cf. KrV, A99;

JL, § 1; and this is quite an important difference between Kant’s and Bolzano’s notions of

intuitions.112

Přihonský takes up Bolzano’s distinction between two types of simple representations:

simple concepts which are not singular, such as [0], [yes], and [no], or syncategorema

such as [something] or [has]; singular but complex concepts such as [God], which has the

components [real] and [unconditioned]; and intuitions which are both simple and singular.

Cf. WLI, §§ 72, 78, note 2. Cf. also Dubislav (1931), who comments that Bolzano does

not add (hinzusetzen) that the single object is real (eigentlich) but infers it from his

characterization of intuitions; cf. WLI, §§ 72, 74 (op. cit., p. 206).

New Anti-Kant 81

impossible on the common belief that if a representation should apply toonly one object, we must think it accompanied by all of the object’sproperties. But, as we have already pointed out above, this belief is false.However, we have still not proved that there are simple representationswith only one object, for this only follows from the subsequent statement.There are certain subjective representations that first and immediatelyappear in us, whenever we attend to the changes which are incited in oursoul by a body, say, a lemon, presenting itself to our senses.113 Therepresentations first and immediately produced in this way must besimple, for complex representations first appear mediately, as severalsimple representations are combined. In addition, they must be entirelysingle, and their only object is the change presently occurring in oursouls.114 Representations with several objects tend to appear at such

113

Přihonský takes up Bolzano’s controversial causal claim: Bolzano claims that we

explain our intuitions by assuming that they have an external cause, and this explanation

justifies our sensory or nonconceptual knowledge. In his determinist view, our cognitive

apparatus is governed by necessary laws which constrain us to assume that sensations

have an external cause from which we infer them; cf. MM4, pp. 139-140. However,

Přihonský’s own view differs from Bolzano’s in that he holds the strong (and problematic)

causal claim that intuitions alone can provide reliable information about the external

world; cf. p. 48, p. 232, and my note 607. Bolzano says that we assume or presuppose

(annehmen, voraussetzen) that our sensations have an external cause (cf. WLI, § 74), and

in § 72 he even claims that a mental change is caused in our mind by an external body.

Yet he weakens this causal claim in § 74, where he asserts that the only way in which we

can explain the occurrence of this mental change is by assuming that it is related to its

object as an effect is related to its cause. Then, in § 77, Bolzano holds an even weaker

version of the causal claim, when he says that we can infer the existence of an object from

a corresponding subjective intuition or sensation. And in a letter to Exner (23 Aug. 1833),

he writes that the simultaneous presence of several intuitions justifies our presumption or

conjecture (Vermuthung) that these intuitions are caused by one and the same object.

Briefwechsel Bolzano-Exner [Bolzano-Exner Correspondence], p. 34). Cf. also his text on

Kant’s construction of concepts through intuitions (1810), where he remarks in § 4: “For

it must first be proved that an external object may correspond to a representation as

ground.” (My translation.)114

In Bolzano’s view there is a double relation between an intuition and its object. (1) One

is the causal claim discussed above. (2) Bolzano argues for another, internal relation

between an intuition and its object by claiming that the immediate object of an intuition is

a mental state. Intuitions are simple, singular ideas which are formed when we are

attentive to a mental change which is induced by an external body. An intuition is the

effect of our attention which is directed onto the mental state or change which is at the

origin of this same intuition. So an intuition is produced by a reflexive act, since it is the

representation of this mental change and its immediate object is the mental change. Thus

attention is a sufficient condition for producing an intuitive representation, but we could

ask whether we can also have intuitions without attention: Bolzano says that a

representation (of the mental change in us) is “the nearest and immediate effect of our

attention”, whenever we direct it “upon the change that is caused in our soul” (WLI, § 72).


occasions; for example, the representation greenish-yellow appears whenwe feel compelled to form the judgment: this (what I see just now)—is—greenish-yellow. It should be clear, though, that we have reached thelatter representation only mediately, through the simple representation weindicated by the little word: this. Consequently, there are at leastsubjective representations [17] which despite their simplicity onlyrepresent a single object and are therefore subjective or mental intuitions.Of course they must have corresponding objective representations of thesame kind (intuitions as such).115 All representations we linguistically

He does not say that attention is a necessary condition for producing intuitions, but we

have no other way of explaining the origin of intuitions unless we assume that attention is

a necessary and sufficient condition.115

Přihonský mentions Bolzano’s claim that “each subjective representation has a

corresponding objective one” (WLI, § 72, my translation), without going into the

difficulties of this view. Why do we need objective intuitions? (Cf. Palàgyi (1902), Kantund Bolzano, Niemeyer, Halle p. 84; Ph. Schwarz (1911), Bolzano's Vorstellungstheorieund Kant's Lehre von der Raumanschauung, Inaugural-Dissertation zur Erlangung derDoktorwürde der Hohen Philosophischen Fakultät der Georg-August Universität zuGöttingen, Hofer, Göttingen, pp. 12-13.) Intuitions are unrepeatable and cannot be

communicated except through our understanding of the concept referring to them, so we

do not need the objective intuitional meaning entities that refer to them. If we accept

Husserl’s perceptual universals, we could argue that Bolzano’s claim about the

correspondence between objective and subjective intuitions guarantees that two subjective

intuitions belong to the same type of objective intuition (cf. Barry Maund, Colours, theirNature and Representation (Cambridge: Cambridge Univ. Press, 1995), but Bolzano does

not accept Husserlian universals: the same intuition cannot occur twice (WLI, § 75), and

every objective intuition has only one corresponding subjective intuition, so there is no

guarantee for the sameness of two subjective intuitions of red, as there is with linguistic

concepts. Since two subjective intuitions are numerically different and have different

objects, they must also have two different corresponding meanings. Two different

subjective perceptions of red also grasp two different objective intuitions. But we may

subsume two subjective intuitions of red under the same concept redness, because we may

not even be aware of the differences between them. Cf. Bolzano’s Replies to Exner, 23

Aug. 1833, pp. 32-33, and 14 Jan. 1834, pp. 42-43, in the Bolzano-Exner Correspondence.

Textor (1996, Bolzanos Propositionalismus, de Gruyter, Berlin) offers a different defense

of Bolzano’s objective intuitions. He claims that objective intuitions are nondescriptive

singular terms which directly present their object (pp. 77-89) and that objective intuitions

serve as characterisations of direct mental attribution and make possible a direct mental

relation with an object (p. 89). But Textor’s view implies a sacrifice (or at least a

reduction) of Bolzano’s causal claim which is crucial for the production of subjective

intuitions. Since objective intuitions are not caused by a mental state, a reduction of the

causal claim would eliminate the asymmetry between subjective and objective intuitions.

However, I think it might be less problematic to reduce the role of objective intuitions to a

formal analogy between objective and subjective representations, which Bolzano needs to

stipulate in order to claim that every subjective proposition has an objective counterpart,

rather than reduce or sacrifice the causal claim about subjective intuitions. For without the

New Anti-Kant 83

indicate by this belong to the class of intuitions. Although we also addseveral other determinations in oral communication, such as: this, what Inow see, smell, or feel, and so on, we are only adding them to makeourselves clear to our interlocutor and not because we consider theaddition necessary for forming the representation itself.

This objective difference between representations, by means of whichwe can divide them into intuitions and pure concepts, also enables us towork out an objective distinction between propositions. For propositionscomposed of nothing but pure concepts are distinguishable frompropositions containing one or several intuitions, and we shall call theformer purely conceptual propositions and the latter empirical

propositions.116 [18] Some generally accepted concepts can bedetermined by applying this distinction to truths, and we shall concludeour introduction by defining the concepts of possibility, coincidence, andnecessity. But first we feel compelled to put forward the extremelyimportant concept of deducibility and to parse it into its components.117

causal claim, we have no explanation of subjective intuitions, and thus their objective

counterpart is be null and void.116

Přihonský reproduces, almost word for word (but without the examples), Bolzano’s

distinction between conceptual and empirical propositions in the Mathematical Method, §

7. In the Theory of Science, he uses his distinction to counter the Kantian distinction

between a priori and a posteriori judgments and claims that, unlike Kant’s, his distinction

is objectively grounded on the inner characteristics of the propositions themselves and not

on the relation between the judgment and our faculty of knowledge; cf. WLII, §133. But

in respect to judgments, Bolzano further distinguishes between immediate perceptual

judgments and empirical judgments: unlike the latter, immediate perceptual judgments are

self-evident (e.g., I am aware that the mental changes or states accompanying judgments,

such as “I have the impression A”, are mine), so here he is in partial agreement with Kant,

as he admits himself, cf. WLIII, 300.12. However, Bolzano restricts the self-evidence of

perceptual judgments to immediate judgments where something is predicated of the first-

person subject.117

Přihonský refers to Bolzano’s notion of deducibility (Ableitbarkeit), now known as

logical consequence or formal implication, and which has been both compared to and

distinguished from the Tarskian notion; cf. for example, John Etchemendy, The Conceptof Logical Consequence (Cambridge: Harvard Univ. Press, 1990); Mark Seibel (1996):

Der Begriff der Ableitbarkeit bei Bolzano (Beiträge der Bolzano-Forschung 7, ed. E.

Morscher and O. Neumaier), St. Augustin; Johan Van Benthem (1984), The Variety ofConsequence according to Bolzano in Studia Logica, XLIV, Reidel, Groningen, pp. 389-

403. In Bolzano’s view, a given class of propositions Q is deducible from a class of

propositions P with respect to certain variable components if and only if all substitutions

for a component C (which produce only true propositions in P) also produce only true

propositions in Q, and the propositions are consistent with respect to their components Cif at least one substitution for C produces only true propositions in P. (Cf. WLII, § 154,

note; WLII, § 155.2.) The main difference between Bolzano and Tarski is that deducibility

is a three-place relation that is not axiomatizable, whereas logical consequence is a two-


The rightfulness of Bolzano’s discovery will hardly be questioned: wesometimes presuppose that certain components in our representations andpropositions are variable and consider the behaviour of theserepresentations and propositions when any other components fill theplace that is presupposed to be variable.118 Of particular importance is thecase where propositions A, B, C, D, . . . and M, whose components i, j, v,. . . are presupposed to be variable, stand in a relation such that everydetermination of these variable parts which makes propositions A, B, C, .. . come out true, also makes proposition M come out true. In this case wesay that proposition M is deducible or follows from propositions A, B, C,. . . or that it can be inferred from them. We also call propositions A, B,C, . . . the premisses and M a conclusion, which is obtained from them.(See Theory of Science, § 155.) The following example may clarify ourpoint. We say: if Bautzen lies to the east of Dresden, then the sun risesearlier in Bautzen than in Dresden. This simply means: the twopropositions stand in a relation to each other such that whenever wereplace the representations Bautzen and [19] Dresden so that the firstproposition still remains true, the other proposition also remains true. Solet us arbitrarily replace Bautzen and Dresden by any other town and, aslong as the first proposition is true, that is, if Bautzen is replaced by atown which actually lies to the east, such as Breslau, Warsaw,Petersburg, and so on, and if Dresden is replaced by Jena, Cologne, Paris,and so on, then the other proposition will also be true and is thereforededucible from the first. This relation is usually expressed by theexample: if A, B, C, . . . obtain, then M obtains, and it has often beenpointed out that the words if–then roughly mean the same as the words:whenever propositions A, B, C, . . . come out true, proposition M alsocomes out true. But philosophers did not know or clearly express the ideathat the propositions A, B, C, . . . cannot come out sometimes true andsometimes false, unless some of their components are considered as

place relation between an expression Q and a class of expressions P which is

axiomatizable: Q logically follows from P if and only if all models of the class P are also

models of the expression Q. Unlike Bolzano, Tarski uses the mathematical notion of a

model: if a d-sequence provides an interpretation for a set of axioms Q by which they all

come out true, then d is a model of Q. Cf. Alfred Tarski, “On the Concept of LogicalConsequence” (1936), 1956, 1983.118

“Variable” translates veränderlich, though neither Přihonský nor Bolzano used the word

“variable” in its contemporary mathematical sense. In their usage, a variable is not a letter

but refers to a constant which can be replaced and produce new statements. Following

Russell and Whitehead, Principia Mathematica, vol. 1 (1910), the statements containing

variable parts are now referred to as propositional functions; cf. Dubislav (1931), op. cit.,

p. 206.

New Anti-Kant 85

variable and that, rather than consider these propositions individually, wecan then consider the whole class of propositions which appears when wearbitrarily replace the variable parts by others.

It is only now that we can explain with sufficient clarity what wemean if we say of an actual object x that its existence is necessary, or thatthe predicate of necessity is occasionally applied to mere propositionsand truths, [20] that is, to things that do not really exist. To be precise,we mean that the proposition: object x exists, follows from purelyconceptual truths.119 Necessary is that which follows from purelyconceptual truths. An existence is accidental if it is not necessary, that is,if it is not deducible from a purely conceptual truth. Possible in the strictsense is that which does not contradict a purely conceptual truth, that is,its nonexistence is not deducible from a purely conceptual truth.120 – Yetwe often use the locutions: that is possible and that may be the case, andso on, in a much larger sense, which is not insignificant for science, whenwe wish to indicate either that the existence of this thing does not, as faras we know, contradict any purely conceptual truth familiar to us or thatits nonexistence is not deducible from a conceptual truth familiar to us ina way which is familiar to us.

119

Přihonský’s example uses both deducibility and the ground-consequence relation (Abfolge): the existence of an object

x is grounded because it necessarily

follows from purely conceptual truths;

cf. WLII, §§ 168, 221. Cf. Dubislav (1931, op. cit., p.

206, note 14).120

Přihonský apparently takes up the Aristotelian view on the relations of contradiction in

De Interpretatione, 9.19b, where propositions are modally qualified by attaching

“necessary” or “possible” to categorical statements. In addition, Přihonský hints at

Aristotle’s distinction between two senses of possibility and takes up the sense in which

“possibly p” is equivalent to “p is neither impossible nor necessary” and “possibly p”

implies or is deducible from “possibly not p”. Cf. also Bolzano’s definition of possibility:

“possible is that whose contradiction does not follow from a conceptual truth”, WLI, §

119.2.c; cf. also WLII, § 182.


5.3. New Anti-Kant

TREATISE

[21] Kant’s Critique is preceded by an introduction (pp. 35–56)*121 whichstarts with the assertion that we must distinguish between two kinds ofcognitions: pure or a priori cognitions and empirical or a posteriori ones.More than anything, he wanted to clarify this distinction beyond doubtand, in addition, to establish his other claim that philosophy is definitelyan a priori science.122 For this was not understood with a sufficientdegree of clarity at the time and is still not completely understood today,except in Germany. English and French scholars of international renownare still unable to decide in favour of such a distinction between kinds ofhuman cognitions.123 Sometimes recently, even German scholars seem to

*

121 Kant’s Complete Works, vol. 2, Leipzig 1838, Modes and Baumann, from which all

our quotes are taken. [Translator’s note: Přihonský refers to the introduction of the second

edition of the Critique, known as B.]122

Přihonský’s assertion is not quite correct because, for Kant, metaphysics or pure

philosophy, is not yet a science but investigates the conditions of possibilities for the

sciences, which are cognitive systems (cf. Prolegomena (1783) preface, pp. 255-256,

Riga, Hartknoch; KrV, B19ff; B869). This is why the main purpose of the Critique is to

show how subjective conditions of thought can provide conditions of possibility for the

cognition of objects (cf. KrV, B122).123

Přihonský’s statement is not quite correct, for this distinction already appears in Locke,

Leibniz and Hume, although they do not express it in epistemological terms, between

kinds of cognitions, but in logical terms, between kinds of propositions, and a priorirefers to internal properties of representations. Locke distinguishes between instructive

and trifling propositions. The former occur not only in the empirical sciences but also in

mathematics and ethics, where they are self-evident, in so far as those domains allow for

judgments in which the predicate expresses a consecutive but not constitutive feature of

the subject; cf. John Locke, Essay concerning Human Understanding, IV, 12, 6-8; IV, 3,

18-20. Cf. two authors who were almost contemporaries of Přihonský, Eduard Martinak

(1884), Die Logik John Locke’s, § 16, and Drobisch (1862), “Locke als Vorläufer Kants”,in Zeitschrift für exacte Philosophie, pp. 1-32. They claim that Locke’s instructive

propositions are a precursor of Kant’s synthetic judgments a priori. In addition, Leibniz

distinguishes between truths of reason and truths of fact, and Hume distinguishes between

relations of ideas and matters of fact. These philosophers distinguished between a rational

or a priori domain and an empirical domain, and one reason why Bolzano criticizes Kant

is that the latter’s notion of a synthetic a priori domain blurred the older distinction

between the domains of reason and fact. Cf. also Scholz (1931, op. cit., p. 207, who

comments that Kant’s criteria for a prioriness (universality and necessity) are based on an

insight which has not yet been replaced by anything better and that even Bolzano’s

distinction between conceptual and intuitive propositions does not achieve what it should

have done, as Přihonský admits below (p. 23). The best attempt to distinguish between apriori and a posteriori truths is Frege’s, in The Foundations of Arithmetic, (1884), p. 4:

“for a truth to be a posteriori, it must be impossible for its proof to avoid an appeal to

New Anti-Kant 87

[22] have lost sight of this important distinction and to considerphilosophical cognition as less than purely a priori. Thus Beneke and hisfollowers want to derive all of philosophy from experience.*124 Andaccording to Fries, Kant’s old supporter who has a number of admirersamongst German scholars, empirical psychology is the inmost basis of allother philosophical sciences.*125 Schelling’s modern school is justwaiting for philosophy to culminate in the combination of both kinds ofcognitions, encouraged by its master’s well-known introductory remarksto Victor Cousin’s preface on this topic.*126 [23] Although ourphilosopher so clearly understood the importance of the distinctionbetween a priori and a posteriori cognitions that he did not hesitate toplace it at the beginning of his Critique, he did not care to define thisdistinction more precisely but was satisfied with quoting examples andexpressions that are no less obscure than the very thing they shoulddefine. (On p. 35 of the introduction) we read that “even though all ourcognition begins with experience, it does not follow that it all arises from

experience”, for it might well be the case that “our cognitive faculty isinduced by sensory impressions” to “give something of itself” to ourcognition, and that this is precisely what he means by a priori cognition.Yet our author asserts soon after this (on p. 36), as though he were sayingthe same thing, that a priori cognition is completely independent ofsensory impressions. He apparently believes that he expresses himselfeven more precisely by adding “independent not of this or thatexperience, but of all experience whatsoever”. “Cognition”, he continues,“is empirical if it is possible only by means of experience”, as opposed topurely a priori cognition, “which is not tinged with experience”. So, for

facts, i.e., to unprovable truths which have no universality and contain assertions about

particular objects. If, on the other hand, it is possible to establish a proof on universal

laws alone, which themselves neither need nor admit of proof, then the truth is a priori.”(My translation.) But even this explanation, which is limited to true judgments, is not

completely satisfactory, since it contains too many indeterminate terms.

*124

In the supplementary pages of the Haller Literary Journal, J. 1837, p. 495, in a review

of Hartenstein’s work, The Problems and Fundamental Doctrines of General Metaphysics,

Beneke explicitly calls the problems and basic doctrines of general metaphysics the most

important difference between Herbart’s philosophy and his own. According to Herbart,

philosophy is a cognition obtained (a priori) from mere concepts, whereas in his view,

philosophy is an extension of mental experience (which he also calls inner experience).

*125

Manual of Psychological Anthropology, vol. 1, 2nd edition, 1837 (Jakob Friedrich

Fries, Handbuch der Psychischen Anthropologie, 1-2, Jena: Cröker, 1820, 1821).

*126

E.g., “it is impossible to attain reality by purely rational means, etc.” (See Victor

Cousin on French and German philosophy, the German translation by Dr. Hubert Beckers,

professor of philosophy at the Lyceum at Dillingen, and the critical preface by Privy

Councillor v. Schelling, Stuttgart and Tübingen near Cotta, 1834.)


instance, the proposition “‘every change has a cause’ is an a priori

proposition but not a pure one, because every [24] change is a conceptwhich can only be derived from experience.”

Here anyone will notice right away that the concept designated by theword experience is already presupposed as familiar, without any need fordefinition. When defining the distinction between a priori and a

posteriori cognitions, is it not fitting to define the concept of experienceitself?127 In our view, the second concept is more complex than the first,and we define experience by virtue of defining conceptual and intuitivepropositions. – Further, what is that supposed to mean: as far as time isconcerned, a certain kind of cognition begins with experience but doesnot arise from it? Are we no doubt supposed to understand by this that apriori cognition is not grounded in experience? But is this true? Are theorigins of our a priori cognitions, such as mathematical cognition (whichKant quite particularly counts among a priori cognitions), not in theleast, or even partly, grounded in experience?128 Surely nobody would or

127

Přihonský has a point that Kant does not give a clear account of experience in B1-3

(introduction I-II), but Kant makes an attempt in the Prolegomena, where he explains that

experience is “the product of the senses and the understanding” (P, § 20) and a “synthetic

connection of appearances (perceptions) in consciousness, so far as this connection is

necessary” (P, § 22). Experience is ordered according to the principles of reason, so that

reason perceives that which it produces according to its own laws. Kant’s notion of

experience as a mental construction also appears in the Dissertation of 1770, where he

says that experience is “the reflective cognition which arises when several appearances are

combined by the understanding” and adds that “the only way from appearance to

experience is through reflection in accordance with the logical use of the understanding”

(§ 5), so that experience as a reflective cognition is already determined by the

understanding. In view of Přihonský’s objections above, it may be useful to keep in mind

that Kant wants to investigate the conditions of the possibility of experience, and that in

his account, experience is synthetic and may be parsed into elements which belong partly

to the senses and partly to the understanding. In addition, in his view experience is

necessary not because, as the British empiricists claim, it is the source of our impressions,

from which we derive our ideas and concepts, but because it is determined by judgments

containing a priori concepts of the understanding, such as [cause]. For Přihonský,

however, experience can only be defined on the basis of the distinction between

conceptual and intuitive propositions.128

Here Přihonský gives a preview of his objections against Kant’s problematic claim that

mathematical cognition is based on pure intuition (cf. NAK, pp. 45ff). But Kant also

explains that mathematics is the only a priori science “arising from reason” which we can

learn (and therefore come to know a posteriori) because mathematical cognition is

cognition through the construction of concepts; cf. KrV, B865. Here Kant refers to his

claim that mathematical propositions are constructed by means of intuition, and

mathematical concepts are generated by schemas providing them with a particular

presentation, such as the drawing of a triangle; cf. KrV, B205, B714-B714. He admits that

“although [mathematical] principles are generated in the mind completely a priori, they

New Anti-Kant 89

could assert such a thing. Otherwise, how could it happen that we onlyobtain a priori cognition once we have had some experience? Why doesall a priori cognition begin with experience if it is not conditioned by itand is thus at least partly [25] grounded in experience? – How can Kantsay that a priori cognition is not only independent of experience but ofall experience whatsoever, if he is forced to admit, as he does, thatexperience is necessary as the cause or condition of its origin? – We arecompletely confused by the added example, according to which theproposition: every change has a cause, is an a priori proposition that isnot purely a priori, because change is a concept that can only be derivedfrom experience. In our view the concept change is as pure as anyconcept, since saying that a thing changes surely only means that atdifferent times it has different and contradictory properties. So theconcept of change simply contains no other components than theconcepts substance, property, time, and so on. Therefore, it consists onlyof pure concepts and contains not a single empirical representation orintuition.

As if he had sensed the insufficiency of his definition of thedistinction between a priori and a posteriori cognition, our philosopherreturns to this topic in the second paragraph of the introduction although,in accord with its title, it no longer deals with the determination of thoseconcepts (it reads: “we possess certain a priori cognitions which even thecommon understanding never lacks”).[26] Kant says: “what matters hereis a reliable mark by which we can distinguish a pure cognition from anempirical one” (p. 36). – Yet we believe that if the inmost nature of thedistinction between a priori and a posteriori propositions had beenascertained, no further marks would be necessary for distinguishingbetween them or, instead, that the clearest marks would be found in thecomponents of their concepts. Kant indicates two marks fordistinguishing between a priori and a posteriori cognitions: necessity anduniversality. “First, if we find a proposition which is also thought asnecessary, then it is an a priori judgment.129 Second, experience neverconfers true or strict universality on its judgments but only assumed andcomparative universality (through induction), so that it is actually

would mean nothing if we were unable to present their meaning [. . .] in empirical

objects.” (KrV, B298-B299)129

Přihonský omits: “and if, in addition, it is not derived from any other proposition except

one which also has the validity of a necessary judgment, then it is purely

(schlechterdings) a priori.” (KrV, B4) Here Kant explains the difference between a prioriand purely a priori (the lack of which Přihonský criticizes above) in terms of logical

consequence.


supposed to mean: whatever we have perceived so far is no exception tothis or that rule. So if a judgment is thought with strict universality, suchthat no possible exception is admitted, then it is not derived fromexperience but is valid purely a priori” (p. 37). How far have we gotwith the two marks?

As to the locution: a judgment has necessity, it is taken in differentsenses. In a certain sense, when we take the necessity [27] of a judgmentas a relation between the latter and our faculty of judging, by means ofwhich we feel constrained to make this judgment, that is, weunsuccessfully attempt to accept its contrary as true—in this sense quite afew judgments are lacking necessity.130 But in Kant’s view we mustconsider them as a priori cognitions, for otherwise his distinction wouldbecome so shaky that it would no longer serve his purpose. In this sense,the judgment “that the relation between diameter and circumference isirrational” is nothing less than an a priori judgment thought withnecessity, since it is known that there were mathematicians who doubtedthat this was the case. Nonetheless, Kant rightly considered allmathematical judgments as a priori cognitions. In yet another sense, Kanthimself distinguished between apodictic judgments expressing necessityand problematic or merely assertoric judgments, so that the proposition:God must reward goodness, is an apodictic or necessary judgment; theproposition: every finite being can sin, is a problematic judgment; andfinally, the proposition: Adam has really sinned, is an assertoricjudgment.131 But it is clear that purely conceptual truths can also belongto all three kinds of propositions. For example: the hypotenuse must

130

This is Kant’s view in the Jäsche Logic (intro. IX. D) and the Dohna-Wundlacken Logicwhen discussing the distinction between truth, which concerns a judgment in relation to

the object of knowledge, and holding-to-be-true (with varying degrees of confidence),

which “is judgment in relation to the subject” (D-WL, p. 732). In fact, both Kant and

Bolzano are concerned with the subjective justification of a judgment, and Přihonský

takes up Bolzano’s criterion of impermeability or indestructibility, which Bolzano uses for

distinguishing between knowing and believing (wissen und glauben) as two mental states

with regard to propositions and judgments. I believe that p if and only if I have cognized

the truth of p and understand it, but my confidence in the explanation of p can be

destroyed. I know that p if and only if I have cognized the truth of p and understand it, so

that my confidence in the explanation of p is indestructible; that is, even if I wanted to, I

could not judge that not p; cf. WLIII, § 321, and Přihonský’s comments on Kant’s

distinction between opining, knowing, and believing, NAK, pp. 206-208, and my note 559.131

Přihonský refers to Kant’s classification of the forms of judgment, more precisely, the

forms of judgment according to the category of modality (KrV, B95). Bolzano criticizes

Kant’s modal judgments in WLII, §§191.2, 193.5, but Přihonský does not take up

Bolzano’s objections here.

New Anti-Kant 91

always be bigger than each of the two legs [of a right-angled triangle];132

it can have a rational as well as an irrational relation to the legs; [28] inthis triangle it really is irrational. There is, however, a sense in which wecan justifiably claim that necessity is not only a mark but also a universaland exclusive mark of a priori truths: namely, if we take the wordnecessity in its most characteristic sense. Still, this sense is so unsuitablefor distinguishing between a priori propositions and a priori truths thatconversely we can only judge whether something is necessary if we canjudge whether the proposition shown to us is a priori. In this sensesomething is necessary if and in so far as it is a purely conceptual truthor can be inferred from a purely conceptual truth or (which is exactly thesame) in as much as the opposite contradicts any purely conceptual truth.

Things are not much better for the second mark, which is equallyambiguous. In a certain sense universality also belongs to empiricalpropositions and indeed to all propositions in general, although inanother sense even a priori truths can lack universality. If we understandby the universality of a proposition merely that its predicate is applicableto all objects falling under the subject-representation, then every true andrightly expressed proposition must have universality. The propertyconstituting the real predicate-representation must always [29]

accompany all objects standing under the real subject-representation.133

For every proposition is included in the form A— has—b where A is thesubject-representation and b the predicate-representation giving a certainattribute to the subject of the proposition (or the lack of such an attribute,in negative propositions).134 Therefore universality is not a specificfeature characterising only and exclusively a priori truths: for example,the empirical proposition: Niobe’s children all died before her, hasuniversality if it is otherwise true. – In another sense, universality isconferred only on propositions whose subject-representation comprisesseveral objects, and in this sense there are a priori truths without themark of universality; for example, God is almighty; the supreme morallaw is a practical truth. Those who (in a third sense) oppose universalpropositions and particular propositions, such as: some men aredepraved, cannot claim that universality is a property which belongs toall a priori truths, because in this last sense, purely conceptual truths

132

[Of a right-angled triangle] is my addition.133

On Přihonský’s use of stehen unter, cf. note 14, above. N.B. In ML, § 5, Bolzano

distinguishes the relations between ideas with respect to their extension, such as inclusion,

subordination, reciprocity, and compatibility.134

Cf. note 4, above, on Bolzano’s canonical form of propositions, which Přihonský refers

to. On negative propositions, cf. note 5, above.


such as: some triangles are equilateral, and so on, must be considered asparticular truths.

However, quite frankly, we do not accept the last distinctionaccording to which logicians divide propositions into universal andparticular ones, [30] in the sense that there are some propositions inwhich the predicate-representation is applicable to all objects of thesubject-representation and other propositions in which the predicate-representation is applicable to only some of them. We think, on thecontrary, that in every true proposition the predicate-representation refersto all objects of the subject, precisely in the so-called particularpropositions, for in them the subject is not “some A”; instead, thesepropositions should be conceived in the form: A cum b has objectuality.The proposition: some humans are erudite, certainly does not mean thatthe predicate erudition applies to every collection of some humans(which is surely wrong) but that to be human and to be erudite forms anobjectual representation.

So although we cannot accept the way in which Kant distinguishesbetween a priori and a posteriori cognitions, we do admit that thereactually is such a distinction and that Kant is the first to have pointed itout somewhat clearly. But there is another convincing view on theinadequacy of his conception of this important distinction.

The way in which Kant attempts to distinguish between a priori and aposteriori propositions is essentially bound to the circumstance that thereare mental propositions or judgments [31] which this or that human beingmakes and holds as true, or that there are true judgments which arecognitions. His distinction is applicable to cognitions or true judgmentsalone, for these are the only ones of which we can assert that in a certainsense they are entitled to necessity and universality. However, a closerexamination shows that both true and false judgments, and evenjudgments and propositions as such, regardless of whether anyone holdsthem as true or whether they are grasped in the mind of some thinkingbeing, are distinguishable in a way that deserves attention and mayprovide a good basis for distinguishing between a priori and a posteriori

propositions.135 There is an essential distinction between both true

135

Přihonský takes up Bolzano’s objection to the Kantian identification of truth with the

knowledge of truth: truth is a property of the content of mental acts rather than of the

mental act itself; consequently a judgment may be true or false, but a cognition

necessarily has a true content. The term ‘cognition’ is therefore synonymous with ‘true

judgment’. Cf. WLI, §§ 26; 36-37. Kant rather unfortunately identifies judging with

knowing when he says that a judgment is the mediate cognition of an object: every

judgment is a cognition, and every cognition is a judgment. So, absurdly, he claims that

we can have false cognitions. Cf. KrV, B93, where Kant defines a judgment as the

New Anti-Kant 93

propositions such as: every limited and reasonable being is subject tovice, the sphere is the most perfect body, every proposition must have acopula, and so on, and also false ones, such as: all created substancesoriginated in time, the diameter of a circle and its circumference are in aratio of 1:20, truth consists in the accordance between representationsand objects, and so on, on the one hand, and true or false propositions:this plant belongs to the genus of lilies, Aldebaran is a twin star, there aremonkeys without tails, and so on, on the other. We believe that in orderto determine what the distinction between these two kinds of propositionsactually consists in, we must establish it [32] as follows: a priori

propositions consist of nothing but concepts, whilst other propositionsalso contain intuitions. If Kant had known this difference and used it as astarting point, he would hardly have made the mistake of giving thoseempty and false definitions and distinctive marks that we have seenabove. It almost appears as though he wanted the determination ofcognitions as a priori or a posteriori to depend on the arbitrary conditionof how we obtain a cognition, either through experience or through merethinking. But how can such a shaky and uncertain distinction provide adecision to the question: to which of the two classes does a cognitionbelong? Are we not obliged to admit that we obtain most and, to someextent, all cognitions, even a priori ones, by means of experience? –What a difference between this and the account that we have given of thisdistinction! For although the scientific definition of an intuition and aconcept is subject to some difficulties, there is hardly anyone who cannotdistinguish between an intuition and a concept, and vice versa, and whofor this reason is unable to determine whether a proposition is composedof only concepts or whether there are one or several intuitions among itscomponents. In addition, as we pointed out, Kant relates everything tomere cognition. Our [33] distinction however has the advantage of beingabsolutely objective, since it does not depend on certain externalrelations but is based on an internal property of the object itself, aproperty which already belongs to propositions as such and not just totheir appearances in the mind (judgments and cognitions). That is why itmight almost be preferable to drop the misleading terms: a priori and aposteriori and to refer to conceptual and intuitive propositions instead.

Once Kant has acquainted us with the essence of a priori cognitions,he puts forward the principle that we possess certain a priori cognitionswhich he corroborates with examples from mathematics and purephysics, and with this we are in complete agreement. For there actually

mediate cognition of an object, or as an action by means of which representations become

cognitions of an object.


are many pure conceptual propositions, both among propositions as suchand among our judgments, which we express in full confidence as trueand right and consider as cognitions, because the statements of allreasonable human beings will concur at all times. And anyone can acceptKant’s assertion “that philosophy needs a science that determines thepossibility, the principles, and the range of all a priori cognitions”(p. 39)136 if they are convinced that the above-mentioned distinctionbetween cognitions is right. [34] Nevertheless Kant brings up one moreimportant distinction between cognitions or rather between truths (truepropositions), which remains yet to be fully appreciated (pp. 42ff.)137. Hepoints out that “in all judgments there is a twofold relation betweensubject and predicate.138 Either the predicate B belongs to the subject A as

136

This is the title of section III of the introduction (B), KrV.137

Přihonský refers to Kant’s distinction between analytic and synthetic judgments, KrV,

B, intro. IV. Cf. Scholz (1931), op. cit., pp. 209-210, on this distinction. He refers the

reader to Couturat’s (1905) critique of the Kantian distinction in Les principes desmathématiques avec un appendice sur la philosophie des mathématiques de Kant, Paris,

reprint Olms, 1979, and Dubislav (1926), Ueber die sog. analytischen und synthetischenUrteile, Berlin, H. Weiss. Scholz also gives an outline of Bolzano’s interpretation of

analytic judgments which is based on the form of statements which I summarise: we

explicate (verständigen uns über) each statement by assuming that it contains at least one

variable and at least one invariable component. We obtain the form of a given statement

by replacing at least one of its components by a sign for a variable. The forms of

statements obtained in this way can be divided into three classes: those which always turn

out true, those which always turn out false, and those which sometimes (at least once) turn

out false. According to Bolzano, a statement is analytic if and only if, by way of a

meaningful charge as regards the content of the empty places in the statement-form, it can

be obtained either through a statement-form which always turns out true or a statement-

form which always turns out false. Otherwise the statement is synthetic. This criterion has

a completely different capacity and efficiency than Kant’s criterion of inclusion. It is

another question whether the distinction obtained in this way is practical and useful.

Scholz suggests that we introduce the concept of perfect statement-form or a form in

which all components which we explicate as variable are replaced by the appropriate

signs. A statement is analytic, if and only if it can be obtained through replacement

(Einsetzung) from a perfect statement-form which always turns out either true or false.

Bolzano calls such analytic propositions analytic in the narrow sense or logically analytic

propositions, and he opposes them to analytic propositions in the broader sense (cf. WLI,

§ 148.2), so Scholz’s suggestion corresponds to the restriction of Bolzano’s analytic

propositions to logically analytic propositions.138

Přihonský slightly modifies the Kantian text which reads “in all judgments in which the

subject-predicate relation is thought (if I consider only affirmative judgments, since a

subsequent application to negative ones is easy), this relation is possible in two different

ways” (KrV, B, intro. IV). Přihonský’s version that “in all judgments there is a twofold

relation between subject and predicate” a) unlike Kant’s text reduces all judgments to the

subject-predicate form and b) is much stronger than Kant’s claim that “this relation is

possible in two different ways”.

New Anti-Kant 95

something that is (covertly) contained in this concept A, or B lies entirelyoutside the concept A, although it is connected with it.” He calls theformer judgment analytic and the latter synthetic. He gives the followingexample of an analytic judgment: all bodies are heavy.139 For in order toconnect extension with the subject: body, it is not necessary to go beyondthe concept of body at all, since I find this predicate in it when I analysethat concept. On the contrary, the predicate: heavy is something entirelydifferent from that which I think in the mere concept of a body ingeneral.

We note that although we take the distinction between analytic andsynthetic judgments as one of the most auspicious and influentialdiscoveries ever made in the domain of philosophical research, it seemsto us that Kant has not grasped it with asufficient degree of clarity.140 Buteven if we were to ignore that, contrary to a well-known logical rule, heincludes metaphors in his definition of concept [35] and uses expressionssuch as “covertly” or “outside the concept”, we cannot help criticizinghis definition of analytic judgments as being too large. For propositionsthat surely nobody would count as analytic, such as: the father ofAlexander, king of Macedonia, was king of Macedonia, and a trianglethat is similar to an isosceles triangle is itself isosceles, and so on, areanalytic by Kant’s definition.141 We could counter this mistake by saying

139

This is a misprint (cf. also Morscher’s 2003 edition, p. 186), for Kant’s famous example

is exactly the opposite: “all bodies are heavy” is a synthetic judgment, whereas “all bodies

are extended” is an analytic judgment.140

Přihonský follows Bolzano, who repeatedly thanks Kant for making the distinction

between analytic and synthetic judgments, whilst developing the implications of the

Kantian distinction as well as criticizing its shortcomings; cf. WLI, § 120; WLII, § 148.

Cf. also my note 106.141

Přihonský uses Bolzano’s example of a nonanalytic proposition where a part of the

subject is included in the predicate: “The father of Alexander, king of Macedonia, was

king of Macedonia” (WLII, § 148). Bolzano uses this example against Kant’s criterion of

inclusion as a mark for distinguishing between analytic and synthetic judgments and

shows that the inclusion of the predicate in the subject is not a necessary mark of analytic

propositions and that although A can include B, it does not follow from this that B is

necessarily a property of A. Inclusion is an insufficient criterion for the (syntactic)

definition of analyticity. Bolzano’s analytic/synthetic distinction is based on the criterion

of substitution, an objectuality constraint, and an equi-veracity constraint: a proposition is

analytic if and only if I can replace either the subject- or the predicate-component of a

proposition without disturbing its truth or falsity, and if and only if the proposition is

objectual. This substitutional account enables him to draw a further distinction between

logically analytic propositions (tautologies) and analytic propositions in a wider,

nonlogical sense, such as “a depraved man does not deserve respect”; cf. WLII, § 148.

As to Kant, we should note (unlike Přihonský) that Kant also bases the distinction

between analytic and synthetic judgments on the principle of contradiction; cf.


that analytic judgments are those in which the predicate is an essentialpart of the subject, but our definition would thus fall prey to the mistakeof being too narrow. It would fit analytic judgments of the form: “A

which is B, is B”, but by no means judgments of the form “A which is B,is A” and “every object is either B or not B” which, obviously, are alsoanalytic. Kant’s definition of analytic judgments and therefore also ofsynthetic judgments which are opposed to them is unsatisfactory and allthe more so as it does not touch the true essence, the difference whichshould be most important to a philosopher. The truth or falsity of certainpropositions (and these are, precisely, only the analytic ones) in no waydepends on those particular representations [36] composing thesepropositions, but they remain true or false, regardless of the changes weproduce in one or the other of their representations, as long as theproposition itself retains its objectuality. The two propositions: aconditioned being is no God and a conditioned being is God are thereforeanalytic propositions. One of them is true and the other false, regardlessof the changes we produce in them by replacing them with therepresentations: angel, man, animal, plant, or any representationwhatsoever, as long as the resulting propositions still retain theirobjectuality.142

Prolegomena, § 2; KrV, B190-191. The young Bolzano (1810) agrees with Kant that

analytic propositions are based on the principle of contradiction; cf. Appendix on theKantian construction of concepts through intuitions, § 1. In addition, Kant made an

important linguistic point, for this principle is not only a logical law, but also a linguistic

rule for meaningful language use. Cf. E. W. Beth (1953) in “Kant’s Einteilung der Urteilein analytische und synthetische”, pp. 253-264. Beth gives the following example: the

principle of contradiction grounds the analytic judgment: “all witches are dangerous”. For

according to custom a woman engaged in magic is designated as a witch only if this

activity is dangerous to others, so that the statement “this witch is not dangerous”

contradicts meaningful language use and is thus rejected on the grounds of the principle of

contradiction.142

Přihonský refers to Bolzano’s objectuality constraint: we can vary the components

[being] and [God] and those variants will be referential where the substitute for the

demonstrative denotes an object among the objects denoted by the [being] and the [God]

substitutes. The problem with representations such as [God] and [angel] is that in order to

conform to the objectuality constraint, we must presuppose either a certain collection of

beliefs or an interpretative act: we must interpret these representations as objectual ones,

but presumably a Catholic priest would not consider this a problem. N.B. Bolzano gives

various proofs of the existence of God in RW I, § 67, Athanasia, note 73, pp. 321-326,

and WLIII, § 330.1, which are based on Kant’s threefold division (ontological,

cosmological, and physicotheological) rather than on scholastic theology, such as

Aquinas’ five ways for proving the existence of God. Cf. W. Löffler (1999), “Bolzanoskosmologischer Gottesbeweis im historischen und systematischen Vergleich”, pp. 295-316.

New Anti-Kant 97

Propositions where this is not the case and which are not composed ofsuch variable representations as: God is omniscient, there is a suprememoral law, the sun has warmth, and so on, are synthetic.143 In this accountof analytic and synthetic propositions, the latter’s existence isundeniable, but this does not result from the Kantian determination. Kantdefines synthetic propositions as those in which the predicate is notcontained in the subject, but he does not provide sufficient proof thatthere are such propositions and only points them out in the domain ofmathematics and of pure natural science. Instead, he should have shownthat there are properties all belonging to objects standing under a certainconcept, but that the [37] concept of these properties is not contained as acomponent in the concept of these objects. For this proposition has thusfar remained unclear to several logicians, which is precisely why they aredetermined to reject the existence of synthetic propositions.*144

But Kant ignores all this, as though he were in a hurry to proceed inanswering the question that is so important for his critique of reason: how

are synthetic judgments formed?145 This is why he distinguishes betweensynthetic judgments of an empirical nature, or empirical judgments, andsynthetic a priori judgments.

Regarding empirical judgments, which are entirely synthetic, it seemsthat Kant believed that their origin could be explained without difficulty.For even if the predicate is not contained in the subject, in suchjudgments the predicate appears in some intuition connected with thesubject, and for this reason we are justified in conferring the former on

143

Přihonský’s statement is incomplete, for in synthetic propositions, the components can

be sometimes true and sometimes false. In addition, in a synthetic proposition not a single

one of the components can be replaced without affecting the truth or falsity of the

proposition; cf. Bolzano’s claim in WLII, § 148.1.

*144

See Dr. Fr. C. Beneke’s System of Logic, Berlin, 1842, pt. 1, p. 156f.; Dr. H. Ulrici’s

History and Criticism of the Principles of Modern Philosophy, Leipzig, 1845, p. 15, where

the separation between analytic and synthetic judgments is bluntly called “unfounded”,

and so on.145

Kant’s question is: “how are synthetic a priori judgments possible?”, in order to find an

answer to the epistemological question: “how is a priori cognition of objects possible?”

and the philosophical question: “how is metaphysics as a science possible?” Cf.

Prolegomena, § 4, KrV, intro. B, sec. III and IV; B19ff. Cf. on this Dubislav (1931), op.

cit., p. 210, who comments that Přihonský interprets Kant’s famous question in the

following sense: a) how does a cognizing human being interpret synthetic a priorijudgments, given that there are such judgments? and b) how can we explain that these

synthetic judgments are a priori truths, unless we presuppose, as Kant does, that they all

belong to the class of immediate cognitions? Hence Přihonský interprets this question in

the sense of the historical Kant who answered it by his doctrine of the sources of

cognitions, which is partly entangled in the doctrine of the old psychology of human

faculties.


the latter. For example, this is how we form the judgment: all bodies areheavy, by taking the mark of heaviness which, apparently, is notcontained in the concept of a body, [38] and conferring it on the latter(pp. 43ff.). – Is this sufficient to explain the origin of empiricaljudgments? Can a single intuition that we have determine our making thejudgment: all bodies are heavy? Surely it only suffices for expressing thejudgment: this body is heavy. Kant nowhere explains how we obtainuniversal judgments by means of a single intuition.146 Kant’s reasoningdoes not even explain the origin of the singular judgment: this body isheavy.147 For if things really were as Kant asserts, the predicate “heavy”would be contained in our intuition of this body, that is, in therepresentation constituting the subject-representation of this judgment,and the latter would be merely analytic, which it should not be, accordingto Kant’s own assertion (p. 44), nor is this the case. In the judgment:“this body is heavy”, the subject-representation “this body” does notcontain anything of the representation of heaviness but intuitions of adifferent kind, such as colour, smell, shape, and so on, and therefore hasthe following form: the object which causes the intuitions of colour,

smell, and so on. Yet the judgment tells us that the same object is also“heavy,” and this means that the object also causes certain [39] other

intuitions that I have, such as the pressure I sense, and so on. Thus it isobvious that many intuitions are presupposed so that this singularjudgment can be formed and, in addition, they must have been repeatedseveral times. In other words, to infer with some probability that thesame object which caused some of my earlier intuitions also causesseveral subsequent ones, I must already have had repeated intuitionsbelonging to the concepts: colour, smell, and so on, simultaneously withthe intuition contained in the concept of a body’s pressure or its weight.Therefore all synthetic judgments a posteriori have the following form:the same object x, which causes the intuitions belonging to the concept A,

probably also causes the intuitions contained under the concept B. In thisway we also explain how judgments originate in general.

146

A single perception of a body does not ground our general judgment that all bodies are

heavy, nor does Kant say this, but Bolzano uses this argument to object against Kant that

the origin of empirical judgments cannot be explained by referring to a connection

between the subject-representation and an intuition; cf. WLIII, § 305.2.147

Likewise, Bolzano objects against Kant that a single intuition is not even sufficient for

grounding a single synthetic proposition, such as “this body is heavy”. For this

proposition requires the intuitions of colours and smells belonging to the subject-concept

“this body”, as well as the intuition of heaviness expressed by the predicate-concept. So

the subject-predicate relation of empirical judgments consists of a conceptual

representation and several intuitive representations; cf. WLIII, § 305.2.

New Anti-Kant 99

But perhaps we can expect that our philosopher will be moresuccessful in defining synthetic a priori judgments, although he himselfconsidered this definition to be far more difficult. – Yet a large part ofthe Critique deals with their explanation,148 and our task will be todetermine whether it is successful. Here Kant only wants to introduce usto the response to his main question concerning the origin of synthetic apriori cognitions, [40] and he does this as follows: “Here everything justdepends (he says, p. 45) on what is the unknown x on which theunderstanding relies when it believes that it discovers, outside theconcept A, a predicate that is foreign to that concept but which itnonetheless believes to be connected with A.”149 But we think that thedifficulty raised here is contrived and that the issue is not as odd as Kantasserts. The reason why our understanding ascribes to a subject A apredicate B which is not contained in the concept A is none other thanthis: we have and know both concepts A and B. We are able to makecertain judgments about these concepts just because we have them. For tosay: we have the concepts A, B, C, . . . surely means nothing other than tosay: we know and distinguish them, which means as much as saying: weassert something of one of them which we do not want to assert of theother, that is, we make a judgment about them. Since this inference isuniversally valid, surely it is also valid when these concepts are simple,in which case our judgments are certainly synthetic. There is no doubtthat by means of synthetic judgments composed of simple concepts, forexample, A, B, C, . . . we can also make various synthetic judgmentsabout complex concepts that are constituted by a connection between thesimple concepts A, B, C, . . . which holds either between these conceptsor between them [41] and other concepts. So it is by means of theconcept we have of an object that we can make a synthetic judgmentabout any object of which we have a concept.

We admit that our explanation of the origin of synthetic judgments isa long way from Kant’s. For as we shall soon see, in the Critique Kantattempts to prove that also in this kind of judgment, the predicate isdiscovered in an intuition connected with the concept, since it is notcontained in the subject. Consequently, the unknown x upon which theunderstanding should rely when making synthetic a priori propositions isan intuition.150

148

Here “explanation” translates Erklärung, rather than “definition”, because the passage

concerns the account of the origin of synthetic a priori judgments.149

Cf. KrV, intro. IV, B13.150

Cf. Dubislav (1931, op. cit, p. 210), who comments that Přihonský’s remark only

applies to Kant’s doctrine of those synthetic judgments which belong to mathematics. This


But if, as shown above, even the origin of synthetic a posteriori

judgments cannot be explained by connecting the subject with anintuition, this method is even less useful with a priori judgments. Forhere, too, we can reply that since the concept of the subject in thesejudgments contains countless, or even infinitely many, objects, whereasthe intuition combined with it can only represent a single object, theinference from a single object to all of them is certainly partly invalidand partly lacking the necessary degree of conviction that its contrary isimpossible, [42] which we commend in judgments of this genre. – Still,Kant would answer: the intuitions through which synthetic a priori

judgments are mediated are of a most particular kind; they are notempirical but pure intuitions, and because of them, the judgment theyground is not merely probable but becomes certain and necessary.Nevertheless, we must confess that we cannot really perceive adistinction of the kind Kant describes (see below) in the sum total of ourhuman intuitions. If, however, there were such a distinction, it would beclear that our philosopher has not sufficiently done his duty by merelyassuring us that pure intuitions ground all right synthetic a priori

judgments, for he should also have shown how these judgmentsoriginated from those intuitions. But as far as we know, Kant has notdone this anywhere, as we will soon demonstrate.

However we must agree with his assertion that “synthetic a priori

judgments are contained as principles in all theoretical sciences ofreason” (p. 46).151 However, such judgments do not only occur inmathematics, pure natural sciences, and metaphysics, which Kant provesirrefutably, but also in logic. And they occur not only in those doctrineswhich belong to a larger concept of this discipline, that is, if we takelogic as a theory of science in Bolzano’s sense, [43] but even in the so-called analytic part of logic, which has been studied since Aristotle. Forinstance, the principles that every judgment has a copula and that thereare nineteen forms of syllogisms are synthetic judgments.152

important restriction was forgotten not only by Přihonský but also by Bolzano (cf. WLIII,

§ 305).151

Cf. KrV, intro. V, B14.152

Bolzano objects against Kant that formal logic (or logic in the large sense, as a theory

of science) must take into account how we obtain knowledge and, consequently, must

include synthetic truths. Cf. KrV, B78. Kant considers formal logic as a closed science

which has a normative use, but a correct application of logical laws cannot increase our

knowledge; cf. KrV, B.viii. According to Kant, dialectical logic contains synthetic

propositions, but it cannot increase our knowledge any more than formal logic can, since

it is a logic of illusion which examines errors of reason. The activity of reason is

restricted to making inferences (cf. KrV, B169); but these inferences are valid only if

New Anti-Kant 101

At the end of the introduction Kant explains how he was led to theidea of a special science for examining the question “whether and to whatextent synthetic a priori judgments are humanly possible and what valuedo they have?” (pp. 49–50)153 He calls this science the Critique of Pure

Reason and divides it into a “Transcendental doctrine of elements” and a“Transcendental doctrine of method” (pp. 52ff.). He calls these parts ofthe Critique and his philosophy as such transcendental, because theycontain cognitions referring to objects of which an experience ispossible,154 as opposed to transcendent philosophy that deals with objects

reason is confined within the boundaries of experience; cf. KrV, B382-385. Only the

analytic part of transcendental logic, which is a logic of truth, examines the conditions

and possibility of knowledge; cf. KrV, B76; B87-88. However, Bolzano claims that if

logical laws are correctly applied, we can discover new truths, and thus the correct

application of logical laws can increase our knowledge. Unlike Kant’s synthetic

propositions a priori, which do not belong to formal logic because they are based on pure

intuition, Bolzano’s synthetic propositions, if a priori, are conceptual and belong to

formal logic. For Kant, inference rules are the pure principles of the understanding and of

reason, and in his view they are synthetic, because the subject-concept is not contained in

the predicate-concept. Cf. JL, § 53. According to Kant, the problem with synthetic

principles of reason is that if they are a priori, reason is unrestricted by experience and

extends itself beyond its legitimate limits. Cf. KrV, B382ff. As a result, reason is caught

in contradiction by making opposed but equally valid inferences (Vernunftschlüsse). Cf.

KrV, B454. In addition, Kant does not distinguish between an inference rule and its

application: the rule is the major proposition in a syllogism; cf. JL, § 58. Unlike Kant,

Bolzano distinguishes between a rule and its application when he says that the rule of

transitivity that from two propositions “A is B” and “B is C”, follows “A is C” is a

synthetic a priori truth or truth-bearer, whereas a particular application of this rule is an

analytic truth; cf. WLIII, § 315.2. Synthetic propositions, for Bolzano, do not have a

component that can be varied arbitrarily whilst conserving the original truth value of the

proposition, as do analytic propositions; cf. WLII, §§ 148, 197. Inference rules are

synthetic because their components are already variables, so that a substitution would be

invalid. They express a relation between indeterminate ideas, such as: “If (if p then q) and

p, then q”, which is not part of the inference itself, but when it is applied in an inference,

the particular application expresses the relation between determinate ideas which can be

replaced by others. If inference rules are synthetic propositions as, by the way, are

axioms, it follows that for Bolzano, analytic propositions are derived from, or depend on,

synthetic propositions. Consequently, in his view formal logic includes analytic and

synthetic propositions, and the former are grounded by the latter. Here he is in direct

opposition to Kant, who claims that formal logic is based on the principle of contradiction

and contains only analytic propositions, because its purpose is “to be a science of the rules

of thinking in abstracto” (JL, intro.).153

This quote does not appear as such in Kant’s text; instead, Přihonský summarises § VII

of the introduction to the Critique, B25-26.154

According to Kant, cognitions are transcendental if and only if they deal, not so much

with objects, as with our mode of cognizing objects, insofar as this is possible a priori (cf.

KrV, intro. VII, B25). He does not mention experience at this point, though Přihonský’s

general interpretation is correct.


that are not given in any experience. In addition, Kant splits the doctrineof elements into a doctrine of sensibility (the “Transcendental aesthetic”)and a doctrine of thinking (the “Transcendental logic”) because, in hisview, sensibility and understanding are the two stems of humancognitions, in that objects are given to us through the first and arethought by the second (p. 56).155 – Here we [44] only wish to remind ourreaders that we cannot accept the claim that “objects are given to usthrough the senses and thought by the understanding”. Still, we couldclaim that sensibility, which we take in the Kantian sense as the capacityfor having intuitions, gives us objects, if we interpret this as follows:from each one of our (subjective) intuitions we are allowed to infer thepresence of a certain object that causes these simple representationswhich refer to it alone (which represent it).156 In our opinion, however, itwould be very wrong to consider intuitions as the only representationswith an object or giving an object, that is, enabling us to make anappropriately true judgment. There are infinitely many otherrepresentations that we have and of which we can ascertain that they havecertain objects standing under them, just as perfectly as we can ascertainthis of intuitions. Occasionally these objects are of a kind that nointuition could acquaint us with or teach us, such as the concepts:something, truth, representation, and similar ones. We will return to thisimportant point several times in the course of our investigations, and wewill have the opportunity to explain our views on this issue in more detailto our readers. – That is why we should proceed without delay to Kant’sTranscendental Doctrine of Elements and begin with its first part, theTranscendental Aesthetic. [45] We may let the following question rest,especially when considering that the answer appears anyway, as a resultof our discussion: is it necessary to put forward the transcendental

aesthetic as a special science dealing with the forms of sensibility (oftime and space), and would it not be more appropriate to classify allthose principles which in Kant’s view should be transferred to thisscience, in another, already existing science? It is no doubt moreimportant and fitting for our purposes to emphasise Kant’s distinguished

assertions in the “Transcendental aesthetic” and go over them point by

155

Cf. KrV, intro. VII, B29-30. In fact, Kant claims in B29 that sensibility and the

understanding are the two stems of human cognition, which may perhaps arise from a

common root that is unknown to us.156

Přihonský’s statement is based on Bolzano’s inferential account of perceptual

knowledge. Cf. for example, WLI, § 77, where Bolzano says that we can infer the

existence of an object from a corresponding subjective intuition or sensation. See my note

113 on Bolzano’s weak causal claim. Cf. also Přihonský’s strong causal claim, NAK, p.

48, p. 233, and my note 608.

New Anti-Kant 103

point and present our objections. So this is what we will do now,informing our readers that we will apply this method of presentation inour entire investigation.

First of all, Kant attempts1. to determine the concept intuition and the distinction betweenempirical and pure intuitions. In addition, he attempts to explain what hemeans by the expression form of sensibility. However, Kant’s words seemso strange to us that we will let our readers hear what he says. [46] Thisis how he begins his presentation:

“In whatever way and by whatever means a cognition may refer to objects,

intuition is that by which a cognition refers to objects immediately and to which

all thought is directed as a means. Intuition, however, takes place only in so far as

the object is given to us; but this, in turn, is possible only, at least to us humans, if

the object affects our minds in a certain way. The capacity (receptivity) to acquire

representations through the way in which we are affected by objects is called

sensibility. So objects are given to us by means of sensibility, and it alone

supplies us with intuitions; but they are thought through the understanding, and

from it arise concepts.157 But all thinking must ultimately refer to intuitions by

means of certain marks, whether straight off (directly) or circuitously (indirectly);thus, in our case, it must refer ultimately to sensibility, because there is no other

way in which objects can be given to us.

The effect of an object on our capacity for representation, insofar as we are

affected by it, is sensation. An intuition that refers to the object through sensation

is called empirical. The undetermined object of an empirical intuition is called

appearance. I call that which in an appearance corresponds to sensation its

matter, but that which [47] allows the manifold of appearance to be intuited asordered in certain relations I call the form of appearance.158 Since that in which

sensations alone can be ordered and placed into a certain form cannot itself be

sensation, and therefore, although the matter of all appearance is given to us aposteriori, its form must lie altogether ready for it in the mind a priori and can be

considered separately from all sensation.

I call all representations pure (in the transcendental sense) in which nothing is

found that belongs to sensation. Accordingly, the pure form of sensible intuitions

in general is found in the mind a priori, wherein all the manifold of appearances

is intuited in certain relations. This pure form of sensibility itself is also called

157

The words gegeben, gedacht, and Materie are emphasised by Kant (cf. KrV, B33) but

not by Přihonský. In addition, I follow Kant in dividing the passage (B33-B35) into

paragraphs, unlike Přihonský’s quote. In this passage, Kant distinguishes between the

matter and the form of appearance: the former corresponds to sensation, and the latter is

an a priori form of intuition which orders sensation. Dubislav (1931, op. cit., p. 210)

mentions two other Kantian uses of the word “appearance”: (1) objects of experience

determined by synthetic a priori principles (i.e., phaenomena, cf. KrV, A249) and (2)

empirical intuitions.158

I follow P. Guyer and A. Wood’s translation of the Critique B34 by adding “intuited as”

which, strictly speaking, is not in the text but helps in presenting Kant’s claim more

accurately.


pure intuition. So if I separate from the representation of a body that which the

understanding thinks in it, such as substance, force, divisibility, and so on, as well

as that which belongs to sensation, such as impenetrability, hardness, colour, and

so on, something still remains for me from this empirical intuition, namely,

extension and form. These belong to pure intuition, which occurs in the mind apriori, as a mere form of sensibility, even without an actual object of the senses or

of sensation.”

Then Kant declares that he intends to deal with these forms of [48]

sensibility in more detail and recalls in advance that it will be found “thatthere are two pure forms of sensibility which are principles for a priori

cognition, space and time.”159

Objections. In the passage quoted above, our philosopher putsforward a distinction between representations he had almost failed tonotice before, by means of which we call them either intuitions orconcepts. But although the distinction between these two kinds ofrepresentations or designations has become a prevailing custom afterKant, at least in Germany, there is still much doubt about whether theconcepts we relate to these expressions have always remained the same.So it is just not that easy to determine what Kant intended by theseexpressions, because he failed once more to give a clear definition. Whatwe can infer with the highest degree of certainty from the quotedpassage, as well as from many others dealing with this subject, is that byintuitions, Kant has always meant only those representations referring toa single and real object. This is why he would have defined arepresentation with several objects, for example: man, triangle, and soon, as a general concept, but not as an intuition. It is equally certain thatby an intuition [49] he always intended a representation whose existenceimmediately entitled us to infer from it the existence of a correspondingobject (represented by it). For, obviously, this is precisely what he wantsto say, when using the locution that an object is given in intuition,immediately given, and even that there is no other way in which an objectcan be given to us (at least immediately).160 But when we examine the

159

Přihonský’s quote of the passage in KrV, B36, is incomplete. Kant’s text reads: “that

there are two pure forms of sensible intuition as principles of a priori cognition, space and

time”. In addition, Přihonský misquotes Kant when he refers to “forms of sensibility”,

rather than to “two pure forms of sensible intuition” (ibid.).160

First, Kant would object to Přihonský that we are not entitled to infer the existence of

an object from the mere existence of a representation. Instead, the object is a construction

or “that in the concept of which the manifold of intuitions is united” (KrV, B137). Second,

for Kant intuitions are single but not simple since they refer to a manifold of sensations

unified by a concept. An object is given in an intuition if and only if the intuition is

combined with a concept (cf. my note 162, below). So, strictly speaking, that which is

immediately given in an intuition is not (yet) an object, but “all our intuition is nothing

New Anti-Kant 105

circumstances under which we are entitled to infer the existence of anobject from the mere presence of a representation, we can see at once thatthis is only the case if the representation is absolutely simple. For whenyou compose a complex representation out of several others through theefficiency of your mind, you must not immediately presuppose that italso has a corresponding object, because such a representation mayfrequently be objectless, as the representations round, square, and so on.In view of this consideration, we agree with Bolzano’s definition of theconcept intuition that it is a representation which, despite its simplicity,

only has a single object.161 The following benefits result from this: first,

but the representation of appearance” and “the things we intuit are not in themselves what

we intuit them to be” (B59). This claim is problematic but not for the reasons given by

Přihonský. Instead, the problem is the twofold nature of intuition as a condition or form of

appearances (i.e., which precedes them) and as something conditioned or affected by the

manifold of sensation (i.e., which results from, or comes after appearances). Third, this is

Přihonský’s own causal claim (which he falsely ascribes to Kant), namely, that intuitions

can provide reliable information about the external world and that we are therefore

entitled to infer the existence of an object from the existence of its representation, cf. my

notes 113-114, Prihonksy’s argument on p. 232, and my note 607, where I mention

objections to this strong causal view.161

Bolzano’s point against Kant is that intuitions are simple and have only one object,

which is the change in our mind caused by an object presented to our senses; cf. notes 17

and 18. Bolzano also objects against the Port Royal rule (cf. note 10) that in intuitions, the

content and the intension of a representation do not stand in an inverse relation, for if an

intuition is both simple and singular, its extension and intension are numerically equal. In

addition, (a) we cannot increase the content of an intuition, for all we obtain is a

redundant representation since intuitions are simple; cf. WLI, § 120. They cannot be

decomposed into parts, because they directly present a mental change caused by a real

object, this red. Intuitions are expressed by demonstratives, such as “this red”, or “this A”.

When we increase the content of the intuition, we do not decrease its extension, for we

obtain “this, which I now see, is red” or “this, which I now see, is A”. But “this, which I

now see, is red” means “this red”, so the so-called complexity of these representations is a

mere redundancy and does not increase their extension; cf. Bolzano’s Reply to Exner, 14

Jan. 1834, in the Bolzano-Exner Correspondence, p. 40. (b) Nor can we increase the

extension of an intuition and thus decrease the content, because intuitions only have one

object; cf. WLI, § 72. Nonetheless, Bolzano’s claim that intuitions are simple is

problematic, for how do we account for “the manifold of sensations”, as Kant would say?

(1) Bolzano could argue that he only postulates the simplicity of intuitions without

proving it. He claims with Locke that our intuition [this red] is a simple idea (cf. Essay,

bk. II, ch. II, § 1; II.XVIII.1-6), yet the object it refers to, though single, is complex. For

example, if we look at a white wall, we do not see a simple or uniform white but a more or

less saturated white, composed of different hues and different degrees of brightness. But if

the object is complex, how do we account for the simplicity of our corresponding

representation of it? (2) Přihonský, who holds a strong causal view, could argue that the

object/intuition relation is isomorphic and that although the intuition, as a representation

of an effect (a mental change) is simple, its object or that which causes the intuition, is


we are even more entitled than Kant is to [50] claim that through eachintuition an object is given, immediately given, and our claim is incomplete agreement with his. But we obtain an even greater benefit byhaving a definition of intuitions by means of which these representationscan be distinguished objectively from all others. This makes it possiblealso to justify a completely objective distinction between a priori and aposteriori propositions, which concerns their basic nature and not theirmere relation to our cognitive capacity. – If Bolzano’s definition ofintuition and the definition of concepts derived from it (see theintroduction, p. 9) are right and if we can assert that they are, while notexactly, still nearly the same as what Kant may have intended by them,then it is indeed peculiar that the latter’s definitions are so extremelyimperfect. He does not even explicitly indicate the mark that an intuitioncan have only a single object, that is, that it must be a singularrepresentation. In addition, not the slightest trace can be found of theother marks indicating that a pure intuition (not mixed with any concepts)must be a simple representation, either in his works or in those of laterphilosophers who accepted and maintained the Kantian distinctionbetween intuitions and concepts. Instead, we find in what follows thatKant often mentions the manifold which lies in intuition.162 In our view,

complex. A mental state is singular but not simple, though it can be the object of a simple

representation. Bolzano claims that the relation between concepts and their objects is

isomorphic (cf. “Aufsatz, worin eine von Herrn Exner in seiner Abhandlung ‘UeberNominalismus und Realismus’ angeregte logische Frage beantwortet wird” (1843), BBGA

I, vol. 18, pp. 63-85). As to intuitions, he would say that we can infer (with probability)

that our intuitions are caused by an object, but he does not say that they are isomorphic.

Cf. WLIII, 303.22; cf. also my notes 113, 114, and 607, on Bolzano and Přihonský’s

causal claim.162

Přihonský’s presentation of Kant’s notion of intuitions and his distinction between

intuitions is a) incomplete and b) not quite correct: (1) in Kant’s view there are three

kinds of representations: general representations which are either concepts of the

understanding (repraesentatio per notas communes) or concepts of reason (repraesentatiodiscursiva) and singular representations or intuitions (repraesentatio singularis); cf. JL, §

1. (2) Although both Kant and Bolzano define an intuition as having a single object,

because it has only one object which is immediately given, they have different reasons for

doing so: Kant bases his notion of singular representations on his psychological

distinction between the genetic origins of intuitions and concepts. These two kinds of

representations are defined as two distinct types of cognitions which originate in two

separate faculties of knowledge: passive sensibility and active understanding,

respectively; cf. JL, intro. V. (In fact, the distinction between percepts and concepts goes

back to Leibniz, who draws this division against the empiricist model of deriving

nonsensory ideas from sensory impressions, and Kant follows Leibniz on this point.) (3)

From this assumption of their different genetic origins, Kant infers that objects are

immediately given in intuition (intuitions are blind without concepts) and thought by the

New Anti-Kant 107

however, a manifold can only lie in a singular [51] representation if thelatter was produced by combining several simple representations. Such acombination can only occur through the mediation of several suchrepresentations, which are not intuitions but concepts. For mereintuitions: ‘this red’, ‘this fragrance’, ‘this sweet’, and so on, do not yetform a singular representation, even if we have them simultaneously, butcan only be unified into one by a concept, as the one indicated by theword: and, or some such other. – We should digress excessively from ouraim if we had to explain what pernicious consequences resulted from thisdeficient determination of the concept of an intuition and how subsequentscholars were misled by this deficiency to introduce a special intellectual

intuitive faculty, to the detriment of science. Thus it should suffice toclarify only those erroneous assertions for which Kant is liable in theshort passage we have shown our readers.

Here Kant asserts repeatedly that an object can be given only throughintuitions. But we have already shown how wrong this is, if it wassupposed to mean that we can be certain that there is an objectcorresponding to a concept, if and only if we have [52] some intuitionthat is subordinate to that concept. Every mathematician rightlypresupposes that the concept of a straight line which extends to infinity

understanding (thoughts without content are empty); cf. KrV, B75. Consequently,

intuitions have objects (phaenomena) if and only if they are combined with concepts, but

since objects can be immediately given in intuition alone, concepts only have a mediate

relation to objects, through intuition; cf. KrV, B33. This is one of the main differences

between Kant and Bolzano. In Kant’s view, seeing involves concepts, since he defines

intuitions as cognitions, whereas Bolzano holds that intuitions do not need to fall under a

concept or to be cognitions (i.e., clear representations) in order to have an object; cf. WLI,

§ 72; WLIII, § 300; MM4. For Bolzano, “a representation is clear if we represent it to

ourselves by way of an intuition.” (WLIII, § 280.4) For Kant, if our intuition is clear, this

is so only because it falls under a concept, whereas Bolzano claims that an intuition

becomes clear when we direct our attention towards it. But attention is a perceptual act

(cf. Bolzano’s discussion of memory-representations WLIII, § 284.2). In addition, the

object of an intuition is not the same for Kant and for Bolzano: roughly, for Kant, an

object is a construction or that in which the manifold of a given intuition is united

(Gegenstand). For Bolzano, an object of intuition is a) a mental state and b) that which

can be perceived and judged or that which is immediately given and about which we can

make a direct or immediate judgment (cf. my notes 106, 114, 165, 250, 251, 307, 329,

612). (4) Another crucial difference between Kant’s and Bolzano’s accounts of intuition,

and one which Přihonský overlooks here and considers as a mistake further on (NAK, p.

163), is that unlike Bolzano, Kant claims that intuitions are complex, because what is

given in a particular intuition is the manifold of sensations which is unified by an act of

synthesis carried out by the mind; cf. KrV, A99. Kant’s intuitions correspond to Bolzano’s

mixed representations, which have both intuitive and conceptual components; cf. WLI, §

73.


on one or both sides is an objectual concept; that is, that there are certainobjects corresponding to it. Yet nobody can have an intuition subordinateto this concept, for surely an infinite line can never be intuited, even inthe largest and most haphazard sense we may attribute to the word“intuition”.163 – Kant’s claim, that the pure form of sensibility (i.e., thatwhich makes it possible for the manifold to be ordered in certainrelations in the appearance) can also be called a pure intuition, showshow haphazard the concept he related to this word must have been.164

According to this, a pure intuition is essentially complex, containing a

163

Kant would agree with Přihonský that the concept of an infinite line is not given in

intuition, for in Kant’s view, mathematical concepts are constructed in intuition by means

of a procedure called the schematism of the pure concepts of the understanding; cf. KrV,

B176. The construction or exhibition of a mathematical concept in intuition occurs by

means of a schema, which not only provides the concept with an image (and thus presents

it) but defines the properties of this concept according to a rule. Unlike Bolzano and

Přihonský, Kant holds that concepts are inadequate for representing mathematical notions

such as infinity and infinite divisibility, and this is why he recurs to a constructive

procedure or introduction of limitations, involving a “cutting out” of smaller line

segments from any given line. So the idea of infinity and of all general concepts of space

is made possible by limitation or intuitive construction, and thus Kant claims that

geometric cognitions are derived from intuition; cf. P, § 12; KrV, B39-40 (quoted by

Přihonský, p. 41), B203-204. Cf. Michael Friedman, Kant and the Exact Sciences (1992),

pp. 69-71. Nevertheless, in Kant’s view the understanding also plays a role in geometric

cognitions, for all geometric propositions must be subsumed under the principle of the

axioms of intuitions to prove their validity. (A subsumption is conceptual, because it is an

operation of the understanding.) This principle says that all intuitions are extensive

magnitudes (cf. KrV, B202), so, for example, the Euclidean axiom “that between two

points only one straight line is possible” is verified first by exhibiting it in intuition, by

means of a schema, but its validity is justified if and only if we subsume our construction

of a straight line under the concept of extensive magnitude, that is, if we grasp the straight

line as an extensive magnitude. For a detailed exposition of this problem and a defense of

an (at least partly) conceptual account of Kant’s geometric propositions, cf. Michael

Wolff (2001), “Geometrie und Erfahrung. Kant und das Problem objektiver Geltung derEuklidischen Geometrie”, pp. 209-232. Cf. also Dubislav (1931, op. cit. pp. 210-211),

who comments that Přihonský’s assertion is understandable only if it is considered in view

of Bolzano’s quasi-platonistic doctrine of propositions and representations as such. For an

outline of the opposition between Kant’s and Bolzano’s accounts of mathematics which

still dominates contemporary mathematics, cf. J. Sebestik: “La dispute de Bolzano avecKant. (Le fragment d’un dialogue sur la connaissance mathématique” (forthcoming).164

Kant’s claim calls for a double role for the pure forms of intuitions, which are the pure

intuitions of space and time. These latter are (a) the conditions and principles of a prioricognition (cf. KrV, B36) and (b) magnitudes “which contain a manifold (of their own)”,

KrV, B160. The forms of intuitions are space and time considered as representations, as

opposed to formal intuitions, which are space and time considered as the objects of

intuitions. The former concern space as a structure (such as Euclidean space, for

example), and the latter concern the relations between different (or the manifolds of)

spatial structures.

New Anti-Kant 109

manifold which it combines into a certain unity. Yet it is certain, as Kanthimself acknowledges (e.g., p. 136) elsewhere, that combination is an actof the understanding, that is, it is only achieved through concepts.165 Inthis way the intuition is a complex representation containing concepts asparts.166 Now then, is it useful to take the concept of an intuition in such alarge and vague sense [53], which must necessarily give rise to confusionin science? – If a pure intuition is (as we must presuppose) reallysupposed to be an intuition, then Kant should at least have proved firstthat the form of sensibility is a real intuition, instead of directly affirming

it, since this does not follow from his own definition that intuitions arethose representations through which an object is immediately given. For

165

Přihonský synopsizes Kant on this point: the synthesis through which an intuition’s

unity is achieved is not an operation of the understanding but a synthesis of apprehension,

which has an intuitive nature and takes place as our experience is ordered for the first time

in the a priori forms of cognition that are time and space; cf. KrV, A99. In KrV, B133-

134, Kant gives the (admittedly twisted) explanation that the unity of an intuition or the

unity of apperception, as he calls it there, is analytical and presupposes a synthetic unity.

This synthetic unity of apperception is transcendental and originates in our self-

awareness, and it produces the representation “I think”, which allows intuitions to become

objects of knowledge. In addition, this transcendental apperception is the condition for the

(second) synthesis operated on intuitions by the understanding, where intuitions can

belong to a subject and can be combined with a concept in a judgment. So it seems that

the synthesis of the manifold is given to the understanding in intuition (§ 21, B144-145),

based on the unity of apperception, but then Kant adds (B145-146) that the unity of

apperception is itself produced by the understanding “by means of the categories” (B145),

in order to restore all mental activity to the understanding. But, in addition, Kant puts

forward a much more interesting claim concerning the sensory manifold in the Critique ofthe Faculty of Judging (KU): “colour and sound are not mere sensations but formal

determinations of the unity of a sensory manifold”, KU, pt. 1, bk. 1, § 14. Here Kant

claims, long before the Gestalt psychologists, that colour sensations have a formal

structure or pattern. We apprehend this structure not through intuition but through thought

or reflection, and not because our manifold of sensations is the perceptual matter given in

an intuition which is subsumed under a concept, but because our auditory and visual

sensations “permit not merely a feeling of the senses but also of reflection upon the form

of these modifications of sense”, KU, pt. 1, bk. 1, § 42. Cf. Anita Kasabova (2004),

“Colour Sensations and Colour Qualities: Bolzano between Modern and Contemporary

Views”: 247-276.166

It seems that Přihonský assimilates Kant’s intuitions to Bolzano’s mixed

representations; cf. my note 162. However, strictly speaking, Kant’s intuitions do not

contain concepts as parts. Rather, intuitions are cognitions (or have an object) if and only

if they are combined with a concept. Still, Kant would agree that the sensory manifold is

united by an activity, which can ultimately be traced back to the understanding, if and

only if this latter is taken in the sense of a faculty for unifying representations (that is,

both intuitions and concepts, since judgments are “functions of unity among our

representations” (cf. KrV, B93). But the sensory manifold itself is not structured; rather it

is a mass of undifferentiated points.


the representations of the form of an object (of the relations in which thelatter’s manifold can be ordered) by no means belong to therepresentations through which an object is given, least of all immediately

given, to us, and this follows even less from the only definition that weconsider as right. – But perhaps the example Kant considered as a helpfuladded explanation will give us more information? According to thislatter, the (representation of the) “extension and shape”167 of a body weintuit168 is a pure intuition, because this representation remains once weseparate from it everything which the understanding thinks about it, suchas substance, force, divisibility, and soon, as well as everythingbelonging to sensation (matter), such as impenetrability, hardness,colour, etc. Quite frankly, this example convinces us even more that Kanthas related the word intuition to a concept that is so indeterminate andwide [54] that no true advantage can be gained from his division ofrepresentations into intuitions and concepts. Impenetrability, hardness,extension, and shape are supposed to be the properties or determinationsof a body which are opposed to substance, force, divisibility, and so on,in such a way that the latter can be cognized by the understanding (thefaculty of concepts), whereas the former can be cognized by sensibility(the faculty of intuitions).169 We can acknowledge this opposition onlyinsofar as it means that we can make the judgments: each body consistsof substances, has certain forces, is divisible, and so on, without havingany intuitions of such a body. On the contrary, we can only make thejudgments that this body is impenetrable, hard, has a certain figure (say,it is spherical) once we have had certain intuitions referring to thatobject. But we absolutely can not accept Kant’s apparent belief that therepresentations: impenetrability, hardness, spherical shape, and similarones are either intuitions themselves or even mixed representations(containing an intuition as a component). They are concepts, even pureconcepts, in exactly the same way as substance, force, divisibility, and soon. The only difference is that substance is a property applicable to allbodies, whereas spherical shape is a property applicable to some bodies

167

Cf. KrV, B35.168

I have used “intuit” for translating angeschaut, so as not to deviate from the

terminology, since Přihonský’s point concerns intuition, although “perceive” would be

more familiar to the contemporary reader.169

Přihonský dissolves Kant’s distinction between the pure (forms of) intuitions and

empirical intuitions, and as a result he puts impenetrability and hardness on the same level

as extension and shape. But for Kant, spatial extension and shape are essential features

which define a body, and this is why he says that these properties remain, even once we

have abstracted all other properties from its representation. For precisely this reason he

considers the judgment “all bodies are extended” as analytic.

New Anti-Kant 111

only, and not to all bodies. However, we can find out only through ourown intuitions whether the latter property applies to the body presently athand. [55] No such distinction governs the way in which this is found outor the way we learn whether a body at hand belongs to the hard ones andthe way we learn whether it belongs to the round ones. Hence we cannotjustifiably name one property an empirical representation and the other apure representation (especially an intuition). And yet if we had told himour objections, Kant might have replied that he did not take the generalrepresentation spherical shape as an intuition but only the representationsthis particular spherical shape, this particular place, this particular

distance, and so on. It will soon become clear that, and why, we cannotaccept this either. For if there is any Kantian doctrine we can accuse ofconfusion, it is the doctrine of spatial objects and the doctrine of spaceand time in general, which is the only content of the transcendental

aesthetic.So in order to bring a little light into this darkness, we consider it

necessary to inform our readers which concepts of time and space we, atleast, consider as right. These are Bolzano’s concepts, as our readers mayhave guessed, and if they disagree with us, they are still free to acceptKant’s theory, or any other. [56] There is no doubt that time and space

are a pair of objects so closely related that it is not surprising if Kantpresents them in a combined doctrine, especially since their connectionwas frequently noticed before his time. But if it is a question of inquiringinto their true nature, it seems to us more useful to begin thisinvestigation with time rather than space, since everything which is inspace must consequently already be in time, but not vice versa (forexample, thoughts, sensations, etc., are temporal but certainly do nothave spatial extension).170 So we may assume that the nature of space

170

Yet Bolzano claims in the Paradoxes of the Infinite that mental entities are

spatiotemporal, so in his view, thoughts, sensations, and other subjective representations

do have spatial extension, and consequently there is only a difference of degree between

mental and material substances. Both are actual and composed of simpler substances

(atoms), and both are located at a single point in space. Bolzano uses this claim to refute

the dualist assumption that the intelligible and the sensible domain are completely

separate and that, consequently, mental and physical substances are completely separate

entities. He calls this assumption the prejudice of the nonspatiality of mental entities,

which has haunted German philosophy ever since Descartes established it and Kant

developed it. The paradox resulting from this prejudice is this: how can mental and

physical substances interact if they are completely distinct? Bolzano solves this problem

by declaring mental entities as spatiotemporal; cf. PU, §§ 55-56. This claim is consistent

with his postulate of a reciprocal interaction between the soul and the body and his view

that the soul can be localized at a single point in space because it is a simple substance

and because its changes must have a cause in other spatiotemporal substances (Athanasia,


cannot be clearly cognized without having first investigated the nature oftime.171 – It is known that we put everything real into a certain time,except the highest being, so that we can in truth ascribe an attribute to areally existing object only by adding a certain temporal determination.172

This assertion is universally valid, so that we may even assert concerningdivine properties that they fit God at a certain time and at all times.Every proposition whose subject-representation refers to a real thing andtherefore has the form “the real A has (the property) b” is true onlyinsofar as we may relate some temporal moment with its subject-representation.173 [57] Propositions such as: “I hear the clock strike two,and feel joy”, “the sun is shining”, and so on, are not completely true iftheir subject-representations are lacking a temporal determination and ifthey have any sense other than this: “At this moment in time, I hear theclock strike two, and feel joy,” “during daytime, the sun is shining”, andso on.” It follows from this that we are supposed to take the word time asthe determination of something real, containing the condition on which

we can in truth ascribe certain properties to the latter. A point in favour

sec. 4, pp. 124ff.; cf. Berg (1997), “Bolzano, the Prescient Encyclopedist”, pp. 13-32).

But Bolzano’s claim that mental entities are spatiotemporal leads to another problem in

his theory, since he claims elsewhere that there are also unextended qualities, such as

internal sensations (cf. note 188, below).171

Kant begins his investigation with space because he first wants to examine external

intuitions and the (spatial) relations between phaenomena before inquiring into inner

sense. In addition, he insists that time is not only the form of inner sense but the basic

principle of all intuitions: “Time is the formal a priori condition of all appearances in

general”, whereas space “is limited as an a priori condition merely to outer intuitions.”

(KrV, B50). Besides, Kant explains in the schematism (which is basically a technique for

concept-application) that the schema (or rule governing the use of a concept) of the

concept of the understanding is the temporal determination of this concept, because time

is the formal condition of the connection of all representations; cf. KrV, B177-179. In

other words, temporal determination is the condition sine qua non of all our

representations. Bolzano, on the other hand, begins his investigation with time and argues

that time and space are neither pure intuitions nor forms of pure intuitions but

determinations of real objects; cf. WLI, § 79; PU, § 55; and Versuch einer objectivenBegründung der Lehre von den drei Dimensionen des Raumes (1815) (VOB), pp. 223-238.

Scholz (1931, op. cit., pp. 211-217) quotes this text in his notes; cf. also my note 181.172

Přihonský takes up Bolzano’s claim that time is a determination of objects and attaches

to the subject-term of statements instead of the predicables; cf. WLI, §§ 25, 75.2, 79.5;

WLII, § 243; WLIII, § 286.8. Cf. Mark Textor on Bolzano’s rejection of the Aristotelian

thesis that tense attaches to the predicate and not to names, in “Caius at Noon or HowBolzano Misunderstood the Copula”, forthcoming.173

Cf. WLI, § 79.5, where Bolzano claims that all propositions in which a contingent

attribute b is ascribed to a real object A have the form [A in t has b] and such propositions

do not express a complete truth unless we specify when A has b. Přihonský repeats this

passage almost verbatim.

New Anti-Kant 113

of this concept is the fact that all temporal properties can be derived fromit without much difficulty. Let us test this by means of the followingprinciple: several contradictory properties can be ascribed to the samesubstance only on condition that there is a difference in time.174 Twocontradictory properties (blossoming and fading) cannot be correctlypredicated of one and the same substance175 or a collection of substances(e.g., a tree) at the same time. For otherwise two propositions withcontradictory predicates of the same subject-representations would bothhave to be true. – As far as the concept of space is concerned, surelyeverybody would accept [58] that it is the collection of all locations.176

But this gives rise to the question: what do we mean by the concept oflocations which are filled by real things? It is unarguably true thateverything real acts and, insofar as it is finite, sometimes undergoescertain changes, sometimes causes them in the objects surrounding it.Obviously the nature of these changes only depends, first, on the forces

of the thing as well as of its surroundings and, second, on their locations.It follows that locations are those determinations of really existing things

we must presuppose, in addition to their forces, so as to explain all the

changes they cause in one another.177 This proposition seems to expressthe right concept of space, precisely because all spatial properties taughtin geometry can be deduced from it.

Anyone who thinks that these definitions of time and space are rightwill easily agree with our claim that the representations of time andspace are pure concepts, since there is no doubt that the components bymeans of which we just composed these representations are altogetherpure concepts.178 But we must still ask whether there are not somerepresentations among the remaining representations we call temporal orspatial [59] that are intuitions or at least contain intuitions ascomponents? Surely nobody will doubt that those representationsreferring to time and space with more than one object, for example, the

174

Unlike Bolzano, Přihonský does not mention the ontological rule that an object at a

different time is really a different object; cf. WLI, § 45.1.a. Cf. also Dubislav (1931, p.

217), who notes that strictly speaking, nothing can be deduced from a concept but only

from the propositions characterizing that concept.175

Following Bolzano, I have added the adverb correctly which Přihonský omits from an

otherwise faithful reproduction of WLI, § 79.5. For the point is, precisely, that we cannot

correctly predicate two contradictory properties of the same thing at the same time.176

Actually, Bolzano (whose text Přihonský follows) says that space is the collection of all

possible locations; cf. WLI, § 79.6.177

This is Bolzano’s definition of location; cf. WLI, § 79.5.178

Cf. WLI, § 79.1-4, where Bolzano argues that space and time are pure concepts. In what

follows, Přihonský faithfully summarises Bolzano’s points. Cf. also PU, § 17.


representation of a moment in general, a time span in general, a point,distance, line, plane, and body in general, are not pure intuitions; for pureintuitions have only a single object, whereas these representations haveinfinitely many objects. A doubt could only arise concerning suchconcepts as all of infinite time, all of infinite space, this definite moment(e.g., the moment in which I am at present), this definite period (e.g., anhour), this definite point (e.g., the centre of the earth), this definitedistance (e.g., between the earth and moon), and so on. Each of theserepresentations has only a single object and if they were also simple, thenby our own definition we would have to count them as pure intuitions.Meanwhile we must recall that every (subjective) intuition arising in oursoul has an actually existing object, and we can infer from this that therepresentations we named are not intuitions. Incidentally, we are free todecide on the question whether they are simple or complex, since theirobject [60] does not exist. To realize that this is self-evident, it is enoughthat we possess these representations and that we understand the meaningof the above-mentioned expressions. Someone who just knows what kindof concepts mathematicians relate to the words time and space mustadmit without hesitation that only the things in time and space are real,but not times and spaces themselves. If he wanted to define times andspaces as realities, he would have to claim that they were effective orcaused something. But which products would these be? – Although weare in the habit of saying that time causes a good many things, such asmaking roses bloom, healing pain and wounds, and so on, still is thereanybody who cannot see that this is said only figuratively to suggest thatthe changes occurring in the course of time have causes which we learnof sooner or later? So, of course, it is not time which causes all effectsand changes in certain periods, but the force of things. This is also howwe should take the locution that cramped space causes us to feel uneasy.It is not cramped space which does this but the narrow walls, a lack offresh air, and so on. – If space and time were something real, theirexistence would be either conditioned or unconditioned.179 If it wereunconditioned, they would be God himself (since only God’s existence isunconditioned and real). If it were conditioned, they would be [61]

creatures which are subject to change. The only things that change arethings in time and space, but not time and space themselves, and only thescholastics could seriously argue about whether time and space arecreated or not. – If time and space were something real, then twomoments or two time spans, two points or two distances would not be

179

Přihonský refers to spatiotemporal objects which are real in the sense that they actuallyexist.

New Anti-Kant 115

internally equal, since no two real things are completely similar.However, all mathematicians agree that all moments and all points arecompletely similar and not only two moments or two points. – Finally, iftime and space were real, there would be something real which lacked asufficient grounding, namely, the existence of a thing at this determinatetime, at this determinate location. For there is no way we can indicatewhy a thing should be at this time rather than at another and at thislocation rather than at another, because internally, these locations andtimes are completely similar. – So if the representations this moment, thistime span, this point, this distance, and so on, are not representations of areally existing object, then they cannot be intuitions either. If no momentin time and no point in space really exist, then, surely, neither do thecollection of all moments and the collection of all points, [62] or thewhole of time and the whole of space. Consequently, we cannot defineeither one of those two representations as an intuition, even though theyhave only a single object (since there is only one infinite time and onlyone infinite space), and we must therefore consider them as pureconcepts. For nobody would believe that they contained an intuition as acomponent. But if we believe we are constrained to consider therepresentation of the whole of infinite time and the representation of thewhole of infinite space as pure concepts, then we must also accept thatthe representations moment, time span, point, distance, and so on, ingeneral belong to the class of pure concepts. For if the representations thewhole of infinite time and the whole of infinite space do not containanything intuitive, how could we call certain parts of theirrepresentations “intuitions”? (Cf. Bolzano’s Theory of Science, § 79.)

Let us now hear how Kant2. presents and proves the doctrine of time and space. The most

important passage on this issue is on pp. 63f. in the “Transcendentalaesthetic”.180 There we read:

“Space is not a discursive or, as we say, general concept of relations of things as

such, but a pure intuition.*181 For, [63] first, we can only represent to ourselves a

180

Cf. KrV, B39-40. On Kant’s doctrine of time and space, cf. Scholz (1924), “DasVermächtnis der kantischen Lehre vom Raum und von der Zeit”, pp. 21-69, and Dubislav

(1931, pp. 217-218). For a pro-Kantian view, cf. Friedman (1992), Kant and the ExactSciences.

*181

Kant should have been discussing the representations of time and space instead of the

objects of these representations (of time and space themselves).

Although Přihonský points this out against Kant, he himself does not apply the distinction

between time and space as objects and the representations of time and space. However,

this issue is problematic in Bolzano’s texts, insofar as we are faced with the following

dilemma: either time and space are determinations and representations, or the


single space,182 and if we speak of many spaces, we mean by that only parts of one

and the same unique space. And these parts cannot, as it were, precede the single

all-encompassing space as its components (from which its composition would be

possible) but can only be thought in it. It is essentially single. The manifold in it,

hence also the general concept of spaces as such, rests merely on limitations. It

follows from this that, regarding space, an a priori intuition of it (which is not

empirical) grounds all concepts of space. Thus also all geometrical principles, for

example, that in a triangle two sides together are greater than the third, are never

derived from general concepts of line and triangle but are derived from intuition

and this a priori, with apodictic certainty. [. . .]

Space is represented as an infinite given magnitude. Now we must think every

concept as a representation contained in an infinite collection of different possible

representations (as their common mark), which hence contains these

representations under itself. But no concept as such can be thought as containing

an infinite collection of representations within itself. Yet that is how space is

thought (for all parts of space, ad infinitum, are simultaneous). Therefore the

original representation of space is an [64] a priori intuition, not a concept.”183

Kant defines time in a similar way:184 “time is not a discursive or, as wesay, general concept but a [pure] form of sensible intuition. Differenttimes are only parts of the same time. Intuition, however, is therepresentation which can only be given through a single object.185 Inaddition, the proposition that different times cannot be simultaneouscould not be derived from a general concept. The proposition is synthetic

representations of time are determinations, but not time and space themselves. In WLI, §

79, and Versuch einer objectiven Begründung (1815) (VOB), § 2.4, Bolzano considers

time and space as determinations of something actual. A determination is something that

is represented, but not a representation. On this I follow Morscher (1973), p. 71, rather

than Sebestik (1992), p. 148, on whose reading of Bolzano determinations arerepresentations because to represent something means to indicate its properties, and the

notions of determination and property are closely related, since each property of an object

can be transformed into a determination of the same object. However, in VOB, § 2.2.

Bolzano says that time is not a property, and Morscher argues that property concepts,

insofar as they are predicate-concepts in a true proposition, constitute a determination of

the subject of this proposition, so that it is the property-concepts which are the

determinations but not the properties themselves. However, it follows from this that the

determinations that are not property-concepts are partial representations belonging to the

subject-concept of a proposition.182

Sich vorstellen translates “represent to ourselves”, which lacks elegance if not

precision, especially when considering Kant’s insistence on self-awareness. For this

reason I follow Kemp Smith, though I add brackets, instead of Guyer’s and Pluhar’s

translations which omit the reflexive pronoun.183

The emphases in this passage are Kant’s, omitted by Přihonský, and so are the divisions

into paragraphs.184

Cf. KrV, B47-48.185

In KrV, B47, Kant writes Anschauung, whereas Přihonský adds “eine” Anschauung.

New Anti-Kant 117

and cannot arise from concepts alone. Hence it is immediately containedin the intuition and representation of time. [. . .]

“The infinitude of time signifies nothing more than that everydeterminate magnitude of time is only possible through limitations of asingle time grounding it. The original representation time must thereforebe given as unlimited. But where the parts themselves and everymagnitude of an object can be represented determinately only throughlimitation, there the whole representation cannot be given throughconcepts (for they contain only partial representations), but they must begrounded by immediate intuition.” –

Objections. Thus Kant does not define each representation of spaceand of time as a pure intuition, but the whole of infinite space and thewhole of infinite time. His main reason is this: “because there is only onesingle space and only [65] one single time; because, when one speaks ofseveral spaces and times, these are only parts of one single space and onesingle time; and because a representation which is given through a singleobject is an intuition.”186 – However, the representations God, universe,supreme moral law, the first prime number in the sequence of naturalnumbers, and thousands of other representations also have only a singleobject.187 But who would call them intuitions because of that? But eventhat which Kant gives as proof of the singularity of space, “namely, thatthese parts can only be thought in it as limitations and they cannotprecede space as components from which its composition would bepossible.”,188 is not very appropriate for convincing us of his claim. Ifthis assertion were correct, we should have to represent the whole ofspace in order to obtain the representation of a single point.189 But who

186

This quote more or less corresponds to KrV, B39, but in this passage Kant refers only to

space, and he says: “one can only represent a single space, and if one speaks of many

spaces, one understands by that only parts of one and the same unique space.” In other

words, he does not claim that there is only a single space or a single time but that we can

only represent a single space or a single time.187

Cf. WLI, § 79.6, as well as the subsequent objections.188

Although he uses quotation marks, Přihonský paraphrases Kant. The passage in KrV,

B40, reads:

“And these parts cannot, as it were, precede the single all-encompassing space as

its components (from which its composition would be possible), but can only be

thought in it. It is essentially single. The manifold in it, hence also the general

concept of spaces as such, rests merely on limitations.”

189 Cf. Bolzano’s objection against Kant: if finite or limited spaces cannot be thought

independently of infinite space, then the representations of point, line, surface, and other

spatial objects should have to contain the representation of infinite space as a component,

and this is not the case, because (1) spatial objects can be thought independently of


would consider this as necessary or even be aware that he should actuallyencounter such a thing in the representation of a spatial part? Nor is ittrue that the representation of a line is required for the representation of apoint, or the representation of a plane for the representation of a line, orperhaps even the representation of a body for the representation of aplane. Nonetheless we usually define a point as the limit of a line, a lineas the limit of a plane, and so on. But are these [66] definitions reallytrue? If they were, then of course we should have to concede that wecannot represent a point without a line, and so on. Even so, is theresomeone who does not feel that this is not the case? After all, there arelines (with a double curvature) of which it is hard to indicate the planethat limits them. – Instead, we must assume that the representation of aspace appears as a part in the representations line, surface, body, and soon, and that the whole of space should be thought and defined as thecollection of all points.190 However, we should reject the completelyungrounded claim that the concept of finite space (e.g., triangles ingeneral) is based only on limitations of infinite space, since there are alsoparts of a whole, such that their concept contains the concept of a whole,as does the concept half a yard. Although there are parts of a whole, suchas the concept half a yard, which contain the concept of a whole, thereare also parts of another kind where this is not the case, such as the partsof a watch or of a steamengine, and so on, and the parts of space calledpoint, line, and so on, belong to the latter kind.191 It is therefore false that

infinite space; (2) the logical or mathematical representations of a line, plane or triangle

do not contain the representation of infinite space as a component; (3) even if we were to

admit that no part of space can be thought without the whole of space, this is merely a

psychological feature of our partial representation of space, and it does not follow from

this that infinite space is an intuition; cf. WLI, §79.6, note. N.B. Unlike Bolzano, Kant did

not allow for an actual infinite.

In Kant’s view, we grasp the concept of space as a quantity or whole consisting of an

infinite number of parts by constructing them through the introduction of limitations or

segments. For Kant, space is not a general concept but a singular a priori intuition with an

unlimited intension-it consists of an infinite number of parts, and the concept of space is

derived from the underlying intuition of space by means of a constructive procedure

which represents the infinite divisibility of space. We do not have an immediate intuition

of a determinate space but a successive grasp of parts which constitute a whole. And the

whole, in this case space, is graspable as a quantity consisting of an additive number of

similar parts. In this way, Kant can claim that the general concept of space is an extensive

quantity without any quality-a claim rejected by philosophers from Berkeley to Bolzano

and Stumpf.190

This is how Bolzano defines space in PU, § 17, and WLI, § 79.3, 79.6, note c.191

Bolzano draws a distinction between parts and their relation to a whole: (1) parts which

can be thought independently of a whole and (2) parts which cannot be thought

independently of a whole. For example, the parts of a watch can be thought independently

New Anti-Kant 119

the parts of space are constituted such that they contain the whole ofinfinite space as a component, and this claim is by no means appropriatefor proving the proposition that space is an intuition. – Likewise it seemswrong to us “that every concept can be thought as a representationcontained in an infinite collection of different representations as theircommon [67] mark, which thus contains these representations under

itself.”192 Not every concept must have several or even infinitely manyobjects; and if it does not have them, how can there be several or

of the whole watch, but half a yard cannot be thought independently of the whole yard. In

addition, the concept [yard] is necessary for defining the concept [half a yard]. Kant’s

limited spaces belong to (2) because the representation [limited space] is contained in the

representation [space], but according to Bolzano this is wrong, since it is not necessary to

represent the whole of space in order to think a single point; nor in order to think a point,

do we first have to think the line which it limits. On the contrary, the representation

[point] is a part of the representation [line] in sense (1). Besides, even if [limited space]

were part of [space] in sense (2), this does not prove that every whole whose parts are of

this kind must necessarily be an intuition, cf. WLI, § 79.6, note c. In addition, Bolzano

objects against Kant’s view that since parts of space cannot be independent, they cannot

be represented independently (of the representation of infinite space in which they are

contained). In Bolzano’s view, it does not follow that if we think a limited space, say, a

cone, without thinking of the space outside this cone, that we deny that there is such an

outside space, ibid., note m.c (on Schultz’s discussion of Kant). This point was taken up

by Schwarz (1911), who argues that if the parts of infinite space cannot be represented by

themselves as Kant claims, the representation of infinite space cannot be complex.

However, Kant also claims that the representation of space consists of a number of partial

representations in it, and therefore it cannot be simple either. So there is a contradiction in

Kant’s proof of the intuitive nature of space; cf. Ph. Schwarz, Bolzano’sVorstellungstheorie und Kant’s Lehre von der Raumanschaung, 1911. While in no way

diminishing the importance of these objections, I think that a slight readjustment of our

perspective of Kant’s aims is appropriate at this point: in the eighteenth century, infinite

divisibility is represented by an intuitive or constructive procedure for cutting out a

smaller line segment from any given one. Kant says that the general concept of space rests

on limitations because, for him, cutting out parts of space is the only way we can

represent the general concept of space (which has an infinite extension but a finite

intension, unlike the singular intuition of space, which is divisible into an infinite number

of parts). So the single intuition of space underlies the general concept of space because

we can only represent the latter by introducing limitations into the former. And what

underwrites this claim is Kant’s cognitive starting point: our cognition of space, that is to

say, our geometrical cognition (which presupposes a Euclidean space), proceeds from the

construction of concepts and mathematical proofs proceed by constructing their object; cf.

KrV, B39-40; B211-212; B741-742; P, §12. Cf. also M. Friedman (1992), pp. 69-71.192

Once again, Přihonský paraphrases Kant’s text whilst purporting to quote him. Kant

writes: “Now we must think every concept as a representation contained in an infinite

collection of different possible representations (as their common mark), which hence

contains these representations under itself.”, KrV, B39-40. “Now we must think every

concept as [. . .].” is a directive which is not quite the same as Přihonský’s version that

“every concept can be thought as [. . .].”, which is an assertion.


infinitely many representations contained under it? – We agree that “noconcept as such can be thought as containing an infinite collection ofrepresentations within itself”,193 if it amounts to saying that no conceptthought by a finite understanding can be composed of an infinitecollection of partial representations, but this applies to everyrepresentation and not only to concepts. Therefore we do not believe thatspace is a representation composed of an infinite number of parts. Thereason “for all parts of space, ad infinitum, are simultaneous”194 does notprove this. For to form the representation “of a whole which is composedof infinitely many parts”,195 I do not need to represent these partsindividually; instead, I have already thought such a whole just bythinking the concepts producing this concept in the appropriatecombination. Nevertheless, there are only a limited number of theseconcepts, as has already been shown (see the introduction). – – As withspace, Kant also asserts of time “that every determinate magnitude oftime is only possible through limitations of a single time grounding it(namely, infinite time)”,196 which is probably supposed to mean that partsof time can only be represented by limitations of (the whole of) infinitetime [68] and that each representation of a determinate part of timecontains the representation of the whole of infinite time as a component.But as we already pointed out against Kant’s similar claim about spaceand its parts, we must reply here as well that we are quite capable ofrepresenting a moment or a time span without thinking of therepresentation of the whole of infinite time. On the contrary, to think thelatter, we must already have the representation of a moment, sinceinfinite time is but the collection of all moments.197 The major premissKant uses here in order to infer the conclusion that time is an intuition isjust as false: “that an object, whose parts can only be represented throughtheir limitation, cannot be represented through concepts”.198 There areenough concepts whose parts are represented only in relation to thewhole (the limitation of the whole in the very sense in which it can applyto time) but which we nevertheless cognize through pure concepts. A

193

KrV, B40.194

KrV, B40.195

This quote does not appear in Kant’s text.196

KrV, B47-48.197

Cf. WLI, § 79.3; PU, § 17.198

The passage in KrV, B48, quoted by Přihonský reads: “But where the parts themselves

and every magnitude of an object can be represented determinately only through

limitation, there the whole representation cannot be given through concepts (for they

contain only partial representations)”.

New Anti-Kant 121

terminus medius in a syllogism can only be cognized by considering thewhole, and yet the representation of a syllogism is a pure concept.

But Kant believes that3. he can confirm the correctness of his view [69] of space as a pure

intuition by means of geometry, “a science that determines the propertiesof space synthetically and yet a priori. [. . .] For from a mere concept nopropositions can be drawn that go beyond the concept, but this doeshappen in geometry” (p. 65).199 Our philosopher sets great store by hisaccount and is very confident about his proofs, as we can gather from theconcluding passage of the “Transcendental aesthetic”, which reads asfollows:

“here we now have one of the required pieces for solving the general task of

transcendental philosophy: how are synthetic a priori propositions possible?Namely, [through] the pure a priori intuitions200 space and time, in which we, if

we want to go beyond the given concept in an a priori judgment, encounter what

cannot be discovered a priori in the concept but which is certainly so discovered

in an intuition corresponding to that concept and can be combined with it

synthetically. But because of this, such judgments can never extend beyond

objects of the senses and can hold only for objects of possible experience.”

(p. 87)201

Objections. That synthetic propositions cannot be derived from mereconcepts without adding an intuition is the πρωτον ψευδοζ of theCritique, and it is not demonstrated anywhere.202 Kant appeals to themathematical sciences, in particular to geometry, in order to support hisclaim, [70] and he assumes that geometrical principles are mediated

199

Přihonský’s quote leaves out the following sentence: “What then must the

representation of space be for such a cognition of it to be possible?” (KrV, B40)200

A coherent translation of this passage in B73 is not easily achieved, and I have had

recourse to Mellin’s edition of the Critique, where the sentence reads nämlich durch reineAnschauungen, rather than using nämlich reine Anschauungen from Hartknoch’s edition.201

KrV, B73.202

Bolzano uses the same expression but in a different objection against Kant. Bolzano

says that the ω ~ψ υ ~of modern German philosophy results from a failure to

grasp the notion of concept with sufficient clarity: a concept was either confused with a

thinking or completely assimilated with the thing that is the object of that thinking. So, in

his view, the first error of Kantian philosophy is a confusion among a mental act, a

concept, and the object of that concept, cf. “Ueber den Begriff des Schönen” (1843), in

Mathematisch-physikalische und philosophische Schriften 1842-1843, BBGA, vol. 18, p.

99. According to Přihonský, the ω ~ψ υ ~of Kant’s first Critique is the

latter’s claim that the objective validity of synthetic a priori judgments is based on pure

intuitions and that Kant fails to prove not only that synthetic a priori judgments are

derived from intuition but also that they cannot be derived from concepts.


through pure intuitions.203 But we have seen that the outlook for hisdoctrine of pure intuitions is not good, and so we understand why our

203

According to Kant, synthetic a priori judgments require pure intuitions, because from

mere concepts no synthetic cognition can be attained (cf. KrV, B64-65). Kant claims that

pure intuitions are necessary for grounding synthetic conceptual truths, and he justifies

this assumption by extending the use of the intuitions of space and time from the domain

of geometry to metaphysical judgments. Přihonský’s first objection is that although Kant

claims that synthetic a priori judgments are grounded on pure intuitions, he does not show

how these judgments originate from pure intuitions (cf. p. 24 above). Přihonský then

argues that the basic error of the first Critique lies in the false guarantee of synthetic

judgments provided by pure intuitions, for Kant’s theory is based on the assumption that

synthetic propositions cannot be composed of pure concepts without adding an intuition,

but he does not show that pure synthetic judgments cannot be inferred from pure concepts

without referring to an intuition.

Přihonský’s “ ω ~ψ υ ” objection is echoed by R. Zimmermann’s claim that

the Critique is based on a mathematical prejudice, namely, that mathematical propositions

are synthetic a priori because they are grounded on pure intuitions, in Ueber Kant'smathematisches Vorurtheil und dessen Folgen (1871), pp. 35-40. Zimmermann, another

Bolzanian pupil, reconstructs Kant’s theory as follows: once Kant had established beyond

doubt that mathematical judgments, which are undoubtedly a priori, are synthetic, he had

also shown the "natural conditions" (P, pp.279-280) of synthetic a priori judgments and

could "ascend from these actual, and, at the same time, well-grounded pure cognitions apriori to a possible condition of the kind that we are seeking, viz., to metaphysics as a

science" (ibid., KrV, B20-21). Kant had to start from a knowledge beyond doubt, such as

mathematical knowledge, and then determine the conditions of this knowledge and extend

them to all possible representations which claim to be knowledge. Thus he based his

enquiry about the possibility of synthetic a priori judgments on mathematical propositions

"which actually exist" (KrV, B20). Having ascertained that there is knowledge a priori, he

could also be certain that there must be necessary conditions of this knowledge and could

look for them. Once we have understood that synthetic a priori judgments are actually

given in these latter, we only need to search for the conditions under which mathematical

propositions are possible to obtain the conditions of possibility of synthetic judgments apriori. In this way, all synthetic a priori judgments are grounded on mathematical

propositions, that is, on the assumption that the latter are not only a priori but also

synthetic (P, §§10-12; KrV, B20-21). Zimmermann objects to Kant that mathematical

propositions are analytic and thus a priori, so no intuitions are needed to mediate between

subject and predicate. Kant assigned a synthetic property to these judgments, which

guarantees the existence of a priori elements of sensibility, so as to hide the shaky

assumption of a priori intuitions under the credibility of the mathematical nature of these

judgments. In fact, none of Kant’s Austrian critics accept either Kant's pure intuitions or

the need for mathematical judgments to be synthetic; cf also Bolzano in WLIII, § 315.3;

MM4, § 96, p. 138.

But Bolzano accepts conceptual synthetic judgments in mathematics and logic; cf.

Beyträge, §§ 19, 21; WLIII, § 315.2. He says that axioms, the principle of extensionality

and the principle of inclusion, or general truths, such as: “Any triangle has the attribute

that the sum of its angles is equal to two right angles”, are conceptual synthetic judgments

(WLII, § 148.1). And thirteen years after Zimmermann, Frege (1884) defends Kant’s

synthetic definition of geometry, as well as his attempt to prove that there are synthetic

New Anti-Kant 123

philosopher was unable to prove how the origin of synthetic a priori

judgments is explicable by means of intuitions of this kind, despite hisfrequent attempts in the Critique. On p. 159, for example, he tries todraw the reader’s attention to a highly important distinction (in his view)

“between an intuition corresponding to a certain concept, such as that of a

triangle, and which originates in a real mark of the object, and an intuition which

arises, or is constructed, in the mind by means of imagination. The former”, he

asserts, “is empirical and neither universality nor necessity are ascribed to it,

whereas the latter is pure and has both properties.”204

But we do not see in what way the representation of a triangle canessentially change if we merely imagine it, rather than draw it on ablackboard.205 Kant seems to have felt this, for now and then he demandsthat this construction should not occur through imagination as such, but

judgments a priori: “In calling the truths of geometry synthetic and a priori, he revealed

their true nature. And this is still worth repeating, since even today it is often not

recognized. If Kant was wrong about arithmetic that does not seriously detract, in my

opinion, from the value of his work. His point was, that there are synthetic judgments apriori; whether they are to be found in geometry only, or in arithmetic as well, is of less

importance.” The Foundations of Arithmetic, trans. J. L. Austen (Oxford: Basil Blackwell,

1986), pp. 101-102 (1st edition, Blackwell, 1950).204

I have not found this quote in the Critique, but I presume that Přihonský summarises a

passage from the “Schematism of the pure concepts of the understanding,” KrV, B176-

180, especially with regard to the vague wording which could be directed against the

“hidden art in the depths of the human soul whose true operations we can divine [. . .]

with difficulty.” (B180). Přihonský’s use of the word “constructed” also suggests a

reference to B746 where Kant discusses the polemical question of defining geometrical

concepts by constructing them through intuitions. Bolzano criticizes Kant on this in

WLIII, § 305.5. But Přihonský inaccurately ascribes to Kant the idea that an intuition is

constructed by the imagination. Rather, an object is exhibited or constructed through

intuitions by the imagination, which provides a concept, say, [triangle] with an image (of

a triangle). Also, an intuition does not originate in the mark (Verzeichnung) of an object,

but an object is given in intuition.205

(1) A mental representation of a triangle is not the same as a physical representation of

a triangle on the blackboard; (2) Kant distinguishes between a schema and an image. The

former is a product of the imagination which can only exist in the mind, “in accordance

with which the image first becomes possible” (KrV, B181), and it corresponds to the

general concept [triangle], unlike the image of a triangle, which only serves to illustrate

the general concept. This distinction is Kant’s solution to the problem of how to

generalize on the basis of a single triangle: the schema is a rule for providing the concept

[triangle] with an image and makes possible a rigorous representation of this concept by

providing concepts “with a relation to objects and thus with significance” (B185), but we

cannot derive a concept from a single image or drawing (which is the object falling under

this concept), and an image cannot “attain the generality of the concept”. This is why Kant

insists that the “schema of the triangle can never exist anywhere but in thought and

signifies a rule of the synthesis of the imagination with regard to pure shapes in space.”

(KrV, B180)


through pure imagination. This is what we call hiding behind vaguewording. – Things [71] do not become any clearer when Kant reminds usthat “imagination as such” (which he calls productive imagination)“produces mere images, but pure imagination produces schemata, and aschema is the representation of a method or rule to provide a conceptwith an image, rather than the image itself.”206 Here it is difficult tounderstand how the representation of a method can be called an intuitionor how this representation can be distinguished from the method toprovide each concept with its image.207 The only way in which we canthink of this method as something specific is to take it as the so-calledgenetic definition of a concept.208 But who would assign this to

206

Cf. KrV, B181. This passage reads: “The image is a product of the empirical faculty of

productive imagination; the schema of sensible concepts (such as figures in space) is a

product and, as it were, a monogram of pure a priori imagination through which and

according to which images first become possible. But the images must always be

connected to the concept, with which they are never fully congruent, only by means of the

schema they designate.” The last part of Přihonský’s quote combines two sentences from

the Critique: “Now this representation of the general procedure of the imagination for

providing a concept with its image is what I call the schema for this concept”, B180, and

[a schema of the imagination is] “a rule for the determination of our intuition in accord

with a certain general concept.” (B180)207

It seems that Přihonský’s difficulty is that he misunderstood Kant, for (1) the

representation of a method is not an intuition. It is a schema which is the product of the

imagination and (2) this method is, precisely, the means (or the rule) for providing a

concept with its image; cf. KrV, B180, and my notes 203 and 204.208

This is Bolzano and Přihonský’s main objection to Kant’s schematism: for them, the

schema is the genetic definition of a concept, whereas for Kant it is a rule for concept-

application. In addition, they reject Kant’s distinctions between schema, concept, and

image. They also claim that there is no necessary relation between Kant’s schema and the

concept it exhibits and reject Kant’s recourse to constructing mathematical concepts

through intuitions. Bolzano says that a schema is the rule according to which an object can

be provided for a concept, so that the concept can be defined, and that there is no reason

why it should be partly sensible or intuitive, merely because it can represent an object. To

define the concept [circle], we follow a rule (a schema) for the ostension of a circle,

“namely, the concept of a line.” (WLIII, § 305.5). The figure of a circle is generated in the

schema, so that the concept [circle] can be defined. Thus the rule (the schema) which

ostends a concept or provides an image for a concept is the genetic definition of a circle,

namely, the concept of a circle. Kant would reply that mathematical concepts are first

given through definition and constructed through intuitions, so he could sidestep Bolzano

and Přihonský’s objection; cf. KrV, B759.

Bolzano and Přihonský object (1) to the partly intuitive nature of the schema, and this

objection is grounded in their rejection of the Kantian distinction between pure and

empirical imagination. According to them, there is only a distinction between the degrees

of vivacity of the representation in question but not between two kinds of imagination.

The problem, as they see it, is that both image and schema are intuitive for Kant, so there

cannot be a generic difference between them, and therefore Bolzano and Přihonský do not

New Anti-Kant 125

intuitions? If the schema of a circle is the representation of the methodshowing how to provide this concept with an object, then surely it is onlythe way and manner in which a circle originates, that is, the well-knowngenetic definition of a circle, or the concept of a line described by a pointthat moves in a plane in such a way that it always maintains the samedistance from a given point. That is why anyone who cognizes the truthof a synthetic judgment by looking at the schema of its subject onlycognizes this truth by looking at a concept and not at an intuition. – Inseveral places in the Critique, we find the assertion that we cannot thinka spatial relation without constructing it. [72] For example, on p. 143 hesays: “we cannot think a line without drawing it in thought, we cannotthink a circle without describing it, we cannot represent the threedimensions of space at all without placing three lines perpendicular toeach other at the same point, and we cannot even represent time without,in drawing a straight line (meant to be the external figurativerepresentation of time), attending merely to the action of the synthesis ofthe manifold through which we successively determine the inner sense,and thereby attending to the succession of this determination in innersense.”209 Is this not exaggerated? Even though such constructions orimages may increase the clarity and vivacity of our representations, andtherefore cause us to become accustomed to them and recall themspontaneously, they are not necessarily related to our concepts and do notoccur everywhere, as these examples show. Even a beginner inmathematics can understand what we want to say when we define adodecahedron as a body which is bounded by twelve identical planefaces. On hearing these words, he will connect a concept with them andthink something true by that. But this will not enable him to understand

accept Kant’s distinction between a sensory image and a nonsensory schema, and in their

view Kant’s claim that the schema makes a concept appear before our eyes, as it were,

implies that the schema itself is also an image. And the fact that the schema is partly

intuitive (because of the ostention or exhibition of a concept) also implies that it has

figurative or illustrative features.

From this objection follows another one: (2) the geometrical construction of a concept is

not indispensable. In other words, it is not necessary that in order to think a line or a

circle, we first need to describe them, that is, to draw them. Such a drawing may increase

their vividness or comprehensibility, but they are not essential for a definition of the

concepts [line] or [circle]. In addition, intuitions are not even sufficient for defining a

concept, for how could we construct concepts such as [dodecahedron] or [a finite spatial

object which is fully determined by two of its points, and each point of which stands in

the same relation to the whole]? In the latter case, as Bolzano says, we may not even know

right away whether this spatial thing belongs to the class of lines, surfaces, or bodies, let

alone produce a construction of it; cf. WLIII, §305.5.209

KrV, B154, Kant’s italics.


how to construct a dodecahedron; he will not know whether the boundingfaces are triangles or pentagons.210 Instead, we must presuppose that hehas clearly grasped the concept of a [73] dodecahedron before he is ableto construct one. We do not obtain the concept of a spatial thing bymeans of a drawing; rather, we already have this concept before we candraw the corresponding object, even if it is only in our imagination. –When mathematicians use certain signs and figures for proving theirtheorems in arithmetic and geometry, it is not because the propositionsare not provable without assistance, but because in that way they cut backon words and facilitate their audience’s efforts at understanding throughillustration. Kant’s doctrine of the symbolic construction of algebraicconcepts is bound to appear extremely forced to anyone taking a closerlook. He would not understand in which sense he is supposed to take theassertions in the Critique that mathematical concepts are expressedthrough intuition by means of the signs x, y, +, –, and so on, insofar asthis is supposed to be particular to algebra. For do not the logician andeven the mathematician occasionally use certain sensory signs toillustrate the concepts in their sciences? The logician, for example, usescircles or lines for explaining the relation of subordination betweenconcepts. – So the outlook for the assertion that arithmetical propositionsare [74] mediated and constructed through intuition is not good! If weask what this intuition is, we hear (p. 47) “that we cannot think theconcept of 12 as a sum of 7 + 5 without going beyond the concept itselfand seeking assistance in an intuition, such as that of five fingers.”211 Butobviously that is not a pure intuition. For this reason Kant says elsewhere

210

Cf. WLIII, §305.5.211

Cf. KrV, B15, intro. V, which reads: “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 analyse my concept of such a possible sum, I will still not find twelve in it. One

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 Senger’s arithmetic) five points, and one after

another, add the units of the five given in the intuition to the concept of seven.” Cf. also

Scholz (1931, pp. 219-220), who notes that Přihonský indicates the ideal of a geometry

without figures discussed by Bolzano (1810) in the Beyträge and the Anti-Euklid (~ 1830),

as well as in WLIII, §§ 305, 306, 314, and that his postulate was introduced by Leibniz

(Nouveaux Essais IV 1): “the cogency of a principle is independent of the diagram, whose

only role is to make it easier to understand what is meant and to fix one’s attention. It is

universal propositions, i.e., definitions and axioms and theorems which have already been

demonstrated, that make up our reasoning, and they would sustain it even if there were no

diagram.” New Essays on Human Understanding, trans. P. Remant and J. Bennett

(Cambridge: Cambridge Univ. Press, 1981). Cf. also Sebestik’s reconstruction of the

dispute between Kant and Bolzano on mathematical knowledge, La dispute de Bolzanoavec Kant (op. cit., forthcoming).

New Anti-Kant 127

that arithmetical propositions are grounded by the pure intuition of time,for when we count, we only add units one after another. But in this waytime should ground all our judgments. We are unable to grasp thoughtsotherwise than in a sequence, one after another, and to represent them bywords. Now what is the difference between an arithmetical presentationand others? Actually there is no essential difference, since in arithmeticwe must infer principles from concepts, as in every scientificpresentation. – It seems as though Kant rightly claims that geometricalpropositions are grounded by intuition, namely, the intuition of space,because here we usually derive inferences from depicted figures. But thequestion is whether this is necessary. We are familiar with thegeometrical description by which the collected truths of geometry can beinferred without the help of figures.

In any case, Kant may declare only those synthetic [75] cognitions asright whose concepts refer to intuitions, and he may repeat that conceptsdo not produce cognitions without a corresponding intuition and thatconcepts are empty, and so on, as often as he wants. We must object andpose the following question: what are these pure intuitions on which

these assertions and similar ones are based? Our philosopher passes

these assertions off as truths and thus as cognitions that are obviously

not merely analytic propositions but synthetic ones.According to Kant, the other part of the doctrine of elements, the

transcendental logic, consists of two divisions: an analytic, which merelypresents the elements of pure cognition of the understanding and theprinciples without which no object can be thought; and a dialectic, whichreveals the false illusion in the hyperphysical use of reason (pp. 88–98).212 The “Transcendental analytic” consists of two books (p. 99), thefirst one containing the concepts and the second one, the principles ofpure understanding.213

Our philosopher begins by reminding us (book 1, p. 99) that “whatmatters in the analysis of our entire a priori cognition of the pureunderstanding is a) that the concepts are pure, b) that they are elementary

concepts [76] which are distinct from the derived concepts of which theyare composed, and c) that their table is complete and fills out the wholefield of pure understanding.”214 He continues by saying that although we

212

Cf. KrV, B87.213

Cf. KrV, B89-90.214

The passage at KrV, B90-91, reads: “I understand by the analytic of concepts [. . .] [the]

analysis of the faculty of understanding itself, in order to explore the possibility of apriori concepts by seeking them only in the understanding as their birthplace and

analysing its pure use as such. [ . . .] Hence we shall trace the pure concepts to their first


cannot presuppose this completeness with certainty by turning to anaggregate brought about by mere experiment, we can nevertheless obtaina complete table of elementary concepts, also called categories, in thefollowing way:

“if we abstract from all the content of a judgment and attend only to the mere

form of understanding in it, we find that the function of thinking in judgment can

be brought under four titles, each of which contains under itself three moments.”

“1. Quantity of judgments: universal, particular, singular.

2. Quality: affirmative, negative, infinite.

3. Relation: categorical, hypothetical, disjunctive.

4. Modality: problematic, assertoric, apodictic.” (p. 103)215

“This function <(activity)>”, we read on p. 110,

“that gives unity to the different representations in a judgment also gives unity to

the mere synthesis [77] of different representations in an intuition, which [,

expressed generally,]216 is called the pure concept of understanding. Thus the same

understanding, 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 analytic

unity, also brings a transcendental content into its representation by means of the

synthetic unity of the manifold of the intuition as such, which is why concepts

that pertain to objects a priori are called pure concepts of the understanding; and

seeds and predispositions in human understanding, [. . .].” In what follows, Kant does not

claim that the pure concepts of the understanding are elementary and are distinct from

derived concepts. Instead, he has two claims: (1) “that the function of thinking in

judgment can be brought under four titles, each of which contains under itself three

moments.” (KrV, B95). Then he presents his table of the forms of judgment or functions

of thinking. This table is divided into four basic forms, namely, quantity, quality, relation,

and modality. He claims (2) that the categories or pure concepts of the understanding can

be derived from these forms of judgment, for “[t]he same function that gives unity to the

different representations in the judgment also gives unity to the mere synthesis of

different representations in an intuition, which, expressed generally, is called the pure

concept of understanding.” (B105) These pure concepts are then presented in the table of

categories (cf. B106). So the understanding gives rise to the functions of judgments and

pure concepts, and transcendental logic has the task of examining a) how they pertain to

objects a priori and b) how the categories can be derived from the forms of thought (in the

transcendental deduction); cf. B117, B124ff., §§ 13-14.

As for the distinction between elementary and derived concepts, which Přihonský

attributes to Kant, of course the latter distinguishes between the categories as the true

genealogical core concepts of pure understanding and their equally pure derivative

concepts, which he calls the predicables of pure understanding, in contrast to the

predicaments. But he immediately adds that in the Critique his concern lies “not with the

completeness of the system but only with the principles for a system”, and therefore he

does not develop this issue; cf. KrV, B108.215

KrV, B95. Přihonský adds (Tätigkeit).216

My insertion, in accord with Kant’s text.

New Anti-Kant 129

this can never be accomplished by general logic.

In this way there arise exactly as many pure concepts of the understanding, which

apply a priori to objects of intuition as such, as there are logical functions of all

possible judgments in the previous table; for the understanding is completely

exhausted and entirely measured by these functions. Following Aristotle, we will

call these concepts categories, for our aim is basically identical with his, although

very different in its execution.”217

The following table of categories accordingly follows from the table ofjudgments:

1. Categories of quantity: unity, plurality, totality.2. Categories of quality: reality, negation, limitation.[78]

3. Categories of relation: inherence and subsistence, causality anddependence (cause and effect), community (reciprocity between agentand patient).

4. Categories of modality: possibility–impossibility; existence–nonexistence; necessity–contingency.218

Objections. Kant obviously intends to establish an entirely newscience which he calls transcendental logic. The analytical part oftranscendental logic supposedly discovers pure concepts of theunderstanding (the categories) and pure principles, whereas thedialectical part allegedly teaches us to uncover the false illusion in theuse of the understanding.219 Here we will leave the question of whether itis necessary to draft a proper science for such a purpose and will justbriefly discuss what we consider unclear and doubtful in the analyticalpart and above all in the doctrine of the categories.

First and foremost we notice that Kant1. does not begin by defining the concept of a category. Only in what

follows and as if in passing, he tells us “the categories are concepts of anobject as such, by means of which we consider that an intuition of it isdetermined for one of the logical functions of judgments.” (p. 126)220

217

KrV, B105. The division into paragraphs is Kant’s, not Přihonský’s.218

KrV, B106.219

Cf. KrV, B82. The transcendental analytic is a logic of truth because it examines “the

agreement of cognition with its object” (B82) and “expounds the elements of the pure

cognition of the understanding and the principles without which no object can be thought

at all” (B87). However, the transcendental dialectic is a logic of illusion because it deals

with the unrestricted hyperphysical use of the understanding and reason; cf. B88.220

Rather than defining the concept [category] in this passage summarily quoted by

Přihonský, Kant explains that the objective validity of the categories rests on the fact that

through them alone experience is possible, since “concepts of objects as such ground all

experiential cognition as a priori conditions; consequently, the objective validity of the

categories, as a priori concepts, rests on the fact that through them alone experience is

possible (as far as the form of thinking is concerned). For they are then related necessarily


[79] This definition is faulty, at least insofar as it does not indicate theinternal difference between the categories and other concepts but merelyspecifies the relation in which these concepts stand to our faculty ofthinking. – We are not denying that categories are pure concepts, sincewe hold that all concepts are pure or a priori, without exception. ButKant claims that categories are “elementary concepts” (p. 99),221

“original and primitive concepts” (p. 112),222 or “genealogical core

concepts” (ibid.).223 It is not clear in what sense he calls them that. Thesenames almost lead us to believe that he considers categories as simple

concepts from which the rest arise through composition, since he thenactually talks of concepts that are derived from them (ibid.).224 But newconcepts can only be formed or derived from concepts alone by means ofcomposition. However, this is contested when Kant asserts on the samepage that, although he knows the definitions of the individual categories,he deliberately ignores them. Nonetheless, definitions in the authenticmeaning of the word can only be provided by composed concepts. HenceKant must consider categories as composed concepts. And indeed, almostall of them are such that they can be dissolved into simpler concepts anddefined through them. So how can we call them elementary concepts,[80] or anyway, what are we supposed to think of the name category, inthe Kantian sense?225

In addition, we cannot agree with the claim that2. the table of categories is deducible from the table of judgments and

that there are as many categories as there are different forms ofjudgment. We cannot accept this kind of derivation. For even admittingthat the concept of plurality is contained in particular judgments, at leastin some of them, how can we possibly show that the concept ofuniversality is contained in universal judgments and the concept of unityin singular judgments?226 The universal judgment: all glass is fragile,expresses nothing but: glass is fragile, so these two judgments areequivalent, but where is the concept of universality in the secondjudgment? Nor can we find the concept of unity in the singular judgment:

and a priori to objects of experience, since it is only through them that any object of

experience can be thought at all.” (KrV, B126)221

Cf. KrV, B89.222

Cf. KrV, B107.223

“Genealogical core concepts” translates Stammbegriffe; cf. KrV, B107.224

Cf. KrV, B108, and my note 214, second paragraph, above.225

In the Kantian sense, the word ‘category’ means “form, according to which experience

is ordered”, “function of unity in judgments”, “objective condition of validity of

judgments”, or “pure concept of the understanding”, cf. P, § 39; KrV, B102ff.226

Cf. WLI, §119; WLII, §188.1.

New Anti-Kant 131

Berlin is the capital of Prussia, for these two representations are verydifferent, namely, the concept of a single object and the representation ofthe city of Berlin.227 – We find it even more difficult to understand howthe three categories of relation can arise from the corresponding threeforms of judgment.228 At the level of relation, each judgment should beeither categorical or hypothetical or disjunctive. Certainly the followingare examples of categorical judgments: a point is that which is [81]

simple in space, a line is limited by two points, the possible is not yetreal, and every concept is a kind of representation. We cannot understandhow the subject- and predicate-representations of these propositions cancontain the concepts of substance and accident, or even only besubsumed under them.229 Nor can we derive the categories of communityor reciprocity according to logical rules from the reciprocal limitation,which in Kant’s view, the two spheres of subject and predicate exert oneach other in a disjunctive judgment.230

In addition, it is wrong3. to divide all categories according to the four moments of quantity,

quality, relation, and modality.231 For the properties of each object andthus also the attributes of categories can be justifiably brought under thetwo rubrics of property and relation, and they therefore belong to eitherquality or relation, and it is a mistake to divide them once moreaccording to quantity and modality.232 For does not the magnitude of anobject belong to its properties and does not modality, which, in Kant’sview, is the relation to our cognitive faculty, obviously belong to arelational moment (relation)? In any case it is inappropriate to considerboth (quantity and modality) separately from their superordinateconcepts.[82] Finally,

227

Unlike Bolzano and Přihonský, Kant makes an interesting point about the distinction

between universal, particular, and singular propositions. He agrees with Bolzano that, for

syllogistic purposes, singular judgments behave as universal judgments: if they are

affirmative, the subject term is distributed: “there is an A which has b-ness”. However,

Kant also claims that a singular judgment differs from general or particular judgments

with regard to knowledge, because in a singular judgment we say something about a

definite object: “this A has b”, but not in a general judgment, cf. KrV, B96.228

Cf. WLII, § 190.1.229

Cf. WLII, § 190.1, where Bolzano objects against Kant that we can show subject and

predicate not only in categorical propositions but in every proposition if we only

understand them correctly.230

Cf. WLII, § 181, where Bolzano opposes Kant’s division of disjunctive judgments with

a distinction between disjunction in the strict sense and a wider sense, which corresponds

to the contemporary usage of inclusive and exclusive disjunction.231

Cf. WLI, § 119.232

Cf. WLI, § 116.2, where Bolzano makes precisely the same point.


4. although Kant asserts that the table of categories is “complete andfills out the whole field of pure understanding”233 and that there thusmust be as many categories as there are forms of judgment, he has notproved this anywhere. For surely the assertion that the understanding actsin the same way when forming categories and forms of judgment cannottake the place of a proof? Certainly, the understanding works in one waywhen uniting representations to form concepts (namely, composedconcepts) and in another way when combining representations so thatjudgments arise from them.234

Kant adds the deduction of the pure concepts of the understanding tothe table of categories (pp. 118ff.).235 But by this deduction heunderstands “the proof that we are entitled to use the pure concepts of theunderstanding <(categories)>, but only in relation to objects of a possibleexperience” (p. 124).236 In his view, the intuitions of time and space donot require a special deduction, “for they contain the conditions ofpossibility of objects as a priori appearances.” But this is not the casewith the categories, “for even without thinking objects, they can be givento us in intuition.”237 – [83] Kant believes he can remove this difficulty

233

Kant does not say that the table of categories fills out the field of pure understanding,

although Přihonský already uses this expression on p. 76, above. The passage at KrV,

B105, reads “the understanding is completely exhausted and its capacity entirely measured

by these functions” (that is, by the categories); cf. also my note 214 and Přihonský’s quote

of this passage on p. 76, above.234

For Kant, the understanding is the faculty not only for thinking but for judging. His aim

is to establish the primacy of judging over thinking by showing that concepts signify

something (that is, they have a relation to objects) only when they are unified in a

judgment. Hence he wants to “trace back all actions of the understanding to judgments”,

and since judgments are “functions of unity among our representations” (KrV, B94), the

understanding is also the faculty of unifying representations.235

Cf. KrV, B116, § 13.236

Cf. KrV, B117, where Kant writes that concepts destined for pure a priori use “always

require a deduction of their entitlement, since proofs from experience are not sufficient

for the legitimacy of such a use, and yet one must know how these concepts can be related

to objects that they do not derive from experience. I therefore call the explanation of the

way in which concepts can relate a priori to objects their transcendental deduction, [ . .

.]”.237

Kant says that neither time and space nor the categories require an empirical deduction,

because “they both relate to objects entirely a priori. [. . .] Thus if a deduction from them

is necessary, it must always be transcendental.” (KrV, B118) He continues by saying that

both space and time as well as the categories require a transcendental deduction: “[w]e

have above traced the concepts of space and time to their sources by means of a

transcendental deduction and explained and determined their a priori objective validity.”

(KrV, B119) In addition, the categories not only require a transcendental deduction,

because they refer to objects “not through predicates of intuition and sensibility but

through those of pure a priori thinking” and “since they are not grounded in experience

New Anti-Kant 133

by pointing out that “all the manifold of sensible intuition necessarilybelongs under the original synthetic unity of apperception, that is, it mustbe grasped under the representation: I think or, more precisely, it must beaccompanied by the awareness that it is my representation.238 [. . .]However, this can only happen by means of judgments; and the logicalforms of all judgments are laid down in the categories.239 [. . .] Thus allthe manifold of an intuition necessarily stands under the categories.240 –The categories do not provide knowledge without application to intuitionbut are empty;241 thus they have no use in cognition except as anapplication to objects of a possible experience.242 [In general], there areonly two ways in which we can think a necessary accord betweenexperience and concepts of their objects: either experience makes theseconcepts possible or these concepts make experience possible. The

and cannot exhibit any object in a priori intuition on which to ground their synthesis prior

to any experience, [. . .] but also make the concept of space ambiguous by inclining to use

it beyond the conditions of sensible intuition, on which account a transcendental

deduction of it was also needed above.” (KrV, B120-121) In what follows, Přihonský

strings together various claims from different parts of the Critique which I comment on

individually, below.238

Cf. KrV, B132, § 16, entitled: “On the original synthetic unity of apperception,” where

Kant argues that “[t]he I think must be able to accompany all my representations”.

Intuitions are representations which are prior to our thinking, but they have no meaning

for us unless they are accompanied by an apperceptive I think, and that is why “all the

manifold of intuition has a necessary relation to the I think in the same subject in which

this manifold is encountered.” Kant’s noted predecessor on the issue of apperception is

none other than Leibniz-although the latter was concerned with explaining distinct ideas

and degrees of awareness by examining the functions and properties of perceptions: thus

for Leibniz, the combined perceptions of a single monad can determine the world it

inhabits (Monadology, §§ 14, 23, 56-60). However, unlike Kant (and Descartes), Leibniz

claims that we are not necessarily aware of all of our perceptions, i.e., we can have

perceptions without apperception (ibid., § 14).239

Cf. KrV, B105, and my notes 214 and 234, above. In addition, and more importantly,

Kant holds that the logical form of all judgments is the apperceptive “I think” related to

representations in a (categorical) judgment by the copula “is”, which is why he says, § 19

(B140), “The logical form of all judgments consists in the objective unity of the

apperception of the concepts contained in them”. The copula expresses more than the

subject-predicate relation: it relates representations to the necessary unity of original

apperception and thus expresses the objective validity of a judgment; cf. KrV, B141-142.240

Cf. KrV, B132. The manifold of intuition necessarily stands under the categories

because it is accompanied by an apperceptive I think; cf. my note 238, above.241

Cf. KrV, B75. The famous passage reads: “Thoughts without content are empty,

intuitions without concepts are blind. [. . .] The understanding is not capable of intuiting

anything, and the senses are not capable of thinking anything. Only from their unification

can cognition arise.” Then, at B126, Kant goes on to explain that the categories are the

conditions of possibility of all experiential cognition”; cf. my note 218.242

Cf. KrV, B126.


former does not occur, for the categories are a priori concepts and henceindependent of experience. Consequently only the latter holds, namely,that the categories contain the grounds of the possibility of allexperience.”243

Objections. 1. It may be true that the judgment: I think (for this Ithink is a judgment and not a mere representation) [84] can accompanyevery one of our representations.244 It’s just that in our view it would bebetter to say that we are able to form a new representation of eachrepresentation that we have, and this representation is such that it refersto each one alone and may indeed be an intuition of it.245 Nevertheless,this claim is false: that each of our representations must be accompaniedby the judgment “I think” (this representation) myself. For actually thisoccurs only if we have elevated a representation so that it is clear for usor if we intuit it.246 In addition, since the judgment: I think thisrepresentation myself, is only of a single kind, it is not easy to see howall categories, that is, all concepts grounding each and every thinkableform of judgment, should be applicable to intuitions because of thissingle judgment.247 Besides, we readily admit that all the manifold of a

243

Přihonský refers to B125-126, where Kant deals with the problem of how the categories

are applicable to objects of experience, except that Kant does not put it this way. Instead,

he justifies the objective validity of the categories as forms of thinking by claiming that

they are concepts of objects in general which ground all experiential cognition.244

Bolzano (1834) says that this awareness of necessity accompanying an a priori intuition

is “already a (vaguely thought) judgment, and a synthetic one at that”, Einschaltung.Etwas über die kritische und einige neuere Philosophien in Deutschland (in the

Religionswissenschaft), § 62.245

Kant’s point, however, is that the apperceptive I think is not an intuition (and does not

belong to sensibility) but an act of spontaneity to which the manifold of intuition has a

necessary relation. The I think is the product of our self-awareness or original

apperception; thus it is a representation and not a judgment; cf. KrV, B132, § 16.246

The characterization of cognition as degrees of clarity goes back to Leibniz: a clear

representation is sufficient for cognition. According to Bolzano, we raise a representation

to a clear one by representing it to ourselves through intuition, that is, by means of a

perceptual act; cf. WLIII, §§ 280.4, 284.2. However, Kant does not subscribe to this

model of cognition as degrees of clarity: for him a representation is clear not because we

intuit it but because it falls under a concept, since intuitions without concepts are blind,

that is, they are not cognitions; cf. KrV, B75. So his model of cognition is rather one of

correspondence between a concept and an intuition, mediated by a schema or rule,

enabling the application of the first to the second: we cognize something when we can

represent it to ourselves according to a rule.247

The reason why Kant introduces the apperceptive I think is because he wants to solve

the problem of the unity of judgment. In Kant’s view, whatever gives unity to a judgment

cannot itself be a judgment but must be a representation. The content of a judgment is a

collection of representations, but the unity of this content is provided by the objective

apperception, and it is expressed by the copula which relates the manifold given in

New Anti-Kant 135

given intuition must stand under the categories, insofar as there is nokind of intuition we have received so far, or may yet receive, that cancontradict any one of our categories. In general, only judgments cancontradict each other, but not representations, and we cannot justifiably

make a judgment that contradicts a pure conceptual truth, based on

certain intuitions we have had. [85] Although we do not really intuitcomposed representations but rather believe that we intuit them, althoughbodies present themselves to our senses as porous, although we alwaysonly perceive finite things, and so on, all this must not mislead us tojudge in contradiction of pure conceptual truths: thus there are no simplesubstances, matter does not continuously fill space, and nothing isnothing infinite. Once the existence of a pure conceptual truth is decided,either because it is a self-evident principle or because it is derivedthrough right inferences, it cannot lose its trustworthiness because theeye points to judgments of a different kind; instead, appearance must becorrected by those completely conceptual truths.*248

intuition to a concept of the object. And the apperceptive I makes it possible for this

manifold of intuition to become an object of knowledge (a cognition); cf. KrV, B140-142,

§§ 18-9. So a judgment is the linguistic expression of a mental unity of representations-a

unity that is brought about by our understanding. Husserl would say that when we

understand or cognize a thing, we connect several representations in a mental act, by

means of an act-quality (Edmund Husserl [1901], Logische Untersuchungen, vol. 2,

Tübingen: Max Niemeyer, 1980; English translation: Fifth Logical Investigation, trans. J.

L. Findlay [London: Routledge, 1970, reprint 2001], § 20). In this way, Kant offers a

reply to an important semantic problem, namely, how we make statements (a meaningful

combination of representations or words) and how we communicate them (or produce the

same combination in someone else’s mind). Bolzano (and Přihonský) provide no

explanation of the unity of judgment. Bolzano claims that there is an objective, mind-

independent order of representations in a proposition: “A has b”, but he does not say how

we make this connection: “If I am asked what this connection is, I can answer only that it

must be a sort of mutual influence of these representations on each other. But I am unable

to determine what sort of mutual influence there must be for a judgment to result from the

presence of several subjective representations” (WLIII, § 291). He then explains that a

judgment is not merely a sum of representations, but a “certain efficacious combination of

representations”. Bolzano calls the degree of this efficacy the degree of confidence

(Zuversicht) with which we make a judgment or the degree of certainty with which we

hold something to be true, WLIII, § 293. Perhaps he means that we understand a statement

when we grasp its components with a sufficiently strong degree of confidence, that is, our

conviction is the glue that holds the components of a judgment in a certain order: when we

hold a statement to be true, we provide an efficacious combination of its components. But

that claim is insufficient for explaining the unity of judgment.

*248

Leibniz’s Monadology by Dr. Robert Zimmermann, Vienna 1847, Baumüller and

Seidel, pp. 172ff.


2. But there is no way we can grant that categories without intuitionsare empty and blind and have no objects.249 Whether a representation hasobjects does not depend at all on whether we know such objects or knowthem through intuition. But frequently we can know that a representationhas an object even though we have not cognized this representationthrough an intuition.250 [86] Thus we know that the concept of a soriteswith a thousand members has objectuality, although we do not have anintuition standing under it. – At several places in the Critique (e.g., p.240), we read that the categories or pure concepts of the understandingdo not represent any objects without all the conditions of sensibility,because the conditions of objective reality are missing.251 Nothing is

249

Here Přihonský got it wrong. Kant says: “Without sensibility no object would be given

to us, and without understanding none would be thought. Thoughts without content are

empty, intuitions without concepts are blind.” (KrV, B75; cf. also my note 241, above.)250

In Kant’s view, we cognize an object of experience or a phaenomenon if and only if a

concept is applied to a corresponding intuition, by means of a schema; cf. KrV, B75,

B181. In Bolzano’s view, intuitions always refer to objects whether or not they fall under

a concept, yet he does not say that they are cognitions; cf. WLI, § 72; WLIII, § 300;

MM4, pp. 139-140. Concepts, however, may or may not have a referent; cf. WLIII, § 271.

A cognition is a true judgment; cf. WLI, §§ 36-37, also WLI, § 26. Bolzano criticizes the

Kantian identification of judging and knowing: for Kant, a judgment is the mediate

cognition of an object, or an action by means of which given Vorstellungen become

cognitions of an object. Cf. KrV, B93. The upshot is that every judgment is a cognition

and every cognition is a judgment, and absurdly, we can have false cognitions. Kant might

reply that a cognition is, first of all, an objective representation with consciousness, and

there are two such cognitions, intuitions and concepts; cf. KrV, B351, B377. So, for him,

cognition is not situated at the level of judgments, unlike knowledge.251

At KrV, B234, the passage which roughly corresponds to Přihonský’s reference

(Critique, p. 240) deals with the category of causality, which is the law or necessary

connection between appearances underlying our experience. For Kant, “objective reality”

means just “to refer to an object and have in that object its signification and meaning”

(KrV, B194), i.e., what Bolzano means by “objectuality”. Perhaps Přihonský intends

Kant’s claim that the objective validity of the categories or of their relation to an object

rests not only on intuition but on empirical intuition (for even a priori intuitions have no

objective validity without an empirical intuition) or data for possible experience. Objects

can only be given to a concept in intuition, and concepts without such data are empty, that

is, they have no significance; cf. KrV, B185, B298-299. On the other hand, “the objective

validity of the categories, as a priori concepts, rests on the fact that through them alone is

experience possible (as far as the form of thinking is concerned). For they are then related

necessarily and a priori to objects of experience, since it is only through them that any

object of experience can be thought at all.” (KrV, B126). So, both concepts and intuitions

are necessary for our cognition of objects-besides, the object is a construction for Kant,

since the Gegenstand (of experience) given in intuition (that is, the appearance) only

becomes an object (for cognition) under the condition of the unity of apperception. “An

object, however, is that in the concept of which the manifold of a given intuition is

united.” (B137). Hence “objective reality” is not the same for Kant as for Přihonský.

New Anti-Kant 137

more wrong than this thought. For by leaving out certain conditions

(which surely can only mean certain closer determinations) from aconcept, it becomes wider, and if the narrower concept, to which thesedeterminations are still sticking, represents certain objects, the wider

concept must do so even more.252

3. The accord between experience and the concepts of its objects isbased merely on the fact that all experiential cognitions are nothing butinferences which we deduce from pure conceptual truths by means ofcertain minor propositions of the form: I have the intuitions x, y, z, . . .How can such a derivation lead to a conclusion that contradicts a pureconceptual truth? –

4. The claim that “all our categories can only be applied validly toobjects of a possible experience” presupposes the concept ofpossibility:253 so before [87] we can understand and use them, we need toknow what we are supposed to understand by the objects of a possibleexperience.254 Now, Kant considers the concept of possibility asbelonging to the categories. So in order to know to which objects thecategory or the concept of possibility is extensible, that is, what ispossible or not, we should already know what is possible and thusunderstand how to apply this concept absolutely (i.e., without limitationto the mere sphere of possibility). If the definition of the possible is that

which does not contradict any conceptual truth, which is the only onethat seems right to us, then assessing what is possible or not alreadypresupposes the cognition of, maybe not all but a great many, pure

252

Přihonský refers to the problem of confusing the components of a representation with

the attributes of its object (NAK, p. 8), but the Port Royal rule, according to which the

content and the extension of a representation stand in an inverse relation (cf. note 10),

does not apply here.253

Cf. KrV, B126, B161, B185, B298-299; and my notes 220 and 251, above.254

Přihonský takes up Bolzano’s view that conceptual truths, such as “A is actual”or “A is

possible”, ground all other propositions. Přihonský follows Bolzano who follows Aristotle

(De Interpretatione, 9.19b) by claiming “possibly p” is deducible from “possibly not p”

and by defining the possible as that whose contradiction does not follow from a

conceptual truth; cf. WLI, § 119.2.c; cf. also WLII, § 182, and my note 120. It follows

from this that our experience of an object A is possible if and only if the proposition “A is

possible” is a conceptual truth, that is, if and only if the object A which falls under the

concept [A] is not impossible. But surely Bolzano would reject this consequence, since he

claims that we can have immediate perceptual (nonlogical) knowledge; cf. § 300.12 and

my note 116, above. As for Kant, when he says that “the categories are the conditions of

the possibility of experience and therefore valid a priori for all the objects of experience”

(KrV, B161), he means (i) that cognition is possible only of the objects of possible

experience and (ii) that experience, as a synthetic connection of appearances, is the

product of an accord between the senses (intuitions) and the understanding (the

categories), P, § 22; KrV, B126.


conceptual truths, especially synthetic ones. Analytic truths or evenidentical truths cannot be considered here. So the assertion that thecategories are applicable only within the domain of possible experienceis really supposed to mean that pure conceptual truths can only refer toobjects which, when we consider them as perceived, do not contradictany pure conceptual truth.255 Therefore, to experience the limits of theapplicability of pure conceptual truths, we must first apply themunlimitedly!! – [88] In the second book of the Analytic, called theAnalytic of Principles, Kant begins by asserting “that in all subsumptionsof an object under a concept, the representation of the former must behomogenous with the latter; that is, the concept must contain that whichis represented in the object that is to be subsumed under it.”256 “Now”, hecontinues, “pure categories and empirical intuitions are entirelyunhomogenous. So there must be a third thing, which is homogenouswith the category on the one hand and the appearance on the other. Thismediating representation must be intellectual on the one hand andsensible on the other. Such a representation is the transcendental

schema” (p. 158).257 “A transcendental time-determination ishomogenous with the category on the one hand, because it rests on a rulea priori, and on the other hand it is sensible. Hence an application of thecategory to appearances becomes possible—by means of thetranscendental time-determination which is therefore the schema of theformer” (ibid.).258 “This schema is not an image but that which firstmakes images possible; it is only the representation of the generalprocedure of the imagination for providing a concept with its image.”

255

Cf. Bolzano’s objections to his view of Kant’s doctrine that we can only make synthetic

judgments about objects that we can perceive (i.e., objects of a possible experience),

WLIII, § 315.5-6. In Bolzano’s view, we can make true synthetic judgments about simple

substances, such as “God is immutable” or “no simple substance can perish in time”; thus

Kant wrongly indicts metaphysical judgments.256

Cf. KrV, B176, entitled: “On the schematism of the pure concepts of the

understanding.” Přihonský already criticizes Kant’s doctrine of the schematism above; cf.

pp. 69-72 and also my notes 204-208.257

Cf. KrV, B177. Přihonský summarises this passage. The schema must be homogenous

with both the category and the appearance so that it “makes possible the application of the

former to the latter” (ibid.).258

Cf. KrV, B177-178. Přihonský leaves out some of Kant’s statements, such as the

following: “Time, as the formal condition of the manifold of inner sense, thus of the

connection of all representations, contains an a priori manifold in pure intuition.”, which

explains why the schema is a transcendental time-determination. In addition, the last

sentence should read: “Hence an application of the category to appearances becomes

possible by means of the transcendental time-determination which, as the schema of the

concept of the understanding, mediates the subsumption of the latter under the former.”

New Anti-Kant 139

(p. 159)259 “The schema of all objects as such is time; the schema of allmagnitudes is number, the representation of a successive addition of oneunit to another; the schema of reality is existence at a determinate time;the schema of necessity is existence at all times; and so on.” (pp. 161f.)260

[89] Objections. We frankly admit that we consider this doctrine of theschematism of the pure concepts of the understanding to be utterlywrong, and we believe that our judgment is partly confirmed by similarjudgments, even by those of several friends of critical philosophy.

1. We know nothing at all about a homogeneity occurring between therepresentation of an object and the concept under which we want tosubsume it. Kant himself answers the question of what the homogeneityconsists in, only by saying that “the concept must contain that which isrepresented in the object”.261 – How could this be possible? The concept

259

Cf. KrV, B179-180: “The schema is in itself always only a product of the imagination;

but since the synthesis of the latter has as its aim no individual intuition but only the unity

in the determination of sensibility, the schema is to be distinguished from an image.” Kant

gives two criteria for distinguishing between a schema and an image: (i) An image is a

particular representation which exists physically and cannot be adequate to a concept:

“thus, if I place five points in a row, . . . this is an image of the number five.” A schema is

a general representation which can exist nowhere but in thought and which can attain the

universality of a concept: “if I only think a number in general [. . .], this thinking is more

the representation of a method for representing a multitude (e.g., a thousand) in accord

with a certain concept [. . .].” (ibid.). Likewise, we can have a schema which is valid for

all triangles, but “no image could ever be adequate to the concept of a triangle in general.”

(ii) A genetic difference: the schema and image originate in two different types of

imagination: a schema is the product of a priori (or productive) imagination (hence Kant

also calls it a pure image), whereas an (empirical) image is produced by empirical (or

reproductive) imagination; cf. KrV, B181.260

Kant writes that the schema “of all magnitudes (quantorum) for outer sense is space; for

all objects of the senses in general it is time. The pure schema of magnitude (quantitatis),however, as a concept of the understanding, is number,” (KrV, B182). Kant distinguishes

between extensive magnitudes (quanta) which are determinate and cognizable, and

indeterminate magnitudes (quantitatis) which are merely forms of representation. This

distinction corresponds to the distinction between formal intuitions and forms of intuition,

(cf. B35-36, B55-56, B137-178, B160; P, § 38). In addition, the schema of necessity is

“the existence of an object at all times” (B184). Bolzano distinguishes between quanta or

extensive magnitudes and quantitas or abstract magnitudes (Kant’s forms of

representation) as follows: quanta are individuals belonging to a species, whereas

quantitas is the particular property of a thing by virtue of which it becomes the magnitude

of a determined species, i.e., a quantum; cf. Sebestik (1992, op. cit., p. 37).261

Cf. KrV, B176. Here Přihonský objects against the confusion between the components

of a representation with the attributes of its object (NAK, p. 8). Cf. my note 106 on

Bolzano’s criticism of the Port Royal rule, according to which the content and the

extension of a representation stand in an inverse relation. Another Kant-critic who

objected against the homogeny claim is Carl Stumpf (1891), who argues that we cannot

justifiably claim that the schema is an intermediary between the categories and


something, for example, is completely simple and yet extends to allobjects. But according to Kant’s assertion, this concept should be themost complex of all concepts (as Lambert mistakenly believed) and since,for example, even this rosebush can be subsumed under it, the conceptsomething contains everything I think or can think of this rosebush!!

2. Nor can we accept that every pure concept of the understanding andevery intuition make up a pair of such unhomogenous things that thelatter can never be subsumed under the former. For is not the concept ofan [90] intuition itself a pure concept of the understanding, under whichevery intuition stands necessarily?

3. In addition, if the category and the intuition are so unhomogenousthat the latter does not fall under the former, how can there be a thirdthing mediating this subsumption? We think that if representation x canbe subsumed under representation M, and this latter can be subsumedunder representation A, logic teaches that representation x can also beimmediately subsumed under representation A.262

4. We think that it is a mistake to consider the transcendental time-determinations as intellectual as well as sensible, insofar as they containthe representation of time (an intuition). We have already shown that therepresentation of time does not belong to the genre of intuitions at all andshould therefore only be considered as a concept.

5. But still, as we also have sufficiently shown, the term schema canmean nothing other than the genetic definition of some concept. So, tomediate between two objects through a schema means to bring this aboutthrough the assistance of a concept or to use a concept for subsuming anintuition under a concept (cf. p. 71, above).263

6. Finally, we cannot accept that the concept of a quantum or amagnitude and the concept of number are supposed to contain theconcept of time. [91] The Theory of Science, § 87, teaches us the truecomponents of these two concepts magnitude and time, respectively.264

appearances simply because it lies between the two. In addition, the reason that time-

determination shares properties of both the categories and intuition is insufficient for

subsuming appearances under concepts. He ironically remarks that such a mediation

would be like using the representation of a glass for subsuming a leaf under the concept of

air, since it shares the property of green with the leaf and the property of transparency

with air; cf. Psychologie und Erkenntnistheorie, p. 11.262

Přihonský refers to the law of transitivity.263

Cf. NAK, pp. 71-72, and my note 208, above.264

Cf. my note 260 on Kant’s (and Bolzano’s) respective distinctions between cognizable

or extensive magnitudes which are objects of intuition (quanta) and the general concept of

magnitude (quantitas). Here we should note that for Kant mathematical concepts are first

given through definition and constructed through intuition; cf. KrV, B759. Likewise, on

New Anti-Kant 141

Kant establishes the principle of contradiction as the supreme

principle of all analytic judgments (pp. 166f.).265 But the principle ofcontradiction is not a principle in the proper sense of the word or aproposition from which all analytic judgments are deducible asconsequences from their ground, nor can it be. Analytic judgmentscannot be derived from this proposition; rather they are merely specificexamples of it. Even Kant does not appear to consider the matter very

his view, mathematical cognition proceeds from the construction of concepts (cf. KrV,

B763) and, as a result of this view, magnitudes are determined as constructions in

intuition. Bolzano objects against Kant that mathematics deals mainly with quanta and

that this is why mathematical cognitions deal with quanta, not because mathematical

cognition and the science of mathematics differ from the other sciences, (1814), p. 866,

Grössenlehre, § 2. So on the one hand, Bolzano criticizes Kant for distinguishing between

mathematical and philosophical cognition, but on the other, he criticizes what he calls the

Kantian confusion between logical and mathematical quantity and quality; cf. (1814), p.

867. In addition, he says that the concept of time has no place in general mathematics; cf.

Pure Analytic proof. (1817), preface II. Be that as it may, perhaps we should take a closer

look at what Kant had in mind:

(i) Kant’s category of quality differs from more contemporary conceptions, since it also

includes intensive quantities or degrees, which differ from extensive quantities in that our

apprehension of them is instantaneous and not successive. Cf. KrV, B210.

(ii) Number, or the successive grasp of parts compounding a whole, is the schema of

quantitative categories. These are extensive magnitudes, such as volumes or durations,

which we grasp by constructing them in intuition. According to Kant, a magnitude is

extensive “when the representation of the parts makes possible the representation of the

whole” (KrV, B202). Extensive magnitudes consist of an additive number of similar parts,

and in order to grasp them, we add up their parts to compound a whole through the

introduction of limitations. For example, we obtain the general concept of space by

segmenting or cutting out parts of space from the singular intuition of space (cf. KrV,

B39-40, and my notes 164 and 190, above). We can reconstruct the following Kantian

argument for this claim: Kant’s principle of the axioms of intuition (the principles

corresponding to the quantitative categories) is that all intuitions are extensive quantities

(extensive Grössen) (KrV, B202). This principle underwrites the construction of

quantitative concepts through intuition. For example, the concept [straight line] is

constructed in the intuition of space by subsumption under the concept [extensive

quantity]. If we say of a single line between two points A and B that it is a straight line,

we do not have an immediate intuition of a line but grasp it as the construction of a

succession of parts which are compounded into a whole. Thus Kant writes: “I cannot

represent to myself a line, however small, without drawing it in thought, that is,

generating from a point all its parts one after the other.” (KrV, B202) And we ascribe to

this line a quality which is, in fact, a quantity, since an extensive quantity is extensible

without extending its quality. We can formulate the following definition: an intuition is an

extensive quantity if and only if it is an additive quantity of similar elements. A spatial

intuition, such as a line, is an extensive quantity that can be increased or decreased

without increasing or decreasing its quality.265

Cf. KrV, B189. Cf. also my note 141, second paragraph, on Kant’s use of the principle

of contradiction as a linguistic rule for meaningful language use.


strictly, for in his explanation he refers only to the contradiction weshould produce by denying that a predicate belongs to a determinatesubject when it is contained in the latter as a component; and of coursewe do not object to that. However, Kant should not have believed, as heactually did, that this explains the origin of analytic judgments andentitles us to make them. For it does not follow from making a judgmentwhich does not go beyond the concept of its subject-representation thatthis judgment must necessarily be true. [92] Even an analytic judgment islacking truth if the subject-representation is not objectual.266 “Areas of acircle are geometrical figures” is an analytic judgment and incontestablytrue. But if there were no areas of a circle, it would not be true that theywere geometrical figures.

This is how Kant expresses the supreme principle of all synthetic

judgments: “every object stands under the necessary conditions for thesynthetic unity of the manifold of intuition in a possible experience.”(p. 171).267

Thus in his view, “synthetic judgments are possible a priori if werelate the formal conditions of a priori intuition, the synthesis ofimagination, and the necessary unity of this synthesis in a transcendentalapperception to a possible experiential cognition as such” (ibid.).268 Whatwe think of Kant’s assertions is sufficiently clear from our precedingpresentation, so we will be brief. Kant’s supreme principle of syntheticjudgments says nothing other than that every object of our cognition isconditioned by an intuition we connect with it, or that we only attain acognition and can only make a true judgment if an intuition can berelated to the subject representing the object of the judgment. But wehave already demonstrated that this assumption does not even show ushow empirical [93] judgments originate, let alone provide an adequatedefinition of synthetic a priori judgments. According to Kant, thepossibility of the latter also rests on intuitions, though not on empiricalones but on the pure or a priori intuitions which are the representations

266

Přihonský opposes the objectuality constraint (cf. WLI, § 148, and my notes 141 and

142, above) to Kant’s criterion of inclusion (of the predicate in the subject), for

distinguishing between analytic and synthetic judgments; cf. KrV, intro. IV, B. But

concerning the necessary truth of analytic judgments, Kant only advances the criterion of

cognizability: a judgment is analytic if and only if its truth is cognizable sufficiently by

reference to the principle of contradiction; cf. KrV, B190.267

Cf. KrV, B197.268

Cf. KrV, B197. Kant ends this statement with his famous claim that “the conditions for

the possibility of experience as such are at the same time conditions for the possibility ofobjects of experience and therefore have objective validity in a synthetic a priorijudgment.”

New Anti-Kant 143

of time and space. We have seen above that this doctrine is false and thatKant has not managed to explain the origin of even a single a priori

judgment. The attempt to achieve this through the so-called schema ofthe concept and the synthesis of both through the pure faculty of

imagination should be considered a complete failure, just as talking abouta synthetic unity in apperception through the representation: I think,should be seen as entirely unjustified.269

In the same way he classifies the categories, Kant then wants toclassify the synthetic principles and dedicates a special passage to thissystematic representation (pp. 111f.).270 “The table of categories”, hesays (p. 173), “gives us entirely natural direction for the table ofprinciples, since these principles are nothing other than rules for theobjective use of the categories.271 All the principles of pure understandingare, accordingly, 1. axioms of intuition, 2. anticipations of perception, 3.Analogies of experience, and 4. postulates of empirical thinking as such”;and they read, as follows:

[94] 1. The principle of the axioms of intuition is “all intuitions are

extensive magnitudes” (p. 174).272

2. “The principle of the anticipations of perception is in allappearances the real, which is an object of sensation, has an intensivemagnitude, i.e., a degree” (p. 178).273

3. “The principle of the analogy274 of experience is experience ispossible only through the representation of a necessary connection ofperceptions” (p. 186).275 These are

a) “[the] principle of the persistence of substance: in all change ofappearance substance persists, and its quantum in nature is neitherincreased nor diminished” (p. 190).276

b) “[the] principle of temporal succession, according to the law ofcausality: all changes occur according to the law of the connectionof cause and effect” (p. 195).277

269

Přihonský’s acerbic remarks leave an open question: did he merely misconstrue Kant or

did he choose to belittle what he failed to understand?270

Cf. KrV, B198.271

Cf. KrV, B200.272

Cf. KrV, B202. Here and in what follows, Přihonský quotes from the second edition of

the Critique.273

Cf. KrV, B208.274

Kant says “analogies”, for there are several.275

Cf. KrV, B218.276

Cf. KrV, B224.277

Cf. KrV, B232.


c) “[the] principle of simultaneity, according to the law of interaction orcommunity: all substances, insofar as they can be perceived inspace as simultaneous, are in thoroughgoing interaction”(p. 211).278

[95] 4. “[The] postulates of empirical thinking as such:”

a) Whatever agrees with the formal conditions of experience (accordingto intuition and concepts) is possible.

b) That which is connected with the material conditions of experience (ofsensation) is real.

c) That whose connection with the real is determined according togeneral conditions of experience is (exists) necessarily (p. 217).279

We cannot communicate the proofs and explanations Kant adds to theseprinciples, not even in an excerpt, but we think that the followingremarks will allow the reader to judge the value of the principles set upby our philosopher.

1. We also accept that every subjective intuition and everyrepresentation as such has a magnitude with regard to itspersistence as well as its vivacity. But that is not at all what Kantasserts in his axioms of intuition.280 There he only wants to saythat each time span and each distance in time have a magnitude,even an extensive magnitude, which is incontestable, of course.But we cannot accept that these magnitudes are intuitions.281 Forthe duration of a second or the distance between my twofingertips is not and cannot be an [96] object of my intuition, anymore than the duration of a millennium or the distance betweenthis fixed star and the moon just gliding past it can be the objectof my intuition. What we intuit are merely certain colours,sounds, tones, and so on, from the presence of which we infer a

278

Cf. KrV, B256.279

Cf. KrV, B265-266.280

Kant’s axioms of intuition are in the plural, since they are a group of mathematical

principles; cf. KrV, B202.281

Kant does not say that all magnitudes are intuitions but that appearances can become

objects of intuition if and only if they possess an extensive magnitude. So extensive

magnitude is a condition of experience and of objects of experience. Thus all intuitions

are extensive magnitudes; cf. KrV, B202. Kant distinguishes between (1) magnitude

(quantitatis), or a pure concept of the understanding which needs a schema (number) in

order to be applied to objects of intuition, and (2) magnitudes (quanta) which are

constructed through intuition. Of the latter kind are extensive magnitudes, which are

grasped when we construct them by successively adding up their parts; cf. KrV, B182, and

my notes 258 and 262.

New Anti-Kant 145

certain temporal flow282 and the existence of certain objects thatcause these representations in us and stand in certain spatialrelations. These inferences are often based on many intermediarypropositions.*283, 284

282 Verfliessen.

*283

Here we recall with approval Dr. Theodor Waitz’s presentation of this object in his

Foundation of Psychology (Hamburg and Gotha, 1846, Perthes, pp. 87f.): “we must deny

that the soul originally and immediately obtains the representation of a surface through

visual sensation (Gesichtsempfindung)” (p. 87). “Surface-representations as such are

neither immediately given through perception nor through the representation of

continuity” (p. 89). “The position and magnitude of objects are not immediately perceived,

but rather judged and estimated” (pp. 90f.).284

Přihonský objects against Kant that we cannot immediately perceive an extension or a

duration but only colours and sounds, from which we infer the existence of certain objects

which stand in spatiotemporal relations. Waitz also claims that the extension and

magnitude of objects are judged rather than perceived, and Přihonský uses this claim as

added support for rejecting Kant’s view that extension and figure are objective properties

of the objects of experience, or phaenomena, whereas colours, sounds, or smells are

nothing but subjective properties or mental changes which cannot be represented without

extension; cf. KrV, B34-35; B45; B209. Here Kant seems close to Locke and Berkeley,

who hold that we cannot conceive or perceive a sensory quality without extension.

However, the Irish philosopher asserts not only that no quality is perceived without an

extension but also that no extension is perceived without a quality, and he claims, unlike

Kant, that since we have distinct sensory ideas of sight and touch, there are also distinct

and different sensory spaces. George Berkeley (1709), A New Theory of Vision, §§ 119,

121, 127 (London: Everyman, 1963), p. 71. The young Bolzano (1810, in the Appendix onthe Kantian Theory of the Construction of Concepts through Intuitions, § 11) argues that

objects which have a shape, also have sensory qualities that fill it, and therefore no

extension is perceived without sensory qualities. So if there is a perceived object such as

x, which is extended, then x has a sensory quality, and what is given in perception is not a

pure form of intuition or empty space as Kant affirms (cf. KrV, B35) but a space filled

with sensory qualities. Although we may separate extension from its sensory qualities

through an act of abstraction, they are inseparable in perception. But while Bolzano

claims that everything which is extended has a sensory quality, he does not claim that

every sensory quality is extended. Bolzano says that only external qualities are extended

and there are also unextended qualities such as internal colour sensations occurring, for

example, when the eye is pathologically affected. Cf. Appendix, § 11; also WLIII, § 286.1.

So Kant and Bolzano agree that there can be qualities without extension. Bolzano’s claim

corroborates his topology of spatial objects, according to which lines, figures, and planes

are collections of points (cf. Sebestik, 1992, op. cit., pp. 56-59), but in PU, § 55, Bolzano

claims that mental entities are spatiotemporal and in the Athanasia (sec. 4, pp. 124ff., and

my note 170, above), he claims that the soul is a spatiotemporal substance, which

contradicts the view that our internal sensations are punctual and unextended.

Anyway, Přihonský’s objections against Kant are based on the “Transcendental aesthetic”

rather than the “Principles of pure understanding”, where his purpose is threefold: (1) He

seeks to distinguish between sensations (the basic parts of our perceptual state or the

“matter” of perception) and quality (a category or basic principle of the understanding

which determines how judgments are applied to objects of intuition). So a sensation is a


2. The principle that every sensation has a certain (intensive)magnitude is entirely right, for we must certainly consider the faculties ofliving beings and thus also sensory faculties and their effects orsensations as intensive magnitudes.

3. We also have no objection to the three anticipations of perception,[97] although it seems to us that Kant has not sufficiently understood thatall causes and effects are always only simultaneous, since he asserts (e.g.,on p. 206) this simultaneity only of the main part of efficient causes.285

basic perceptual state; discrimination of a quality is dependent on concepts and categories;

and the combination is a discriminative cognition of something we see which is a part of

the process of coming to know objects; cf. KrV, B210. For Kant, a sensory quality is the

reality degree of a sensation, and this is what we can anticipate or know a priori about a

sensation: a sensation is empirical, but its phaenomenal reality is dependent on a prioriconcepts and principles (such as the Anticipations) and can therefore be determined apriori or anticipated, from its beginning, at point 0, to any degree of intensity. In other

words, we can present the process of generating the reality degree of a given sensation of

red from 0 to any required level. Cf. KrV, B208-210; B218. (2) Kant’s purpose underlying

this distinction is to justify his claim that it is possible to anticipate perceptual qualities

we have not (yet) sensed. He argues that we can perceive appearances if and only if they

possess an intensive magnitude or degree. The quality of appearances is determined by the

anticipations of perception which allow perception to take place, by “ascribing a degree to

all that is real in appearances” (KrV, B217). We can abstract from extension when we

conceive sensory qualities by presenting their degree. And it is because we can present

their degree that we can anticipate sensations we have never seen; cf. KrV, B218. What

we know a priori of a sensation is its reality-degree which is a quality or property of the

appearance of the sensation. The quality attaches to the appearance of the sensation, and it

characterizes our experience of a sensation. In other words, a sensory quality is a feature

or property we attribute to an object; it is that which we discriminate in something we see.

Sensory qualities characterize our experience of visual or auditory events; cf. Anita

Kasabova (2004) “Colour Sensations and Colour Qualities: Bolzano between Modern and

Contemporary Views”, 247-276, (3) Kant’s third purpose is to object to the empiricist

claim that sensory qualities are extended. For him, sensory qualities have no extension

and, in addition, it is possible to increase or decrease the degree of a sensory quality

without modifying its extensive quantity or surface: if we can conceive a quality redwithout having seen it, we can also conceive its degree of intensity without spatial

extension; cf. B210, B211. Therefore Kant admits extensive quantities without qualities

and qualities without extension: just as extensive magnitudes have no intensive degrees,

sensations have no extension; cf. KrV, B209, B215; and my note 264, above.285

Cf. KrV, B248. Kant says that “[t]he majority of efficient causes 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 the causality of its cause,

since if the cause had ceased to be an instant before then, the effect would never have

arisen”. But the principle of causal connection among appearances is their succession,

because the temporal order in which we perceive them is that of earlier and later, and this

order must be distinguished from the lapse of time between the causality of the cause and

its immediate effect which “can be vanishing (they can therefore be simultaneous), [. . .].

New Anti-Kant 147

But things really stand as follows: every effect that produces a finitecause consists in a change, and therefore only happens in a certain time.As long as it lasts (as long as something changes), the cause must also bepresent in us and be effective. But in common usage we often use theterm cause for that which is only a partial cause, that is, one of thoseobjects which can produce an effect only in combination with certainother causes.286 And it is certainly true of such partial causes that theyoften exist a long time before the real effect happens and sometimes lastlonger than it does. This is the case, for example, with stone and steel,which exist prior to the spark that is generally said to be their effect orproduct and that, consequently, are its cause. But basically steel andstone are only partial causes of the spark, for to cause the latter, it takesalso a force to beat them against each other, as well as air: in short, theirwhole and complete cause which only comes into existence at the sametime as the effect.

[98] But there is just as little ground for claiming that the effect lastslonger than the cause, since it stops as soon as the cause has stoppedacting. Here it is pointless to object that the building lasts longer than itsbuilder or that each body will move without stopping if it is set in motionby some force in precisely the direction and with precisely the velocity it

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 dynamic 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 from it.” (B248-249)286

In Bolzano’s view, the principle of causality says that “for every change we presuppose

a cause which has produced it” (WLIII, § 379), and it expresses an objective relation

between truths: Bolzano relates the two types of concepts, cause-effect and ground-

consequence, by deriving the concepts of cause and effect from the concepts of ground

and consequence. He claims that causal propositions are determined by a ground-

consequence relation (Abfolge) between other propositions, so that the causal proposition:

“x causes y” actually expresses the following ground-consequence relation: “the truth that

x exists is related to the truth that y exists as a (partial) ground is related to its (partial)

consequence” (WLII, § 168). Cf. also Bolzanos Wissenschaftslehre in einer Selbstanzeige,

p. 82, where he claims that his doctrine of objective relations between truths is a new way

of explaining causality. He also makes an almost Kantian claim concerning cognition: the

principle of causality is based on the ground-consequence relation, which also expresses

the principle of sufficient reason (Satz vom Grunde). This latter is a cognition of reason

(Vernunfterkenntnis) and not of the understanding because “it is the first expression of

waking reason always to ask for the ground” (WLIII, § 311.2). This claim is very similar

to Kant’s view that the causality principle is a principle of reason, though for different

reasons, and Bolzano explicitly rejects an account of conceptual truths or grounds as mere

cognitions; cf. WLII, § 211. But he points out that our cognition of this truth is the first

sign of awakening reason; cf. WLIII, § 311.2.


received at the last moment the force had an effect on it, unless itencounters an obstacle on its way, and so on. For in these examplessomething is considered as the effect of a cause without actually being aneffect but rather the product of many other factors. For example, thecontinuing existence of the building is not the effect of the builder but ofthe forces of the matter the building consists of. For a body to change itslocation or move, a velocity is necessary which lies in the body as themovement’s proper and immediate cause and a force which produces achange by acting on the body’s velocity. This velocity persists orincreases as long as this force has an effect. Once the force stops havingan effect, the change or velocity stops increasing. So the velocity a bodyreceives at the last moment in time remains unchanged, and because [99]

velocity is the proper cause of the change in location, it continues topersist and the body moves without stopping. – Rather than refuting ourprevious claim of the simultaneity of cause and effect, these examplesconfirm it. Besides, this theorem is not without importance for science:for example, when deciding the metaphysical question of whether theworld’s eternal existence can be considered an effect of God.287

4. Finally, we cannot avoid rejecting the three so-called postulates of

empirical thinking as entirely false. In our view, what is possible is onlythat which does not contradict a pure conceptual truth.288 We do notconsider a surface bounded by two straight lines (bilinium) possible butdeclare it to be impossible, because such a surface contradicts a pureconceptual truth and not an intuition (namely, the pure intuition ofspace), as Kant believes.289 Besides, we should not worry about becomingentangled in fantasies because of this assumption if we are sufficiently

287

Přihonský refers to Kant’s fourth or modal antinomy of pure reason or conflict of thetranscendental ideas, which deals with the question whether the world has a first cause or

not: the two following claims are opposed: “something belongs to the world that, either as

a part of it or as its cause, is an absolutely necessary being” and “there is absolutely no

necessary being existing anywhere, either in the world or outside the world, as its cause”;

cf. KrV, B480-481. Bolzano criticizes the antinomies and Kant’s denial that human reason

has the capacity to judge suprasensible objects, in WLIII, § 315.6-7, cf. also my note 255,

above.288

On Kant and Přihonský’s differing notions of possibility, cf. my note 254, above. In

addition, Kant’s postulates for empirical thinking are principles for employing the

category of modality: they form the “conditions of experience” which are also the

“conditions for the objects of experience”, and they govern the way empirical cognition is

related to the conditions of possible experience (cf. KrV, B266).289

Kant would reject Přihonský’s critique by advancing his method of constructing

mathematical concepts through intuitions; cf. on this method, my notes 128, 164, 204-207,

and 264, above.

New Anti-Kant 149

careful in making our judgments.290 – To presuppose that a concept isobjectual, it is obviously not enough to think that it contains nocontradictory marks, such as b and not b; nevertheless it is excessive toassert [100] that we can only ascertain the objectuality of this concept ifexperience teaches us to cognize an object standing under it.291 Thisconclusion is inferred from the false claim we have already refuted; thatwe are entitled to make a synthetic judgment if and only if we connectthe subject-representation with a pure or an empirical intuition and thepredicate which we want to ascribe to the subject.

This same book of the “Analytic” also contains the important partentitled “On the ground of the distinction of all objects as such inphaenomena and noumena” (pp. 236f).292

Here Kant repeatedly assures us “that we cannot cognize any objectother than those which can be intuited through sensibility, and they arecalled phaenomena.293 Noumenon is the concept of a thing which is notan object of sensible intuition, and it is merely problematic; we do notknow whether there is an object corresponding to it, since objects cannotbe given to us in any other way than through intuition and our intuitionsare only sensible, not intellectual; and we do not even know whether

290

Kant restricts the postulates for empirical thinking to a “merely empirical use”, since

they not only express the form of thinking but concern things; cf. KrV, B267; cf. also my

notes 255 and 287, above. Kant restricts these principles to a regulative use (they show

how the unity of experience may arise from perception or how experience is organized [cf.

B223]) but does not allow their constitutive use (unlike the axioms of intuition and the

anticipations of perception, they do not govern the existence or the constitution of

appearances; cf. KrV, B221).291

Kant and Přihonský have different notions of experience: for Přihonský, experience is

that which is given through the senses-that is, it is synonymous with perception. For Kant,

however, experience is “cognition by means of connected perceptions”, and “the

categories are the conditions of the possibility of experience” (KrV, B161). So experience

is constructed by means of the categories, since it is a synthetic connection of appearances

(or perceptions)” (P, § 22); cf. also my note 127, above. On the relation of standing underwhich holds between objects and concepts, cf. note 14.292

Cf. KrV, B294. In Kant’s view, the division of objects into phaenomena and noumena is

merely problematic: i.e., it can only be permitted in a negative sense, since the noumena is

merely a boundary concept, and we are unable “to posit anything positive outside the

domain of [sensibility]” (KrV, B311). Problematic judgments correspond to the modal

category of possibility (i.e., when an assertion or denial is considered as merely possible),

and Kant distinguishes them from apodictic and assertoric judgments; cf. B100. Cf. also

Dubislav (1931, pp. 220-221), who comments that, according to Bolzano, Kant’s

distinction belongs to the chimeras.293

Kant also distinguishes between phaenomena as appearances or objects which “are

thought in accordance with the unity of the categories” and noumena as “merely objects of

the understanding that, nevertheless, can be given to an intuition, although not to sensible

intuition”; cf. KrV, B306.


there are such intuitions.” (p. 247)294 [101] Nonetheless Kant groundsseveral important principles on this problematic distinction, such as theprinciple of human freedom, as we shall see further on!

This main part also contains an “Appendix on the amphiboly ofconcepts of reflection, through the confusion of the empirical use of theunderstanding with the transcendental” (pp. 254f.).295 – Kant says herethat a preliminary reflection or comparison of the concepts belonging tothem is necessary for all judgments.296 This comparison is merely logical

if it concerns the relation between concepts as such or transcendental if itconcerns the kind of cognition to which the concepts belong. Therelations in which the concepts to be compared can belong to each otherare those of identity and difference,297 of agreement and opposition, ofthe determinable (matter) and determination (form). If the concepts arecompared merely logically, we run the risk of establishing a relationlacking transcendental validity, and this is the relation Kant calls anamphiboly.298 He sets up the following rules on this issue:

294

Přihonský’s account of Kant’s noumena is not quite correct, for this passage reads:

“The concept of a noumenon, i.e., of a thing that is not to be thought of as an object of the

senses but rather as a thing in itself (solely through a pure understanding) is not

contradictory; for one cannot assert of sensibility that it is the only possible kind of

intuition. Further, this concept is necessary in order not to extend sensible intuition to

things-in-themselves, and thus to limit the objective validity of sensible cognition [. . .];

however, we have no insight into the possibility of such noumena, and the domain outside

of the sphere of appearances is empty (for us); i.e., we have an understanding that extends

farther than sensibility problematically, but no intuition, indeed not even the concept of a

possible intuition, [. . .] about which the understanding could be employed assertorically.

The concept of a noumenon is therefore merely a boundary concept, to limit the

pretension of sensibility, and therefore only of negative use.” Cf. KrV, B310-311. Cf. also

A289/345, where Kant explains that the illegitimate (or positive) use of noumena is to

ascribe objectuality to them-either by making a transcendent use of the categories and

treating them as if they were either noumena or applicable to noumena (cf. P, §§ 33, 45),

or by transcending the limits of experience and inappropriately applying them to

suprasensible objects (cf. P, § 45). Přihonský also criticizes the noumena in the Appendix,

p. 232; cf. my note 606.295

Cf. KrV, B316.296

Kant says that “all judgments, indeed all comparisons, require a reflection, i.e., a

distinction of the cognitive power to which the given concepts belong.” (KrV, B317)

Přihonský uses Erkenntnisart instead of Erkenntniskraft, which slightly modifies the

meaning of the sentence.297

I follow Guyer & Wood and Kemp Smith in using “identity” for translating Einerleiheit(KrV, B317). Pluhar translates “sameness”.298

The amphiboly of concepts of reflection is a fallacy which occurs either when the

empirical use of the understanding is confused with the transcendental use or when a

concept of reflection is mistakenly applied to sensibility or the understanding; cf. KrV,

B317-318.

New Anti-Kant 151

1. Identity and difference.299 An object with the same innerdeterminations is always the same if it is an object of the pureunderstanding (noumenon). But if it is an appearance (phaenomenon),[102] the difference of the places of these objects is, at the same time,still an adequate ground for assuming that there are several objects.300

Leibniz’s principium identitatis indiscernibilium (that there are no twothings completely equal to each other) applies to objects of the

understanding but not to appearances, although he wants it also appliedto them.

2. Agreement and opposition.301 As merely logical relations, realitiesof the pure understanding cannot be in opposition to each other, althoughthey can be in opposition as appearances. – Hence Leibniz’s assertionthat evil is a mere limitation again fails to apply to appearances.

3. The internal and the external.302 An object of the pureunderstanding must have inner determinations that bear no relation toanything that is different from it. However, the inner determinations ofan appearance are nothing but relations. This induced Leibniz to assumethat monads are internally distinct substances because of their particularrepresentations, even though their existence is unprovable preciselybecause they are not appearances.

4. Matter and form.303 In a concept of pure understanding, matterprecedes form; hence Leibniz grounds external relations on theassumption that there are things gifted with powers of representation(monads). But if it is only sensible intuitions in which we determine allobjects as appearances, [103] then the form of intuition or space and timeprecedes all matter (the sensations).

Objections. 1. We cannot accept the claim that every judgment ispreceded by a reflection which compares the representations belonging toit, for such a comparison or cognition of identity or difference, and so on,is itself a judgment and would thus presuppose a comparison or ajudgment and so forth, to infinity.

2. Anyway, we do not think that the four pairs of concepts Kantcompiled under the name “concepts of reflection” are really a whole ofthe kind he claims. Kant deduces these concepts from the familiar fourmoments under which he subordinates all forms of judgment. Since,

299

Cf. KrV, B319.300

Kant writes: “an adequate ground for the numerical difference of the object (of the

senses) itself.” (KrV, B319)301

Cf. KrV, B320.302

Cf. KrV, B321.303

Cf. KrV, B322.


however, this division of judgment appears wrong to us, we cannotapprove the division of the concepts of reflection. These are supposed torepresent mere relations between concepts of representations. However,we certainly do not express the relation between two representations byclaiming that two representations of one and the same real object arecaused by it; that is, by claiming that what we presently perceive is thesame as that which we have perceived at another time. Instead, therepresentation we are interested in is of a relation in which tworepresentations stand to a real object we want to represent [104] when weapply the two concepts of identity and difference to them in the above-mentioned way. However, the concepts of agreement and opposition firstand immediately refer to propositions and not to representations: we canassert only of propositions but not of representations that they agree with,or are opposed to, each other. We can say of certain representations, A

and B, that they are opposed to each other only insofar as the twopropositions “x is A” and “x is B” are in opposition or cannot both be trueat the same time. The concepts external and internal express two kinds ofproperties which are not properties as such but which can only be objectsof representations. Finally, form and matter are concepts equallyapplicable to representations and hundreds of other objects. For do wenot say that gold, silver, marble, and wood, and so on, are the matter ofwhich some work of art is manufactured? However, we usually call theway in which those masses are shaped into a whole, the form of a workof art.304 – Someone could argue in Kant’s sense that “these concepts mayrefer to various objects: as a whole, they belong to logic only if they areapplied to representations and judgments. The understanding produces auniversal or particular judgment, insofar as the relation in which the[105] objects of the subject- and predicate-representation stand to eachother,305 appears as a relation of identity and difference. Theunderstanding affirms or denies to what extent it maintains that thesubject and predicate agree or are opposed to each other. Theunderstanding judges categorically or hypothetically or disjunctively towhat extent the relation between subject and predicate is an internal orexternal one. Finally, the understanding expresses a problematic orassertoric or apodictic judgment, depending on whether it believes that itshould discover the ground of the connection between the tworepresentations in form or in matter.” –

304

Přihonský assumes that matter is a physical substance, rather than a category contrasted

with form, which is what Kant has in mind.305

Přihonský writes Subject- und Prädicatvorstellung, but he probably intends

Vorstellungen (of subject and predicate, respectively).

New Anti-Kant 153

We object to this: even if we could endorse the way of explaining theorigin of universal and particular judgments, as well as affirmative andnegative judgments (although it sounds a bit odd to say that theunderstanding produces a universal or a particular judgment through theinsight that the subject- and predicate-representations either stand in arelation of identity or of difference), we cannot accept that the distinctionbetween categorical, hypothetical, and disjunctive judgments consists ina relation between subject and predicate which is an internal one in thefirst and an external one in the latter two. Do we not use the conceptsinternal and external truly arbitrarily by claiming that the proposition: theplanet Mercury is nearest to the sun, expresses [106] an internal relation,whereas the propositions: if Mercury is nearest to the sun, it receives agreater intensity of light than the earth and Mercury either has a strongeror a weaker light than the earth—express an external relation? Even lessacceptable is the assertion that we make a problematic judgment whenthe relation between subject and predicate is grounded in the form,whereas assertoric and apodictic judgments are grounded in the matter.We may judge only problematically that “the moon has no rivers”, yetthis judgment is not grounded in the form alone but also in the matter.306

3. Although it has not yet been strictly proved, Leibniz’s principle ofthe identity of indiscernibles has an infinitely high degree of probabilitybecause of the fact that two substances are completely identical if and

306

It seems that Přihonský refers back to his earlier discussion (NAK, p. 81) of Kant’s

table of judgments and the (Bolzanian) objection (WLII, § 190.1) that all propositions

have a subject and a predicate, rather than categorical judgments alone; cf. my note 229,

above. But in the amphiboly, Kant is concerned with the fallacious use of concepts of

reflection [matter] and [form] when we fail to distinguish between their conceptual and

intuitive applications. The matter of a judgment concerns what is determinable (the

content), whereas the form of a judgment is its determination; cf. KrV, B322. Kant on the

one hand accuses Leibniz of applying concepts of reflection to objects of sensibility that

belong to the understanding and assimilating appearances to concepts, and on the other

hand he accuses Locke of assimilating concepts to sensory impressions by abstracting the

first from the second; cf. KrV, B326-327.

In addition, at the end of the amphiboly Kant introduces the important distinction between

the concepts [something] and [nothing], as well as the distinction between the logical

concept [nothing] (resulting from negation, according to the principle of noncontradiction)

and [nothing] resulting from a real opposition between two forces which cancel each other

out. The first [nothing] cannot be thought because it is opposed to possibility, whereas the

second [nothing] can be thought as the state of rest between two opposed forces. Both are

empty concepts, since they have no object or “example from experience” (B347). Kant

then discusses the question of whether the object of a judgment is something or nothing;

cf. KrV, B348-349. Přihonský fails to mention this passage and comments only on the

(formal) logical aspect of Kant’s text, and as a result his critique is a little one-sided.


only if they have two completely identical destinies besides theiroriginally identical dispositions.

4. The assertion that logical assertions cannot contradict each other isan error resulting from the idea that there are only analytic truths or that“A has b” is true if and only if the representation b is contained in therepresentation A as a component. If the opposite occurs, [107] then thepropositions: each A has b and no C has b can both be true, even if therepresentations A and C do not contain any negation, and then therepresentation of a thing which is at the same time A and C would becontradictory, even if it is composed of nothing but affirmations. This iswhy we do not sustain the usual definition of the most perfect being as abeing that unites all thinkable forces in it. Instead, we define the mostperfect being as a being that unites all forces that are possible side byside and are united in the highest degree in which they can subsist side byside.

5. We deny that all determinations of things of appearance can onlyconsist in relations, because we humans also count as appearances, andthe property of thinking we are all aware of is usually no mere relationbut an inner determination.307

6. The significance of the question whether matter precedes form orform precedes matter can certainly not be understood by means oftemporal precedence. It probably only means this: which of the twogrounds or conditions the other? – So if, as is necessarily the case here,we consider matter as substances and form as the relation of substancesin time and space, we can neither accept the view Kant [108] ascribes toLeibniz, namely, that the temporal and spatial relations betweensubstances are grounded in the inner determinations of the latter; norKant’s own position. For it does not follow from the inner determinationsof a substance at what time it has these determinations and in whichposition it is located. In our view, it is the locations of things which arethe relations between them that we must think in addition to their forces,to explain why things, at this or that time, have precisely this or thateffect on each other.308

307

Kant does not use the word Erscheinungsdinge. When referring to Dinge, he specifies

that these are noumena or things-in-themselves. Appearances, however, are phaenomena

or objects of experience, and here Kant uses the word Gegenstand. These latter become

Objekte when intuitions are united in a concept, through the unity of apperception, when

the manifold of intuition is related to the I think; cf. KrV, B122, B136-137, B197; cf. also

my notes 238 and 251, above.308

Přihonský takes up Bolzano’s definition of space as a collection of locations (WLI, §

79, and Přihonský’s discussion, NAK, pp. 61-62), as well as Leibniz’s view that

substances are related through forces of attraction and repulsion and that space is the

New Anti-Kant 155

The second part of the transcendental logic is called transcendentaldialectic. It is preceded by the doctrine of transcendental illusion (pp.278f.).309 For Kant’s dialectics is nothing other than the “logic ofillusion”. Thus he first raises the question of how illusion and errororiginate. He replies: not through the senses, for the senses do not judgeat all. Not through the understanding by itself, for if the understandingacts merely according to its own laws, the judgment must necessarilyagree with these laws, and then the judgment is not false [109] but right.So error is caused only by the unnoticed influence of sensibility on the

understanding, through which it happens that “the subjective grounds ofthe judgment join with the objective ones310 and make these latter deviatefrom their destination.” (p. 277)311 In addition, Kant remarks that he isconcerned not with empirical illusion, but with transcendental illusion.The first arises from the mere disregard of logical laws and disappears assoon as we pay attention, whereas the second does not stop even if wehave uncovered it and clearly seen its nullity through transcendentalcriticism. The cause of this is that in our reason there lie basic rules andmaxims for its use which look entirely like objective principles, andthrough them it comes about that the subjective necessity of a certainconnection of our concepts for the benefit of the understanding, is takenfor an objective necessity for the determination of things-in-themselves(pp. 278ff.).312

order of coexistences; cf. G. W. Leibniz, Third Letter to Clarke, §§ 4-5; and New Essayson Human Understanding, bk. 2, ch. 13, § 17, where Leibniz criticizes the precritical

Kant’s Newtonian position on space.309

Cf. KrV, B349.310

I follow Guyer and Wood in using “join with” for translating the verb

zusammenfliessen, although Přihonský’s use of the noun das Zusammenfliessen (below) is

more appropriately translated as “confluence”.311

Cf. KrV, B350-351.312

Cf. KrV, B353. Přihonský confuses Kant’s distinctions between empirical and

transcendental illusion and between logical illusion and transcendental illusion, because

he omits the passage where Kant explains that transcendent principles of pure

understanding (which bid us to overstep the boundaries of experience) are incorrectly used

transcendentally (a use that reaches beyond the boundaries of experience) and should only

be used empirically (remain within the boundaries of experience); cf. B352. Kant’s

“Transcendental dialectic” aims at uncovering judgments which appear to be true but are

illusory, and here he presents a critique of dialectical illusion. Transcendental illusion

arises from the “basic rules and maxims of our reason” when they have “the appearance

of objective principles” (KrV, B353). Empirical illusions, such as optical illusions, occur

when the faculty of judgment is misled in the empirical use of otherwise correct rules of

the understanding through the influence of imagination (B351-352). Take the example of

the Müller-Lyer illusion: we see two arrows and mistakenly judge that they have different


Comment. 1. We cannot approve Kant’s way of trying to explain theorigin of error. For he does not answer the question of how sensibilityexerts its influence on the understanding and causes a confluence ofsubjective and objective grounds. We do not even [110] have a clear ideaof what this confluence is supposed to be.

2. If the definition of error presupposes certain maxims for the use ofour reason which (falsely) look like objective principles, then the groundsfor the existence of a certain error are already established in thispresupposition, namely, the error by means of which we take the merelysubjective necessity of a conceptual connection for an objectivenecessity.

3. A completely different and, in our view, more right approach istaken by Bolzano, who begins by pointing out that propositions as suchor objective propositions can stand not only in the relation of deducibilitybut in the relation of mere probability, which is entirely independent ofour faculty of judging. Thus for instance the proposition: “Caius hasdrawn a black ball from an urn” stands to the proposition “there areninety black balls and ten white balls in this urn” in a probabilityrelation, such that the degree of this probability is 9/10 and thisindependently of whether anyone holds one of these propositions as true(Theory of Science, § 309).313 Bolzano further claims that if a proposition

lengths. However, logical illusion (or the illusion of fallacious inferences) “arises solely

from a failure of attentiveness to the logical rule”, KrV, B353.313

Bolzano discusses the probability relation between propositions as such in WLII, § 161,

where he gives the example mentioned by Přihonský and distinguishes between

probability and confidence. He defines probability as a relation between given

propositions, where it is not presupposed that these propositions should be represented or

believed by a thinking being: “being a relation between two collections, the relative

satisfiability or probability of a proposition has a certain magnitude which, if it is

determinable at all, can be represented by a fraction whose denominator and numerator are

related as these two collections” (§ 161.2). However, confidence (Zuversicht) can range

from negative to positive values, with zero indicating indecision; cf. § 161, note 1. In

WLIII, § 309, Bolzano discusses grounds for the possibility of error and circumstances

which endorse error. In Bolzano’s view, “error arises whenever it happens that something

which we have recognized as probable, following perfectly correct rules of inference, and

which we expect and take as true, turns out to be false” (§ 309.3). However, unlike

Přihonský I think that this view seems surprisingly close to Kant’s discussion of empirical

and logical illusions since Bolzano claims (i) that the error lies in our judgment and (ii)

that the main cause of error lies in our misdirected attention. According to Kant, empirical

and logical errors or illusions reside in our misdirected faculty of judging in its use of

otherwise correct rules of the understanding, either because of the influence of

imagination, or because we fail to attend to a logical rule; cf. KrV, B353 and my note 312,

above.

New Anti-Kant 157

M stands in the relation of probability to other propositions A, B, C, D, . .

. which we take to be true, this probability relation is a feature of ourfaculty of judging, if the propositions stand in a relation we havecognized as a predominant probability [111] (i.e., when we estimate thisprobability as greater than one-half) and we not only make the judgment“M is probable” but the judgment “M” itself and make the judgment Mwith the degree of confidence that corresponds to the perceived degree ofprobability. Nobody will deny that this is so, or that in making this

judgment (by expecting that which has a higher degree of probability),we do not yet necessarily make a mistake, although it is understandablehow the possibility of error arises from this mental feature.314 For theprobable, however high its degree of probability may be, does not haveto be true. So, when we feel obliged to expect that which has a highdegree of probability and thus hold it to be true, we occasionally holdsomething to be true which in fact is not true, that is, we err.

Kant defines pure reason (the seat of transcendental illusion) as thefaculty of principles (synthetic cognitions from pure concepts), p. 280.315

In drawing inferences, reason strives to reduce the great manifold ofcognitions to the smallest number of principles and to bring about thehighest [112] unity in this way (p. 284).316 The supreme principle of purereason is the proposition: “when the conditioned is given, then the wholeseries of conditions subordinated one to the other (a series that is henceitself unconditioned) is also given” (p. 286).317 But the question is: doesthis principle has an objective validity or is it merely a logical

(i) In the first part of his almost Kantian claim, Bolzano says that “we do not err when we

estimate that the probability degree that a black ball will appear is 99/100 but only when

we expect the appearance of a black ball (on the basis of this estimate or for some other

reason)” (§ 309.3). The probability (Wahrscheinlichkeit) of an error is its illusion

(Schein), and this illusion arises from judgments which were inferred according to

probability inferences; “every error is a proposition which stands in a certain relation of

probability to the other propositions taken as true by the being who makes the error” (§

309.4). Hence error occurs in both empirical and conceptual judgments. (ii) In the second

part of his claim, Bolzano says that error is caused, for example, by ignorance, lack of (or

misdirected) attention, an inappropriate association of ideas which then misdirects our

attention, faulty memory, imagination that is either too strong or too weak, and our own

will when it misdirects our attention; cf. § 309.6.314

Cf. WLIII, § 309.2. Bolzano uses the word Seele (soul) whereas Přihonský uses Geist(mind). When Přihonský mentions Bolzano’s view that error resides in a feature of our

faculty of judging, he fails to notice that Kant makes precisely the same point (although in

a different context, i.e., the critique of dialectical illusion and the problem of

transcendental illusion); cf. KrV, B351-352, and my notes 211 and 312, above.315

Cf. KrV, B355-356.316

Cf. KrV, B362-363.317

Cf. KrV, B364.


prescription for bringing the highest possible logical unity into ourcognitions? This is what Kant investigates in the two books of thetranscendental dialectic: the first book deals with concepts of reason andthe second book with syllogisms of reason.

Kant would have us call ideas those concepts of pure reason which gobeyond the possibility of experience. In his view, Plato already used theword idea in this sense (p. 289),318 and we do not wish to argue aboutthis, even though we think that this word can be taken in a moreappropriate sense. Kant says that we can expect that the forms of

syllogisms will contain transcendental ideas in the same way the forms ofjudgment provided the categories (p. 295).319 The series of prosyllogisms

or the ascending series of conditions, which may be either determinate orindeterminate, has to be presupposed by reason as complete,

unconditioned, or absolutely true (pp. 301f.).320 – Kant derives the

318

Cf. KrV, B370. By drawing a distinction between ideas (or pure concepts of reason

which serve for comprehension (Begreifen) and go beyond the possibility of experience)

and categories (or pure concepts of the understanding which serve for understanding

(Verstehen) of perceptions) (cf. KrV, B368; P, § 41), Kant attempts to find a middle way

between Plato’s and Aristotle’s accounts of ideas: he criticizes Plato for hypostatizing

ideas by making them into “archetypes of things themselves” and using the word “idea”

for “something that not only could never be borrowed from the senses but that even goes

far beyond the concepts of the understanding” (B370). On the other hand, he criticizes

Aristotle’s using the word “idea” for concepts of the understanding and limiting it to

experiential cognition (ibid.). In Kant’s view, the categories of the understanding relate to

possible objects of experience, whereas the ideas of reason relate to “the absolute totality

of all possible experience”, i.e., the construction of experience which “is not itself an

experience but a necessary problem for reason”, since it must be represented by reason,

and for this representation reason needs different concepts from those of the

understanding (cf. P, § 40).319

Cf. KrV, B378. Here Kant uses the word form in the singular and refers to the “form of

judgments” and the “form of syllogisms”.320

Cf. KrV, B387-389. While Přihonský mentions only the prosyllogisms, Kant says that

the same action of reason leads to a ratiocinatio prosyllogistica or series of inferences

“that can be continued to an indeterminate extent either on the side of the conditions (perprosyllogismos) or on the side of the conditioned (per episyllogismos).” (B387) The

prosyllogisms or ascending series of inferences are related to the faculty of reason

differently from the episyllogisms or descending series of inferences. Prosyllogisms

proceed to the unconditioned (from secured consequences to indeterminate grounds), in

which case the inferred cognition (the conclusion) is either complete and necessary and

therefore needs no grounds or, if it is derived, it is a member of a series of grounds that is

itself unconditionally true. Episyllogisms proceed to conditioned cognitions, and if a

cognition (the conclusion) is regarded as conditioned, the descending series is considered

as a potential progression a parte priori (from sufficiently secured grounds to

consequences). Cf. B387-389.

The reason Kant makes this distinction is to explain that transcendental inferences of

reasons are ascending series, since cognitions of reasons are unconditioned or determined

New Anti-Kant 159

transcendental ideas from the following remark (pp. 302f.):321 everyrelation to a representation is only a twofold relation, either to the subjector to its objects, and the latter can be either mere [113] appearances orobjects of thinking. So every relation of representations is threefold: 1)to the subject, 2) to the manifold of the object in appearance, and 3) to allthings as such. Consequently all transcendental ideas can be broughtunder three classes: 1) the unconditioned unity of the thinking being (theidea of the soul); 2) the absolute unity of the series of conditions ofappearances (the idea of the world); and 3) the absolute unity of thecondition of all objects of thought as such (the idea of God).

Comment. Could anything be more forced than this derivation of thethree ideas: soul, world, and God? Even the leniency in the relationsbetween representations is forced and wrong! – A relation to the thinkingsubject only occurs in representations in the subjective sense of the word,that is, in appearances in the mind of a thinking being. However,representations as such are also related to objects.322 In addition, we

a priori, and reason can arrive at cognitions from principles by ascending from the

conditioned synthesis to the unconditioned (cf. B390). He shows the relation between

ideas and the form of the syllogism on the one hand and that the legitimate use of the

ideas is only regulative and not constitutive, on the other. However, the understanding

“always remains bound” to the conditioned synthesis and can never reach the

unconditioned (ibid.).321

Cf. KrV, B390-391. Kant says that “what is universal in every relation that our

representations can have is 1) the relation to the subject, 2) the relation to objects, and

indeed either as appearances or as objects of thinking as such. If we combine this

subdivision with the above division [between representations in general and between

concepts and ideas in particular (cf. B376-377)], then all the relations of representations

of which we can make either a concept or an idea are of three sorts: 1) the relation to the

subject, 2) to the manifold of the object in appearance, and 3) to all things as such.” (KrV,

B391)322

We should note that Kant and Bolzano-Přihonský have different conceptions of the

notions “objective representation” and “subjective representation”. Kant’s representations

are subjective, in the (Bolzano-Přihonskian) sense that they occur in a thinking being.

Kant’s notion of representation concerns cognition, and all cognitions are representations

consciously referred to an object (cf. Jäsche Logic, § 1). For Kant, a subjective

representation is a sensation (sensatio), whereas an objective representation is a cognition

(cognitio). Cognitions are either intuitions, concepts of the understanding, or concepts of

reason (ideas). Intuitions relate immediately to objects (of experience), concepts relate

mediately to objects (of experience), and ideas transcend the possibility of experience; and

their objects are the soul, the world, and God, which are, respectively, the objects of

psychology, cosmology, and theology; cf. KrV, B376, B391.

For Bolzano-Přihonský, subjective representations are mental events but not mental

pictures, with the possible exception of intuitions; cf. WLI, §§ 48.3, 52.3-5, 72; LogischeVorbegriffe (BBGA 2, A, vol. 5), § 4.2-3. Objective representations are correlative

contents of subjective representations and components of propositions. They are meanings


should recall that not every representation represents an object(regardless of whether it is taken in a subjective or an objective sense),for example, the representations nothing, √-1, and others.323 We shouldalso not forget that the distinction between objects as appearances andobjects of thinking is extremely illogical as such, since the first term iscompletely contained in the second. – Even more peculiar is the way inwhich [114] the two ideas world and God are supposed to be derivedfrom the two latter relations. Is not the world the collection of all finite

substances, rather than the collection of appearances alone?324 It is alsowrong to define God as a thing (being) containing the highest conditionfor the possibility of all that is thinkable. First, not all objects of thoughtreally exist, for example, truths as such. And it is wrong to make thepossibility of truths as such, especially necessary a priori truths, dependon God’s thinking and cognizing.325 God cognizes conceptual

of signs (Bedeutungen von Zeichen) which arouse, or are grasped by, subjective

representations, and these meanings can have different linguistic expressions; cf. WLI, §§

48.3, 50.3; WLIII, §§ 270-271; “Ueber den Begriff des Schönen” (Prag, 1843), pp. 87-217.

Cf. also Přihonský’s introduction where he presents some basic Bolzanian concepts, NAK,

pp. 7-17.323

Kant does not call the concept of nothing a representation, because it is not a cognition

but belongs to the concepts of reflection and is subordinate to “the concept of an object as

such (taken problematically, leaving undecided whether it is something or nothing).”

(KrV, B346) However, Kant would agree with Přihonský that not every concept has an

object, adding that the reverse is also true, for not every object has a concept.

Accordingly, he divides the concept of [nothing] into four and distinguishes two thought-

entities of [nothing] and two nonentities of [nothing], respectively: (1) a logical [nothing]

which cannot be represented (ens rationis), and (2) a concept without an object or the

cancellation of opposed predicates (nihil negativum). In addition, [nothing] can be (3) an

empty intuition without a concept (ens imaginarium) or (4) an empty object without

concept (nihil negativum), which is an empty concept, like (1). However, (2) and (4) are

empty data for concepts. Cf. KrV, B347, and my note 306, above.324

Kant would reply to Přihonský that the world is not a whole existing in itself outside

our representations and that it cannot be the object of knowledge. Therefore we cannot

inquire into its composition or define it as the collection of all finite substances, since it is

precisely this sort of inquiry which generates the antinomies of pure reason; cf. KrV,

B455. Instead, the word “world”, taken in the transcendental sense, “signifies the absolute

totality of the sum total of existing things”, but these are appearances (not noumena), and

for this reason “by ‘world’ is understood the sum total of all appearances”, KrV, B446.

Přihonský could then reply that although the world is not an object of cognition, we could

still inquire into the composition of the world in order to progress in our cognition.325

Přihonský’s point is not quite correct: Kant does not make the possibility of truths as

such depend on God’s thinking and cognizing, (1) because in the first Critique, Kant

discusses the transcendental idea [God] which is a regulative concept (of reflection) and

not a constitutive concept or category. In addition, he restricts knowledge of God to

analogy and says that in this case, knowledge must make room for faith (KrV, B, preface,

xxx.) (2) Kant does not have a notion of objective truths or truths as such in Bolzano-

New Anti-Kant 161

truths, pure conceptual truths, the law of gravitation, the supreme moralprinciple, and so on, because they are true. But it is not the case that theyare true because God cognizes them: instead, they are entirelyindependent of God’s or the creator’s will. Therefore not everythingexisting has a ground of its existence and its possibility in God, for atleast God must be excluded from this.

As for the dialectical inferences of pure reason, (according to p. 307)there are supposed to be three classes, just as there are three ideas inwhich their conclusions result.326 In the first class, I infer the absoluteunity of a subject from the concept of a subject that contains nothingmanifold. [115] Kant calls this inference a paralogism of pure reason.327

The second class deals with the absolute totality in the series ofconditions for a given appearance. Kant calls it the antinomies. In thethird class, I infer a being of all beings, and Kant calls this inference ofreason the ideal of pure reason. Besides, he already points out on p. 283

Přihonský’s sense (cf. my note 602, below). In Bolzano’s view, truths as such (or true

propositions) are thought and cognized by God: “every proposition [. . .] is at least

thought and represented by God and, if it is a true proposition, God will also acknowledge

it as true. Thus it occurs in the divine understanding either as a mere idea or as a

judgment.” (WLI, § 19.1.b) It seems that Přihonský uses this claim as the starting point of

his objection against Kant and incorrectly infers from the Bolzanian view that Kant made

truths as such depend on God’s thinking and cognizing. (3) Kant nowhere claims that

objects of thought actually exist; on the contrary, he rejects this claim when he argues that

noumena are used illegitimately, if we ascribe objectivity to them; cf. KrV, B345. Cf. also

P, § 45, where Kant says explicitly that one such illegitimate objectification arises when

reason represents “the objects of experience in a series stretching so far that no experience

can comprise the likes of it” and, to complete the series, looks for noumena entirely

outside experience (i.e., a transcendental idea, such as God) to which it can attach the

chain. Přihonský also makes an incorrect inference from “conditions of possibility” to

“existence”: even if God were defined as containing the highest condition for the

possibility of all that is thinkable, this definition does not imply that objects of thought

actually exist.326

Cf. KrV, B397-398. Cf. Scholz (1931, p. 221), who comments that the classical doctrine

of inference, on which the Kantian one is based, was radically transformed by modern

logic so that a detailed critique of this doctrine has become almost pointless.327

Kant writes “transcendental concept of a subject” and “transcendental paralogism”

(KrV, B398). In Aristotle’s view, a paralogism occurs when we falsely infer the truth of

an antecedent from the truth of a consequent: “Whenever, if one thing is or happens,

another is or happens, men’s notion is that, if the latter is, so is the former-but that is a

false conclusion.” (Poetics, 1460a.21) Kant distinguishes between a logical and a

transcendental paralogism: a logical paralogism is “a syllogism which is fallacious in

form, regardless of its content” and a transcendental paralogism is “one in which there is a

transcendental ground, constraining us to draw a formally invalid conclusion” (KrV,

B399). The paralogisms in the transcendental dialectic belong to the second kind.


and repeats on p. 296 that these three kinds of inferences correspond tothe categorical, hypothetical, and disjunctive inferences.328

Comment. It would be taking things too far if we were to completelyrefute Kant’s doctrine of dialectical inferences by showing theinadequacy of Kant’s doctrine of inference as such. Therefore thefollowing remarks shall suffice. – In our view, the principal division ofinferences into immediate and mediate inferences, which appears inKant’s Logic and is maintained by his pupils, is ineffective insofar as theformer are considered as inferences from a single premiss and the latter

as inferences from several premisses.329 For if the division of inferences

328

Cf. KrV, B360 and B379.329

Cf. Jäsche Logic, § 42, where Kant says that an immediate inference is the derivation of

one judgment from the other without a mediating judgment, whereas a mediate inference

contains not only the concept contained by the judgment itself but also other concepts

which are needed for deriving a cognition from it. He then classifies inferences of the

understanding as immediate inferences and inferences of reason or inferences of the

faculty of judging as mediate inferences; cf. JL, § 43. In inferences of the understanding,

the judgment is derived directly, without intermediate premisses, cf. JL, §§ 46-54, where

Kant discusses inferences from the universal affirmative to the particular affirmative; the

denial of the contradictory, contrary, and subcontrary; as well as the rules of conversion

and contraposition. In determining inferences of reason, Kant follows the Aristotelian

syllogism, i.e., of a valid argument where the conclusion is different from any of the

premisses: “Reason [. . .] is the faculty of inferring, i.e., of judging mediately (through the

subsumption of a condition of a possible judgment under the condition of something

given). The given judgment is the universal rule (major premise, major). The subsumption

of the condition of another possible judgment under the condition of the rule is the minor

premise (minor). The actual judgment that expresses the assertion of the rule in the

subsumed case is the conclusion (conclusio).” (KrV, B386) According to Aristotle, a

syllogism must have more than one premiss, and every conclusion follows from two

premisses which relate the terms of the conclusion to a a third or middle term; cf. PriorAnalytics, i.25 (41b.36). Contemporary logic allows for syllogisms with 0 or 1 premiss;

for example, “he breathes, therefore he is alive”. So contrary to Přihonský’s belief, there

is a distinction between inferences based on a number of premisses, albeit one that

traverses the history of logic.

Bolzano draws a different distinction between immediate and mediate inferences, which

Přihonský may have in mind, although he does not mention it here but on pp. 174 of the

NAK (cf. also my note 331, below). Bolzano claims that “all our judgments come about in

two ways: either we make them immediately, i.e., when we are not induced to make a

judgment by certain other judgments from which they can be derived; or we derive them

from other judgments.” (WLIII, § 308) So for Bolzano, in an immediate judgment there is

no inference, whereas a mediate judgment is the conclusion of an inference. In his view,

immediate judgments are either judgments of perception containing an indexical

statement, such as “This (what I now see) is A”, or conceptual judgments; cf. WLIII, §

312. Immediate judgments are infallible, in that they express a true proposition, and they

are basic judgments, from which other judgments are inferred. He clearly adheres to an

axiomatic conception, since he claims that we can only understand the existence of

New Anti-Kant 163

is grounded on the number of premisses, there is no telling whyinferences from one premiss are referred to a special class, whereasinferences from two, three, or more premisses are all brought into thesame class. At best, we could do this only if the inferences from morethan two premisses include no peculiar, [116] single, or atypicalinferences, but this is not the case. And by what right do we call someinferences immediate and others mediate? Does not each conclusionimmediately follow from its premisses? – Since we also consider thedistinction between judgments according to the four well-knownmoments of quantity, quality, and so on, as wrong, we cannot accept itsuse for the division of inferences. Does it not appear extremely arbitrarywhen it is applied to immediate judgments? Here, in immediateinferences, there are supposed to be ways of inference belonging to eachmoment, whereas there is supposed to be a different explanation formediate inferences, and the division is supposedly grounded on themoment of relation alone!330 Even assuming that the inferences belongingto the moment of quantity do not provide a useful division, it is not clearwhy the quality of propositions in an inference does not ground a usefuldivision. The different modi of syllogism named by logicians indeedfollow the quality of the premisses. And what are these modi if not termsof a division? As regards modality, inferences must be subject to thesame differences as propositions in general, insofar as by the modality of

mediate judgments on the basis of the existence of immediate judgments; cf. WLIII, §

300.3. On Bolzano’s axiomatic conception, i.e., that there are basic truths from which

other truths are derived, cf. A. Von Duhn (2002), “Bolzano’s Distinction between

Confirming (Gewissmachen) and Justifying (Begründen) Propositions” forthcoming.330

Cf. JL, § 60, note 1. Somewhat surprisingly, Přihonský does not mention that Kant fails

to distinguish between an inference rule and its application: the rule is the major

proposition in a syllogism. Cf. JL, § 58. For Kant, inference rules are the pure principles

of the understanding and of reason, and in his view they are synthetic, because the

subject-concept is not contained in the predicate-concept. Cf. JL, § 53. According to Kant,

the problem with synthetic principles of reason is that if they are a priori, reason is

unrestricted by experience and extends itself beyond its legitimate limits. Cf. KrV,

B382ff. As a result, reason is caught in contradiction by making opposed but equally valid

inferences (Vernunftschlüsse). Cf. KrV, B454.

Unlike Kant, Bolzano distinguishes between a rule and its application when he says that

the rule of transitivity that from two propositions “A is B” and “B is C” follows “A is C”

is a synthetic a priori truth or truth-bearer, whereas a particular application of this rule is

an analytic truth; cf. WLIII, § 315.2. Synthetic propositions, for Bolzano, do not have a

component which can be varied arbitrarily whilst conserving the original truth value of the

proposition, as do analytic propositions. Inference rules are synthetic because they contain

variables, so that a substitution would be invalid; cf. WLII, § 148, § 197. Cf. A. Von Duhn

(2001), “Theoretical Laws and Normative Rules: Kant and Bolzano’s Views on Logic”,

pp. 3-12.


a proposition we understand the greater or lesser degree of certainty withwhich we [117] hold it to be true. – In addition, it is not easy to see whythe syllogism is the only simple (mediate) form of inference and whythere cannot be several forms. Irrefutable examples show that amonginference forms with two premisses, there are several forms that do notbelong to syllogism. If not, how can we subordinate the inference derivedfrom the following two premisses to the syllogistic form:

“that which has a, has b” and

“that which has b, has a”;

the inference: “every object of one of the representations A and B” (for A

and B denote the concreta of the abstract representations a and b) “is an

object of both” or “A and B are variable representations”?*331 – Should

we not rather admit that this is quite a simple inference form which is

completely different from syllogism? – The same applies to the

inference:

“that which has a, has b,

that which does not have a, has b,

therefore every something has b”; and to many others; and these are

not only inferences based on two premisses but inferences based on three

or more premisses. –

[118] Moreover, we think that the division of syllogisms intocategorical, hypothetical, and disjunctive is also wrong.332 The so-calledhypothetical syllogism is only a type of categorical inference, namely,modus Barbara if we use modo ponente and modus Camestres if we usemodo tollente. Therefore it should not be assigned to the categoricalinferences but subordinated to them. It is completely incomprehensiblehow the inference type called disjunctive syllogism could be assigned tosyllogisms. The medius terminus which is essential in syllogisticinferences is missing and would be wrong here, given that there are twotermini extremi which should be united in the conclusion. For if we inferusing modo ponente, then from the two premisses: “only one of the

*

331 For example: “that which has equilaterality also has equiangularity; that which has

equiangularity also has equilaterality; therefore [equiangular] and [equilateral] are

variable representations.”332

Cf. JL, § 60.

New Anti-Kant 165

propositions: A, B, C, . . . is true” and “A is true”, we derive theconclusion: “therefore B, C, . . . are false”. Modo tollente results from thetwo premisses: “only one of the propositions: A, B, C, . . . is true” and “A

is false” and the conclusion: “therefore only one of the propositions B, C,

. . . is false”. So where would be the terminus medius or the two extremithat would be united in the conclusion? Although the representation A

occurs in both premisses, in the major it does not occur as subject- orpredicate-representation and thus not as terminus medius (cf. Bolzano’sTheory of Science, § 223ff.).

Someone who considers all this and draws from it the conviction[119] that the doctrine of inferences in use till now is inadequate andwrong, in particular its division into categorical, hypothetical, anddisjunctive inferences, will hardly be inclined to see a special warrantyfor its deeper grounding in the nature of the human cognitive faculty inthe fact that those “three dialectical inferences should correspond to thethree inferences of reason”.333

But let us consider in detail what Kant teaches on the three classes ofdialectical inferences and begin with his assertions about the paralogisms

of pure reason (pp. 308f.).334 The (transcendental) concept I is notcatalogued in the table of transcendental inferences (he says on p. 309),because it is the vehicle of all concepts as such: the I think is the onlytext from which rational psychology should develop its entire wisdom(p. 310).335 In addition, Kant repeatedly explains that this I think isalready an empirical proposition (p. 323, note).336 – If the table of

333

Cf. KrV, B390. Kant refers to the transcendental (not the logical) dialectic when he

says that it is “supposed to contain both the origin of certain cognitions from pure reason

and inferred concepts, whose object cannot be given at all, and so lies wholly outside the

faculty of the pure understanding. We have gathered from the natural relation that the

transcendental use of our cognition must have in its inferences, as well as in its

judgments, that there are only three species of dialectical inferences relating to the three

species by which reason can arrive at cognitions from principles;”. Rather than

Přihonský’s correspondiren, Kant uses the expression sich auf . . . beziehen, which

translates as “relating to”.334

Cf. KrV, B399.335

Cf. KrV, B399-400. Kant says that the concept or, if one prefers, the judgment I think“can have no special title, because it serves only to introduce all thinking as belonging to

consciousness.” It serves to distinguish the following two objects: “I, as thinking, am an

object of inner sense and am called “soul”. That which is an object of outer sense is called

“body”. Accordingly, the expression I as a thinking being already signifies the object of a

psychology that could be called the rational doctrine of the soul, if I do not seek to know

anything about the soul beyond what [. . .] can be inferred from this concept I insofar as it

occurs in all thinking.” (B399-400)336

Cf. KrV, B422. Kant discusses the Cartesian cogito ergo sum and I think as containing

within itself the proposition I exist: “I exist thinking” is an empirical proposition, for


categories is applied to this I (= soul), we obtain the following fourpropositions: 1) the soul is substance; 2) in its quality, it is simple; 3) inthe different times in which it exists, it is numerically simple, that is, ithas unity; and 4) it is in relation to possible objects in space. These aresupposedly the four paralogisms of the transcendental doctrine of thesoul (p. 311).337 For [120] 1. That the I can always be considered assubject in our thinking is an identical proposition, but it is not the casethat this I is a substance, for this demands entirely different data(p. 314).338

2. That the I is singular is likewise a merely analytic proposition, butthat it is a simple substance is a synthetic proposition which merelysensible intuitions are not entitled to establish (ibid.).339

3. Even the identity of the subject of which I am conscious in all myrepresentations does not concern the intuition of it, through which it is

“how should it be possible to go beyond experience (of our existence in life) through the

unity of consciousness with which we are acquainted only because we have an

indispensable need of it for the possibility of experience [. . .]?” (B420) Kant

distinguishes between rational and empirical psychology and claims that the former is

engaged in a fruitless search for a thinking substance or absolute subject of I think and

that the substantiality, identity, and simplicity of the soul rests on a fallacious syllogism

or paralogism; cf. KrV, B399, B421. Empirical psychology, on the other hand, has a place

within anthropology.337

Cf. KrV, B402-404. Kant defines a transcendental paralogism as a syllogism “in which

there is a transcendental ground, constraining us to draw a formally invalid conclusion.”

(KrV, B399) The transcendental paralogisms of the doctrine of the soul “are falsely held

to be a science of pure reason about the nature of our thinking being” (ibid.). In these four

paralogisms, the I think (the transcendental ground of thinking) is the basis for inferences

about the substantiality, simplicity, and identity of the soul. In Kant’s view, this I is an

empty expression which denotes the thought of an absolute but logical unity of the

subject; cf. KrV, A356. “Through this I, or He, or It, (the thing) which thinks, nothing

further is represented than a transcendental subject of thoughts = x” (B404). The nature of

this I can only be known by analogy, not by abstraction, nor is it a necessary ground

underlying cognition and experience; cf. A356, B422. This is why he rejects Descartes’

claim in the Discours de la méthode (1637) that we can prove existence by means of self-

awareness, although he admits that the I think (which remains uncognizable) ultimately

results from a process of abstraction from experience.338

Cf. KrV, B407. Kant says that the I that I think can always be considered as subject isan identical proposition, but this “does not signify that I as object am for myself a self-subsisting being or substance. The latter [. . .] demands data that are not encountered at all

in thinking, and thus (insofar as I consider merely what thinks as such) perhaps more than

I will ever encounter anywhere (in it).”339

Cf. KrV, B108.

New Anti-Kant 167

given as object, and thus cannot prove the identity of the substance, forwhich several synthetic propositions would be necessary (p. 315).340

4. That I distinguish my own existence from other things outside meis likewise an analytic proposition. But I do not thereby know whether Ican exist without these things outside me (ibid.).341

Then Kant attempts to show that there are false conclusions in theusual argument for the substantiality of the soul, as well as inMendelssohn’s proof of the persistence of the soul (pp. 316f.)342

Comment. Our readers already know that we consider Kant’sassertion that we are only entitled to analytic judgments aboutnonsensory objects but not to synthetic judgments (which is why we fallinto nothing but paralogisms when we venture to make such judgments)itself as a [121] paralogism.343 Hence we can justifiably confine ourselvesto recalling the following remarks:

1. In the introduction to this part of the Critique, Kant says that whatthese paralogisms of rational psychology have in common is that fromthe transcendental concept of subject, which does not contain a manifold,they infer the absolute unity of this subject. However, the discussionshows that only one of these (four) paralogisms draws the conclusion ofthe unity of the soul and for an entirely different reason from the one justsuggested, namely, from the identity of consciousness. The other threeparalogisms have entirely different conclusions.

2. If we are still undecided about whether there is a simple substance

or a collection of several substances in which thinking occurs, we can

340

Cf. KrV, B408-409. Kant writes “[t]he proposition of the identity of myself in

everything manifold of which I am conscious is equally one lying in the concepts

themselves and thus an analytic proposition; but this identity of the subject, of which I can

become conscious in every representation, does not concern the intuition of it, through

which it is given as object and thus cannot signify the identity of the person, by which

would be understood the consciousness of the identity of its own substance as a thinking

being in all changes of state; to prove that would demand not a mere analysis of the

proposition I think but various synthetic judgments grounded on the given intuition.”341

Cf. KrV, B409. Kant says that the proposition “that I distinguish my own existence, that

of a thinking being, from other things outside me” is an analytic proposition, “for other

things are those that I think of as distinguished from me. But I do not thereby know at all

whether this consciousness of myself would even be possible without things outside me

through which representations are given to me, and thus whether I could exist merely as a

thinking being (without being a human being).” (B409)342

Cf. KrV, B413.343

Přihonský refers to KrV, B408-409. N.B. Kant’s main point is that the inference from

the formal conditions of thought to a substance of thought is a paralogism; cf. B399-402,

B421.


claim that there is one substance with the same degree of certainty withwhich we can infer the existence of any substances whatsoever.

3. Of course our intuitions do not teach us immediately that our soulshould be a simple substance, any more than they teach us immediatelythat the things around us are complex.

4. Likewise we accept that certain synthetic judgments are needed forproving the identity of the soul, but we have shown that Kant has notmanaged to prove the impossibility of such synthetic cognitions. [122]

5. The question is not about knowing whether I can exist withoutthings outside me but merely about whether my soul is a substance that isnumerically different from one of the substances of these things.344

6. Although there are mistakes in the proofs used up till now, namely,in Mendelssohn’s, the mistakes are not the ones Kant accuses the proofsof having. But someone should show which fallacious inferences weremade in the proofs of substantiality, simplicity, unity, and persistence ofthe soul, which Bolzano carried out in the Athanasia (Sulzbach: Seidel,1838, 2nd ed.).345 The Critique gives the following explanation of the

344

Kant would reply that the question is not whether my soul is a substance that is

numerically distinct from one of the substances of the things existing outside me, but (1)

whether self-awareness is possible without things outside of me and (2) whether I can

exist merely as a thinking being (without being a human being); cf. KrV, B409. Kant

rejects Descartes’ cogito ergo sum, or the claim that existence can be inferred from self-

awareness, whereas Přihonský is concerned with the question whether there is a numerical

difference between the soul and other substances; so the two authors are dealing with

different issues.345

Bolzano’s main argument in the Athanasia is an attempt to prove the immortality of the

soul by proving its substantiality: (1) the soul is a simple substance; (2) no simple

substance passes away; (3) therefore no soul passes away. However, Bolzano points out

himself that to prove the persistence of the soul, it is insufficient to prove that no soul

passes away. In addition, we need to show that we are conscious of our experience and

can recall our present state; cf. Athanasia, p. 9; cf. Mark Textor (1999), “Bolzano über dieUnvergänglichkeit der Seele”, pp. 269-290, who criticizes Bolzano’s argument.

So Bolzano takes us back to Kant’s point about consciousness and the problem of the

unity and simplicity of the I think. In Kant’s view, the unity of the I think is falsely

inferred from the simplicity of the thinking being. For although I think an absolute logical

unity of the subject (simplicity), I do not cognize the real simplicity of the subject. The

concept [substance] is merely a valid condition of our cognition but not of the object that

is to be specified; cf. KrV, A356. In other words, we are committing a transcendental

paralogism by inferring from the formal conditions of cognition to the substance of

cognition. But cf. Scholz (1931, pp. 221-222) who comments that even in the most

favourable interpretation, Kant only shows that the proofs of the immortality of the soul

published before 1781 do not stand up to a severe critique. Kant did not provide us with a

proof of indemonstrability, such as mathematics provides for the Euclidean axiom of

parallel lines with regard to the other axioms in a complete Euclidean axiomatic system,

for the simple reason that such a proof is practically impossible since it requires an

New Anti-Kant 169

antinomies of pure reason (pp. 330–331).346 When reason is applied to theobjective synthesis of appearances, what happens is that it soon findsitself involved in contradictions which lead to an antithesis of purereason. In its progress, reason necessarily runs into propositionsaccompanied by an illusion which is not only artificial but natural andinevitable. – If we now consider the table of categories and emphasizethose which are accompanied by a series in the synthesis of the manifold,we find that there are no more than four cosmological ideas: the absolute

completeness [123] of 1) the composition of a given whole of allappearances, 2) the division of a given whole in appearance, 3) the origin

of an appearance as such, and 4) the dependence of the existence of whatis variable in an appearance (p. 336).347 These ideas lead to the followingantinomies (contradictory propositions).348

First antinomy. Thesis. The world has a beginning in time and inspace it is also enclosed in boundaries. Antithesis. The world has nobeginning and no bounds in space but is infinite with regard to both spaceand time (pp. 344–345).349

Second antinomy. Thesis. Every complex substance in the worldconsists of simple parts, and nothing exists anywhere except the simpleor what is composed of simples. Antithesis. No complex thing in the

established axiomatic system, and in philosophy this presupposition is not supplied. We

should ask the prerequisite question which neither Kant nor Bolzano have raised: what do

we actually mean when we talk about the immortality of the soul? And we will find that

every attempted explanation is based on the presupposition that the soul of a human being

is a substance which is bound to the body of this human being in such a way that the

human being is characterized without any further description as a product of its soul and

body. Now, this interpretation of the soul does not go beyond an incomprehensible

combination of words, and anyone who does not see this is either incoherent, or else he

must renounce an examination of the immortality question on the grounds that it makes no

sense. A similar possibility must be considered concerning the question of God, for it is

difficult to say what the assertion “there is a God” means, since it cannot mean that we

can determine a location in space at which he can be found at any time.346

Cf. KrV, B432. Kant refers to antinomy in the singular. An antinomy is a rhetorical

form of presentation used by Quintillian, in which opposed forms of argument are

presented side by side; cf. Institutio oratoria, bk. VIII, ch. 7. Kant uses the antinomy in

the transcendental dialectic to show that reason can make opposed but equally valid

inferences. His point is that reason illegitimately goes beyond the limits of experience in

making inferences about cosmology and its object, the world, and thus “finds itself

involved in such contradictions that it is compelled to relinquish its demands in regard to

cosmology” (B433).347

Cf. KrV, B443.348

Kant also calls the antinomy of pure reason, the antithetic of pure reason and the

conflict of the transcendental ideas; cf. KrV, B454.349

Cf. KrV, B454-455.


world consists of simple parts, and nowhere does anything simple exist inthem (pp. 350–351).350

Third antinomy. Thesis. Causality according to the laws of nature isnot the only one from which all the appearances of the world can bederived. To explain these appearances, it is also necessary to assumeanother causality through freedom. Antithesis. There is no freedom, andeverything in the world happens solely according to the laws of nature(pp. 358–359).351

Fourth antinomy. Thesis. To the world there [124] belongs somethingthat, either as a part of it or as its cause, is an absolutely necessary being.Antithesis. There is absolutely no necessary being existing anywhere,either in the world or outside the world, as its cause (pp. 364–365).352

Kant takes each of these eight propositions as a proof in which no

false conclusion whatsoever should be provable, unless it were the falseconclusion that we assume the right to make synthetic judgments aboutnonsensory objects.

Excerpts of these proofs read as follows:1. On the first thesis. The world necessarily has a beginning in time,

for if it has no beginning, then up to every given point in time, aneternity has elapsed and hence an infinite series of successive states ofthings in the world, each following another, has passed away. But suchan infinite series of successive states of things is impossible, since theinfinity of a series consists precisely in the fact that it can never be

completed by a successive synthesis. – In addition, the world is notinfinite with regard to space but enclosed in boundaries. For an infiniteaggregate of real things cannot be regarded as a given whole or as agiven whole of simultaneously existing things. For in order to think theworld that fills all spaces as a whole, the successive synthesis of the partsof an infinite would have to be regarded as completed, [125] that is, in

the enumeration of all coexisting things, an infinite time would have to be

regarded as having elapsed, which is impossible.353

On its antithesis. The world has no beginning. But let us suppose thatit has a beginning. Since its beginning is an existence preceded by acertain time in which the thing does not yet exist, that is, an empty time.

350

Cf. KrV, B462-463ff.351

Cf. KrV, B472-473.352

Cf. KrV, B480-481. Cf. Dubislav (1931, p. 222), who comments that for Kant, the

dissolution of the antinomies can only be achieved by transcendental idealism, and at the

same time indirectly grounds the latter. Cf. also Scholz (1931, p. 222), who comments that

Přihonský’s analysis of Kant’s doctrine (pp. 124-131) of the antinomies is one of the best

interpretations of this difficult part of the Critique.353

Cf. KrV, B454-457.

New Anti-Kant 171

But nothing can arise in empty time, because no part of such a timeprecedes any other with the distinctive condition of nonexistence. So aseries of things can begin in the world, but the world itself cannot have abeginning and hence is infinite with regard to time. – In addition, theworld cannot have a boundary in space. But, supposing that the world isfinite and bounded by space, then it follows that it exists in an empty

space which is unlimited. Therefore there should be a relation not onlybetween things in space, but between things and space. However, such arelation, and thus a boundary of the world by empty space, is nothing, sothe world is not bounded by space.354

2. On the second thesis. Every composite thing consists of simpleparts. For supposing that composite things do not consist of simple parts,[126] then, if all composition were removed in thought, no compositepart and, since there are no simple parts, no simple part and thus nothingat all would remain.355

On its antithesis. No composite thing consists of simple parts. Forevery part of the composite must occupy a space. However, space doesnot consist of simple parts but of spaces. Thus the simple, if there weresuch, would occupy a space. But then the simple would be a composite.356

3. On the third thesis. If there is a causality according to the laws ofnature or necessity, or if everything happens according to the laws ofnature, then there is only a subordinate but never a first beginning, andthus no completeness of the series on the side of the causes descendingone from another, in that every present state of things can only be theeffect of a previous state. So the claim that there is only one causalityaccording to the laws of nature is self-contradictory, and therefore theremust also be another causality, namely, a ground of causes, which hasabsolute spontaneity and begins from itself a series of appearances thatruns according to natural laws; hence there must be a transcendental

freedom.357

On its antithesis. There is no freedom, [127] but everything happenssolely according to the laws of nature. For suppose there were a freedomin the transcendental sense, as a special kind of causality according towhich the occurrences of the world could follow, namely, a faculty ofabsolutely beginning a state, and thus also a series of its consequences:then not only will a series begin through this spontaneity but thedetermination of this spontaneity itself to produce the series, that is, its

354

Cf. KrV, B459-461.355

Cf. KrV, B462.356

Cf. KrV, B463. Kant says “substantial composite”.357

Cf. KrV, B472-474.


causality, will begin absolutely, so that nothing precedes it throughwhich this occurring action is determined according to constant laws. Butthis contradicts the law of causal connection.358

4. On the fourth thesis. The world of sense is a series of changes, thatis, of conditioned things. Any conditioned thing presupposes, with regardto its existence, a complete series of conditions up to the absolutelyunconditioned, which alone is absolutely necessary. Thus there mustexist something absolutely necessary. This necessary something,however, belongs to the world of sense. For supposing it were outsidethat world, then the series of changes in the world would derive itsbeginning from a necessary cause itself belonging to the world of sense.But this is impossible. For since the beginning of a time series can be

determined only through what precedes it in time, the supreme condition

of the beginning of a series of changes in the world must exist in a time

when the series [128] was not yet. Consequently, the necessary cause ofchanges belongs to time, hence to appearance, and cannot be thought ofas detached from the world of sense.359

On its antithesis. There is absolutely no necessary being existinganywhere, either in the world or outside the world. Either in the world.Suppose that either the world itself were a necessary being or that therewere a necessary being in it, then in the series of its changes, either therewould be a beginning that was unconditionally necessary and hencewithout cause, which is self-contradictory,360 or else the series itselfwould be without a beginning, and although conditioned in all its parts, itwould nevertheless be absolutely necessary and unconditioned as awhole, which is self-contradictory, because the existence of a multiplicitycannot be necessary if no single part of it possesses an existencenecessary in itself. Or outside the world. Suppose that there were anabsolutely necessary cause of the world outside the world, then thiscause, as the supreme member in the series of causes of changes in theworld, would first begin these changes and their series. But then it wouldhave to begin to act, and its causality would belong in time, and for thisreason in the collection of appearances, that is, in the world;

358

Cf. KrV, B473.359

Cf. KrV, B481-483.360

Kant writes: “a beginning which is [. . .] without a cause, which conflicts with the

dynamic law of the determination of all appearances in time; or else the series itself would

be without any beginning, and [. . .] it would nevertheless be absolutely necessary and

unconditioned as a whole, which is self-contradictory, because the existence of a

multiplicity cannot be necessary if no single part of it possesses an existence necessary in

itself.” (B481)

New Anti-Kant 173

consequently, the cause itself would not be outside the world, whichcontradicts the presupposition.361

Kant explains this highly remarkable appearance of the antinomies asfollows (p. 394).362

If the conditioned is given, then through it a regress [129] in theconditions for it up to the unconditioned is not always given, but it isgiven to us as a problem.363 The unconditioned is given along with theconditioned, if and only if both are things-in-themselves. But if we aredealing with appearances, the complete series of conditions is notnecessarily given with the conditioned, for this could only be given as asensible thing if the regress to the unconditioned were really achieved,but this is impossible if the series is infinite.364 – The major premiss ofthe transcendental inferences of reason takes the conditioned in thetranscendental sense of a thing-in-itself, and thus the unconditionedshould be given, as well. However, the minor premiss only mentionsappearances. So the mistake made here is really a so-called sophismafigurae dictionis. The contradiction in the antinomies is removed byshowing that the idea of absolute totality, which is valid only as acondition of things-in-themselves, is applied to appearances that existonly in representation.365 – Next, Kant presents us with the followingresolutions of the cosmological antinomies (pp. 407ff.).366

The characteristic of the first two resolutions, he says, is that the twocontradictory propositions are, to a certain extent, false, whereas thecharacteristic of the last two [130] is that the two contradictorypropositions are, to a certain extent, true, namely:

1. The idea of the totality of the composition of the appearances of aworld-whole (p. 407).367 The world has no beginning in time and noboundaries in space, but nonetheless it is not infinite. The first: theperception of such a beginning and such a boundary is impossible,because empty time and empty space cannot be perceived. The second:

361

Cf. KrV, B481-483.362

Cf. KrV, B525.363

Kant writes that “then through it a regress in the series of all conditions for it is given

to us as a problem” (KrV, B526).364

Cf. KrV, B526-527.365

Cf. KrV, B527-528. Kant says that the major premiss “takes the conditioned in the

transcendental sense of a pure category, while the minor premiss takes it in the empirical

sense of a concept of the understanding applied to mere appearances; consequently, there

is present in it that dialectical deception that is called sophisma figurae dictionis. This

deception is, however, not artificial but an entirely natural mistake of common reason”.366

Cf. KrV, B545.367

Cf. KrV, B545.


because an infinite magnitude cannot be the object of a possibleperception.368

2. The idea of the totality of division of a given whole in intuition

(p. 411).369 If I divide a whole given in intuition, then the possibility ofthe division must go to infinity; despite this I am not permitted to saythat the whole consists of infinitely many parts.

3. The idea of the totality of the derivation of occurrences in the

world from their causes (p. 416).370 One and the same effect can beconsidered as free with regard to its intelligible (i.e., its nonapparent)cause and yet at the same time, with regard to appearances, as their resultaccording to natural necessity. “For since appearances are not things assuch, they must be grounded by a transcendental object which determinesthem as mere [131] representations: so nothing prevents us fromascribing a causality to this object besides the property through which itappears; a causality which is not appearance, although its effect is seenin an appearance. This acting subject would have no temporaldeterminations, nor would it fall under the law of all temporaldeterminations. It could begin its effects in the sensible world by itself,and these, as appearances, would still have their empirical conditions inthe preceding time.” (pp. 421–422)371 Such is the case with human

368

Přihonský slightly mispresents Kant’s point. The latter writes: “I will also not say that

the regress from a given perception to everything bounding in it in a series, in space and

in past time, goes to infinity; for this presupposes the infinite magnitude of the world; nor

will I say that it is finite; for an absolute boundary is likewise empirically impossible.

Accordingly, I will be able to say nothing about the whole object of experience (the world

of sense) but only something about the rule in accord with which experience, suitably to

its object, is to be instituted and continued.” (KrV, B548) We notice that Kant refers to

experience, whereas Přihonský refers to perception. Unlike Přihonský, Kant distinguishes

between the two: experience is “the product of the senses and the understanding” (P, § 20;

cf. also my note 127). A perception is a representation which combines sensation and

consciousness, that “which is immediately represented, through sensation, as actual in

time and space” (B147; cf. also B207, B272).369

Cf. KrV, B551.370

Cf. KrV, B558.371

Although Přihonský marks this passage as a quote, it does not appear as such in the

Critique but roughly corresponds to excerpts from KrV, B573-574. In Kant’s view, we

may assume that natural causes are at the same time effects which have intelligible causes,

“as long as the action in the appearance of this cause accords with all the laws of

empirical causality. For in this way the acting subject, as causa phaenomenon, would have

all its actions linked with inseparable dependence to the natural chain of causes, and only

the phaenomenon of this subject (with all its causality in appearance) would contain

certain conditions that, if one would ascend from empirical objects to transcendental ones,

would have to be regarded as merely intelligible. This intelligible ground does not touch

the empirical questions at all but may have to do merely with thinking in the pure

New Anti-Kant 175

beings. That reason has a special causality is clear from the imperatives

of our reason. The ought expresses a possible action which is notgrounded in things of experience. Thus our actions result necessarily onthe one hand from empirical causes and on the other, from freedom (pp.426–427).372

4. The idea of the totality of dependence of appearances regarding

their existence in general (p. 434).373 All things in the world of sense arecontingent, but the whole series has a nonempirical condition; thus thereexists an unconditionally necessary being, not as a member of the worldof sense but as an intelligible being (p. 435).374

Comment. 1. It is certainly perplexing that our human [132] cognitivefaculty is supposedly equipped in such a way that when we are thinking,which we do not do arbitrarily but consider a necessary and peremptorytask in the development of our mental forces, we should fall into four

self-contradictory propositions.2. Surely we do not need to call attention to the contrived manner in

which the four cosmological ideas are derived from the table ofcategories. Who could guess that the fourth idea arises from the conceptof contingency and not of dependency or causality? Or that the second

idea belongs to the categories of quality (reality, negation, andlimitation)?

3. Quite a few people will find mistakes of a very different kind in theproofs Kant made up for his antinomies, even those who do not wish todeny our reason the right to judge nonsensory things.

understanding; and, although the effects of this thinking and acting of the pure

understanding are encountered among appearances, these must nonetheless be able to be

explained perfectly from their causes in appearance, in accord with natural laws, by

following its merely empirical character as the supreme ground of explanation; and the

intelligible character, which is the transcendental cause of the latter, is [. . .] indicated

through the empirical character as only its sensible sign.”372

Přihonský probably means to indicate p. 426. Cf. KrV, B575. Kant says that the

causality of reason (to which the dynamical law of nature is not applicable) “is the

persisting condition of all voluntary actions under which the human being appears.” It is

the intelligible character of the human being to which no before or after applies, and every

action “is the immediate effect of the intelligible character of pure reason.” (B585) In this

sense, it is a causality through freedom (from empirical conditions). But “freedom is

treated here only as a transcendental idea through which reason thinks of the series of

conditions in appearance starting absolutely through what is sensibly unconditioned”, and

it involves itself in an antinomy by following its own laws, “which it prescribes for the

empirical use of the understanding.” Kant’s point is that this antinomy can be resolved by

showing that nature “does not conflict with causality through freedom” (B586).373

Cf. KrV, B587.374

Cf. KrV, B588.


a) In the proof of the first thesis

α) Kant tacitly presupposes that a collection cannot exist unless there issomeone who has counted through it or who could count through it

(cf. p. 346, note).375 If this were so, then it would indeed beunquestionable that an infinite collection cannot exist anywhere,for an infinite collection is precisely that which is bigger thanevery number and therefore cannot be counted.[133] But why should it belong to the possibility of a certaincollection that there is someone who counts through it or evenindividually represents all units that compose it? Does thisassertion not define all irrational relations as impossible? Couldthere be, for example, only one square, since every squarerepresents an infinite collection of fractions of its diagonal, knownas the magnitude √2? We can see that Kant’s assertion revokes thewhole of mathematics.

β) In addition, Kant wrongly presupposes that there can be only series in

time and that the infinity of a series consists only in the fact that itcan never be completed through a successive synthesis of its

members.376 The concept of a series alone by no means includesthe concept of time, and there can be infinitely many finite as wellas infinite series which have nothing to do with time at all. Of thiskind is the series in which both sides extend to infinity, . . . –4, –3,

375

Cf. KrV, B456. Kant writes: “The concept of a totality is in this case nothing other than

the representation of the completed synthesis of its parts, because, since we cannot draw

the concept from an intuition of the whole (which is impossible in this case), we can grasp

it, at least in the idea, only through the synthesis of the parts up to their completion in the

infinite.” This note accompanies a passage where Kant says that we cannot have an

intuition of an infinite whole of simultaneously existing things (cf. KrV, B455), and he

comments that in this case we cannot construct or draw the concept of the totality from

intuitions but can only grasp the idea of this totality through the completed addition of

units to each other (i.e., the synthesis of the parts). It seems that the underlying problem is

the difference between Přihonský’s (and Bolzano’s) notions of addition (based on the law

of association (a + b) + c = a + (b + c)) and collection or sum (based on the law of

commutativity or the indifferent order of the members of a collection) (cf. WLI, § 84;

WLIII, §§ 305.5, 315.3), and the Kantian notions of addition and sum, which are synthetic

and based on the “successive synthesis of part to part in the process of its apprehension”

(KrV, B204). Kant distinguishes between a connection as such, or the addition of 7 and 5

(which is a synthesis), and an associative connection, which is not a synthesis (cf. KrV,

B15, B205-206), even though he falsely grounds this distinction on the sign +. Cf. Benno

Kerry (1886), who criticizes Bolzano for wrongly replacing Kant’s synthetic notion of

addition by an associative notion, in: “Ueber Anschauung und ihre psychischeVerarbeitung”, 2, pp. 438-440.376

Cf. KrV, B480-481; cf. also NAK, p. 123, above.

New Anti-Kant 177

–2, –1, 0, +1, +2, +3, +4, . . . (cf. Theory of Science, §§ 85 and 87,note).377

χ) The assertion that infinite time cannot be a past time is so completelyfalse that we should rather claim that an infinite time has passedbefore each moment. For, as every mathematician knows, in everymoment, the whole of time can be divided into two parts, [134]

one of which is past, whereas the other one is future, but both ofthem are completely identical in all their inner properties. So if wewanted to assert that no infinite time has passed, we should alsoassert that there is no infinite time in the future.

δ) Finally, the whole thesis is as wrong as Kant’s proof for it. However,we find the antithesis that the world cannot have a beginning intime or a boundary in space, entirely right, as well as provable,although on different grounds than the ones given by Kant, whichare based on the false assumption of an empty time and an emptyspace.

b) The thesis of the second antinomy—“every composite substanceconsists of simple parts, and nothing exists anywhere except thesimple or what is composed of simples”378—seems to us to be true,but the antithesis is false: “no composite consists of simpleparts.”379 Kant infers the latter from the claim that space does not

377

In Kant’s view, the “true (transcendental) concept of infinity is that the successive

synthesis of unity in the traversal of a quantum can never be completed.” (KrV, B460)

The young Bolzano, however, even denies the existence of infinite series. In Derbinomische Lehrsatz, § 2, he defines a series as the sum of terms which is formed

according to a determinate law, adding that if the series is infinite, the number of its terms

can be increased arbitrarily. In his later work, e.g., the Paradoxes of the Infinite, he gives

a quantitative explanation of the concept [infinite], by saying that it applies to collections

(i.e., things which exhibit magnitude or multitude); cf. PU, § 2. Yet when defining “a

truly infinite quantity”, Bolzano gives the example of a straight line unbounded in either

direction, i.e., “the magnitude of the spatial entity containing all points determined solely

by their abstractly conceivable relation to two fixed points” (PU, § 11). This example does

not seem very different from Kant’s. Cf. also WLIII, § 315.7, where Bolzano rejects the

proof that the world has a beginning in time as contradictory: “the infinity of a sequence

does not consist in the fact that it cannot be completed through a successive synthesis [. .

.]. The concept of time does not belong to the concept of an infinite sequence at all, since

there are sequences of things, finite as well as infinite, which are not in time at all. [. . .]

But if there are infinite sequences of numbers and magnitudes in general, then there must

also be infinite sequences in time, indeed in both temporal directions [. . .].” Cf. also

Scholz (1931, pp. 222-223).378

Cf. KrV, B462.379

Cf. KrV, B463.


consist of simple parts but only of spaces, that is, of a manifold ofelements external to one another.380 This is false, if we take it inthe sense in which it must be taken here, for there are two parts ofcomplex things: equal parts, or those which belong to the conceptto which the whole refers, and unequal parts.381 If, for example, wecrush a piece of saltpeter, [135] we have equal parts. But if, fromsaltpeter, we produce hydrochloric acid and potash, thecombination of which saltpeter consists, then we obtain unequalparts. Likewise, when we are talking about parts which are equalto a whole, there is no doubt that all of space (everythingextended) consists of mere spaces (extensions), a line consists oflines, an area consists of areas, and so on. But if we are talking ofparts in general, such that we also refer to unequal parts, then wemust admit that every space, regardless of whether it is a line or anarea or a body, also consists of simple parts, namely, of points; forspace really is nothing but a collection of points.382 In order tomake this point clearer, we note that all mathematicians teach thatthere are points in every line, area, and body, which means nomore and no less than that points constitute their parts. Still, onecould object that they are not component parts (partes integrantes),since it is impossible to produce a line, area, or body from points,however many there may be. Our reply is, certainly, if there isonly a finite collection of points; nor do we deny that not everyinfinite collection of points forms an extension, for an extension isonly formed by a collection of points such that, at every [136]

point, there are, for every (even the smallest) distance, one orseveral points which really have this distance. This is the onlysystem of points which is a true continuum,*383 regardless of

380

I have added “of elements”, conforming to KrV, B463, because Přihonský’s “external to

a manifold outside one another” does not make sense, and the omission is probably an

oversight.381

Cf. WLI, § 87.382

Cf. NAK, p. 58. In Bolzano’s view, space is a collection of possible locations; cf. WLI,

§ 79.6, and my note 177. Cf. also WLIII, § 315.7.c.

*383

The concept of a continuity is the concept of a property of a spatial object such that it

has a consequent according to which each point has a neighbouring point for each

sufficiently short distance, or in other words, each point has so many neighbours that none

of them is its nearest neighbour. For such a proof of the impossibility of nearest points in

space and thus of spatial continuity, cf. Leibniz’s Monadology, by Dr. Robert

Zimmermann (Vienna, 1847), p. 167.

But cf. Scholz (1931, p. 223) who comments on Přihonský’s note that this continuity

concept is insufficient for analysis if we require that one should speak of a continuum

only when speaking of completeness (i.e., when there are no gaps).

New Anti-Kant 179

whether it is a line, an area, or a body. So whoever talks aboutparts in general without referring to a particular kind, that is, equalparts, is mistaken when he claims that space consists of infinitelymany parts and not of simple parts.384

c) The theses of the third antinomy are expressed in an extremely obscureway. According to the proof of the antithesis, “by freedom weunderstand a faculty of absolutely beginning a state and thus aseries of consequences, so that nothing precedes it through whichthis occurring action is determined in accord with constantlaws.”385 Here we can only wonder how [137] such a faculty can becalled causality, since the latter word refers only to a causal nexuswhich does not occur here. In addition, it is presupposed that everycause precedes its consequence, which is false, as we have alreadyshown. Even more importantly, the definition defines somethingcompletely different from that which is understood by the usage ofthe word freedom. For we always consider it as a certain propertyof our voluntary faculty. But if we take the concept in the sensewhich is closest to this definition (that is, in the sense ofindeterminism), then freedom is the possibility of a voluntarydecision without a determining ground, so that the same decisioncould also have ceased in the same circumstances. In this sense386

it can be said that every free decision of the will begins a series ofconsequences, but we cannot claim that freedom occurs where aseries of consequences begins absolutely. A series of consequencesbegins absolutely wherever a cause is acting which is lacking anyfurther causes, such as the highest being, or God, to whom we cancertainly not attribute freedom in the indeterminist sense. – In theproof of the thesis, it is asserted that “there is no othercompleteness of the series on the side of the causes descending

384

In Kant’s view, the intuition of space is singular but not simple; cf. KrV, A99; JL, § 1.

On Bolzano’s problematic claim that intuitions are simple representations, cf. my note

161; NAK, p. 49. The general concept of space, on the other hand, consists of infinitely

many parts because it is constructed through intuitions by a procedure of limitations; cf.

my note 163; NAK, p. 52; KrV, B39-40; where Kant discusses space as an infinite given

magnitude (Přihonský refers to this passage in the NAK, pp. 63-64). Cf. also Přihonský’s

and Bolzano’s objections to Kant’s notion of space; NAK, pp. 66-67, and my notes 190

and 192.385

Cf. KrV, B473. Přihonský omits part of the sentence: “[. . .], a series of its

consequences, then not only will a series begin absolutely through this spontaneity, but

the determination of this spontaneity itself to produce the series, i.e., its causality, will

begin absolutely, so that nothing precedes it [. . .]”.386

Přihonský uses Bedeutung, whereas before he used Sinn.


from one another, if everything happens according to the laws ofnature”,387 that is, if everything that [138] happens has a causewhich is not itself something that happens. However, from thishypothesis it merely follows that the series of events must beinfinite but not that it is incomplete and cannot really exist, for aswe have pointed out above, infinite series can also exist and haveceased to exist. – Further, if there is an infinite series of causes inthe narrowest sense, then it occurs simultaneously with itsconsequences. If the members are conditions rather than causes,then every preceding member must be a condition of thesucceeding one, but infinite time is in no way necessary for thecourse they take. Every movement through which a body travels,even the smallest distance, is an example of a series consisting ofinfinitely many members, where the preceding member conditionsthe succeeding one, even though the entire period of time duringwhich all members are moving can be very short.

d) Finally, the proof of the fourth antinomy falsely presupposes that theseries of occurrences in the world has a beginning in time if it isgrounded in something else.388 Instead, it should have beenpresupposed that the substances of the world exist at all times butonly as a consequence of a substance which has unconditionalreality. – The conclusion is also hastily drawn that a substanceacting in time belongs to appearances and thus [139] to the worldof appearances. No simple substance can be considered as asensory thing and even less so the substance to which creation isattributed. Nonetheless substances act in time, even though theyare nonsensory, for example, our souls. Even God can act in time,even though he is not an object of the senses.389

387

Cf. KrV, B472. Kant writes: “if everything happens to mere laws of nature”.388

Cf. KrV, B482.389

This is precisely Kant’s point when he discusses the causality of freedom: the ground of

intelligible determination is not reduced to the causality of nature: God can act in time,

but he is not qualified by time. The fourth antinomy opposes two theses concerning the

modal limits of the world: does an absolutely necessary being belong to the world as its

cause, or not? In Kant’s view, both theses are dialectical (as are the other opposed

arguments; cf. KrV, B534), and the antinomy is removed by showing that the conflict is

due to an illusion arising from the fact that reason has mistakenly applied ideas valid only

of things-in-themselves, to appearances (ibid.).

As for the sophisma figurae dictionis, Kant says that this is a natural mistake of common

reason, which (in the major premiss) presupposes that the conditioned and the condition

are given simultaneously and (in the minor premiss) regards appearances as things-in-

themselves “given to the mere understanding”, and thus mixes rational and empirical

New Anti-Kant 181

4. Although these proofs contain many substantial mistakes, they are theonly kind of fallacia figurae dictionis accepted by Kant. They consist infreely talking of things sometimes as mere appearances, sometimes asthings-in-themselves. But things as mere appearances are not mentionedin any of these antinomies,390 neither in the propositions nor in theproofs.391 Appearances are mentioned only in propositions expressingrelations between things and our sensory organs, such as: sugar is sweet,and so on.392 Where do propositions of this kind occur in those proofs?

5. This is why we think that Kant’s solutions to this fourfold antinomyare not satisfactory.

a) As we have already mentioned, we can accept only the proposition thatthe world has no beginning in time and no bounds in space, but wemust reject the other proposition that nonetheless the world and itsgiven ground cannot be infinite, because an infinite magnitude

[140] cannot be the object of a possible perception. For theassertion that something cannot exist if it is not perceived isfalse.393 In addition, Kant contradicts himself because of theregulative principle of reason on which he bases the assertion that“in the empirical regress no experience of an absolute boundary

can be encountered” (p. 407).394 How does he intend to prove thisimpossibility if not from pure conceptual truths which have amore general application than to experience alone and which areapplied unconditionally?

concepts. For the “synthesis of the conditioned with the condition [. . .] carries with it no

limitation through time”, whereas the empirical synthesis is necessarily “given only in

time, one member after another” (B528).390

Denn in the text, though the sense of the sentence corresponds to “but”.391

There seems to be a terminological misunderstanding on Přihonský’s part. Kant refers

to things-in-themselves (noumena) and appearances (phaenomena), but he does not have

the notion of “things as appearances” (Dinge als Erscheinungen). Cf. my notes 293-294,

above. For Kant, a thing is a substance with an undetermined existence and which cannot

be known, as opposed to an object (Gegenstand); cf. KrV, B279.392

Here the syntax is awkward: the sentence begins, “Nur solche Sätze [. . .]” and ends

with the passive “wird von den Dingen als Erscheinungen gesprochen.”393

Esse est percipi et percipere is Berkeley’s claim, not Kant’s. In addition, it does not

follow from the proposition “an infinite magnitude cannot be the object of a possible

perception” that this infinite magnitude can exist if and only if we perceive it. Kant says

that “we can intuit an indeterminate quantum as a whole, if it is enclosed within

boundaries, without needing to construct its totality through measurements, i.e., through

the successive synthesis of its parts.” (KrV, B454) Cf. my notes 260, 264, and 281, above,

on Kant’s notion of magnitudes and the construction of mathematical concepts through

intuitions; cf. also notes 32, 109-112, and 281 on Kant’s concept of infinity.394

Cf. KrV, B545.


b) Regarding the second antinomy we wish to assert on the contrary thatnot every whole given in intuition is infinitely divisible, at leastnot if we are talking of a possibility in all respects (that is, also inconsideration of our limited forces and instruments). But throughreasons395 we can ascertain a priori that every body big enough to

cause sensible intuitions in us is composed of an infinite collection

of simple parts. To object to this, as Kant does, that such a wholeis divisible to infinity and that even so the whole does not consistof infinitely many parts, is a remarkable contradiction.396 For if thecollection of parts is only finite, then the business of [141]

division must come to an end after a determinate number ofrepetitions.

c) Nothing could be more unfortunate than Kant’s attempted unificationof freedom and natural necessity by distinguishing between man asan appearance and as an intelligible being.397 That which is initself without cause, that is, in fact and in truth, can onlymistakenly appear as produced by a cause.*398 We should think thatany unbiased person ought to sense immediately what is wrong

395

Gründe.396

Cf. KrV, B552.397

In the third antinomy, Kant distinguishes between natural and free causality. From a

theoretical point of view, freedom is “the power of beginning a state spontaneously”

(KrV, B561), undetermined by an antecedent cause according to the law of natural

causality. From a practical point of view, freedom is “the independence of the power of

choice from necessitation by impulses of sensibility” (B562). Here Kant sets the basis of

his practical philosophy, according to which the human will is a kind of causality

belonging to humans insofar as they are rational. The will is autonomous, i.e., it can be a

law unto itself; cf. Grounds for the Metaphysics of Morals, third section, pp. 446-447. In

the first Critique, he argues that an acting subject has an empirical character (through

which its actions, as appearances, are connected with other appearances according to

natural laws) as well as an intelligible character (through which it is the cause of those

appearances, as a noumenon). Such a subject thus has an empirical and an intelligible

causality, both of which apply to the same effect; cf. KrV, B566-567. Every human being,

as an appearance, stands under empirical laws and thus has no freedom with regard to his

empirical character. However, every human being also has an intelligible character, i.e.,

the faculty of reason which has (free) causality and conditions our voluntary actions by

determining them through grounds of reason; cf. KrV, B574-578ff. But cf. Bolzano’s

objection to Kant’s practical postulation, WLIII, § 315.8.

*398

Kant claims that it is possible “that one and the same appearance is the effect of an

intelligible cause without the least interruption of its connection with natural causes”

(p. 420) (KrV, B572). But on which intuition does this evidently synthetic judgment rest?

And how can we rightfully apply the category of possibility to something that is certainly

not an object of a possible experience?

New Anti-Kant 183

with this thory.399 If a significant number of German scholars havenonetheless adhered to it, this only proves that often we humansgradually get used to a very wrong way of definition, and in theend we even persuade ourselves that things do indeed stand thisway. – In addition, the assumption of the existence of intelligible

things according to the principles of critical philosophy iscompletely unjustified. For how else can we infer [142] theexistence of certain things-in-themselves grounding appearances ifnot by inferring a cause from the effect? And yet it says here thatthis inference is only valid in the world of appearances.400

399

Kant could reply that principles of the understanding which anticipate experience are apriori synthetic modes of knowledge because they make experience possible by

determining how an appearance will manifest itself to our sensible intuition. Cf. also my

note 289 and Přihonský’s discussion in the NAK, pp. 93ff., above, as well as my notes

252 and 285 on the authors’ respective conceptions of possibility. Kant would reply to

Přihonský that we cannot have a sensible intuition of nonsensory objects.

Přihonský’s critique is probably based on Bolzano’s objection to Kant’s doctrine that we

cannot make any synthetic judgment about nonsensory objects, i.e., God, the soul, and

immortality, or other objects that we cannot perceive. But if this is so, Přihonský

misunderstood both Kant and Bolzano, since their point of disagreement is, precisely, that

we cannot make synthetic judgments about objects we cannot perceive, i.e., which are not

given in sensible intuition. In Bolzano’s view, we can make true conceptual synthetic

propositions about metaphysical objects, such as “there is a God”, and in Kant’s view, we

must recognize these objects as ideas of reason which are regulative and not constitutive

principles; cf. KrV, B537. We can make speculative judgments about these ideas, but they

cannot be the objects of a possible experience (hence they cannot have a corresponding

intuition), and by treating them as such, reason mistakenly frees itself from the limits of

experience; cf. KrV, B382-385. Bolzano and Přihonský claim against Kant that not all

existential propositions are synthetic (cf. WLII, § 142; NAK, pp. 147-148; and my note

417), but concerning the proposition “God exists”, we could argue that Kant and Bolzano

would accept that such existential assertions are synthetic: both authors would agree with

Frege (cf. Foundations of Arithmetic, 1953, pp. 59, 65) that existence is not a real

predicate, because existential propositions express relations between concepts and not

between individuals, so when asserting that God exists, we do not predicate a real

property of God but only assert that “there is an x such as (Fx)”. In other words, existence

is a second-order predicate or property of the concept [God], which Kant calls a “logical

predicate”; cf. KrV, B626. In Bolzano-Přihonský’s view, “God exists” is a synthetic

proposition because it has no component which can be varied without modifying the truth

or falsity of the proposition. In Kant’s view, we cannot attribute existence to God because

he is not the object of a possible experience; but the proposition “there is an x such as

God” is synthetic. Cf. also E. Morscher (1974), “Ist Existenz ein Prädikat?”, Zeitschriftfür philosophische Forschung, pp. 120-132.400

Kant does not infer the existence of certain noumena from their effects; instead, he says

that the empirical character is the supreme ground of explanation, and the intelligible

character is “entirely unknown, except insofar as it is indicated by the empirical character

as only its sensible sign.” (KrV, B574) This sensible sign is the ought of reason’s practical


d) This same inference, which is unjustified by the principles of theCritique, is also made in presupposing the existence of God as anintelligible being that is the absolute condition of the existence ofthe whole world of sense.401 In addition, we think of God not onlyas the cause of the world of sense but also as the cause ofintelligible beings or things-in-themselves.

As Kant puts it, he understands by an ideal “an idea not merely inconcreto, but in individuo, that is, as an individual thing which isdeterminable, or even determined, through the idea alone.” (p. 440)402

Presumably he wants to say: the representation of an individual which isdeterminable a priori through mere concepts according to a rule. Butunclarity is Kant’s very unfortunate mistake.

Our philosopher asserts that the only true ideal of which humanreason is capable “is the concept of a thing that unifies all of reality,omnitudo realitas, God in the transcendental understanding” (pp. 445–446).403 We strongly doubt whether God can be defined as a being whichunifies all realities, especially [143] insofar as we understand by realities(as Kant seems to do) all those properties which can be represented by apositive concept (not including a negation). At least this much is certain,that there are quite a few properties of God which may be represented bya negative concept. Thus the one property of God from which all the restcan be most easily cognized is obviously negative, that his existence hasno ground. In our view, Bolzano proceeds more rightly and prudently inthe Theory of Religion (vol. 1, § 74), by replacing the concept of the most

real being with the concept of the most perfect being and by taking thelatter (as we already noted) as the one property from which it follows that

imperatives, which “expresses a possible action, the ground of which is nothing other than

a mere concept, whereas the ground of a merely natural action must always be an

appearance. Now of course the action must be possible under natural conditions if the

ought is directed to it, but these natural conditions do not concern the determinations of

the power of choice itself but only its effect and result in appearance.” (B576) Cf. also

Bolzano’s objection to Kant’s doctrine of duty, WLIII, § 315.8.401

Cf. KrV, B587.402

Cf. KrV, B596.403

Cf. KrV, B603. Kant explains that God (as the transcendental ideal) is the ensrealissimus, or the individual being whose existence is adequate to the idea of the being of

beings. That is, God is the only true ideal which satisfies the condition of being

determined in and through itself and known as the representation of an individual. Kant

does not attempt to prove that God is a transcendental ideal, since in his view, God must

be presupposed by reason, not as the “existence of a being conforming to the ideal, but

only the idea of such a being, in order to derive from an unconditioned totality of

thoroughgoing determination the conditioned totality” (A578-579/606-607), i.e., the being

to which all beings can ultimately be referred.

New Anti-Kant 185

God possesses all forces which can coexist and possesses them to thehighest degree in which they can coexist.

Kant proceeds from the concept of God to “the grounds of proof of

speculative reason for inferring the existence of a highest being” andsubjects it to his critique (p. 451).404 Without proving it, he claims thatthere are only three proofs for the existence of God: the ontological

proof, which infers the existence of a highest cause entirely a priori405

from mere concepts; the cosmological proof, which only [144] grounds iton some existence; and the physicotheological proof, which sets out fromthe special constitution of our world of sense.406 He wants to prove thatall three proofs are untenable.

The impossibility of an ontological proof of God’s existence

(p. 456)407 supposedly results from the following inference. “Somethingis necessary that follows from mere concepts or whose nonbeing isimpossible, that is, self-contradictory.”408 “Thus, for example, it isnecessary that a triangle has three angles. But this and similar examplesare taken only from necessary judgments and not from things.409 Theunconditioned necessity of a judgment is only the conditioned necessityof the predicate, namely, that under the condition that a triangle exists, itnecessarily has three angles. But no contradiction arises if we cancel thesubject. It is the same with the concept of an absolutely necessary

being.410 Surely it is contradictory to say: ‘the absolutely necessary beingis not omnipotent’, but it is not a contradiction to say: ‘the absolutelynecessary being is not’.”411 – If someone should want to say that hethinks of God as the most real being and that the concept of such a beingalready has reality, so we cannot negate it without contradiction, Kantreplies

404

Cf. KrV, B661.405

Cf. KrV, B618.406

I follow Guyer and Wood’s translation of Beschaffenheit.407

Cf. KrV, B620.408

Cf. KrV, B620. The entire passage is a summary of Kant’s text, rather than a quote.409

Cf. KrV, B621.410

Cf. KrV, B622.411

Cf. KrV, B623, where Kant writes: “Omnipotence cannot be cancelled if you posit a

divinity, i.e., an infinite being which is identical with that concept. But if you say God isnot, then neither omnipotence nor any other of his predicates is given; for they are all

cancelled together with the subject, and in this thought not the least contradiction shows

itself.”


“that it is already a contradiction to bring the concept of existence into the

concept of a thing.412 [145] Every existential proposition is synthetic. [. . .] Being

is not a real predicate of a thing but merely the positing of a thing which adds no

new property to it. In the proposition “God exists”, I add no new predicate to the

concept of God. A hundred real dollars do not contain the least bit more than a

hundred possible ones; otherwise they should not be dollars. [. . .] So whatever

our concept of an object may contain, we must still go beyond it to provide it with

existence. Lastly, not even the possibility of God follows from the mere statement

that realities (positions) do not contain a contradiction, for we must also ask

whether the connection (synthesis) of these realities is possible; but this cannot be

established for an object that lies beyond all experience.” (pp. 456–464)413

412

Cf. KrV, B624-625. In Kant’s view, we have produced a contradiction (and a tautology)

by bringing the concept of existence into the concept of a thing “which you would think

merely in terms of its possibility.” (B625) For if we presuppose existence as belonging to

possibility and then infer existence from inner possibility, we have committed a tautology,

largely due to our failure to distinguish between existence and reality. “For if you call

positing [. . .] “reality”, then you have already posited the thing with all its predicates in

the concept of the subject and assumed it to be actual, and you only repeat that in the

predicate.” (ibid.) Kant distinguishes between existence and reality: existence (Sein) is the

way in which an object of possible experience comes into being through an act of thinking

(i.e., through the understanding), whereas reality (Wirklichkeit) is the property an object

of experience acquires by coming into being in time and space (in this way the object

becomes perceptible); cf. Kant’s second postulate of empirical thought, KrV, B266. But

the problem that seems to be at issue in the ontological argument (originally presented by

St. Anselm in the Proslogion; this argument reappeared in Descartes’ Fifth Meditation) is

God’s reality or actuality (Wirklichkeit) rather than his existence, and Kant has not shown

the impossibility of the ontological proof of God’s reality by showing that existence is not

a real predicate. In addition, existential propositions have a locative character, at least if

we analyse the verb “exist” as “being somewhere”. Cf. J. Barnes (1972), where he refers

to Locke’s use of the verb “to exist” as “being not something but somewhere (and hence

somewhen).” (The Ontological Argument, pp. 63-65)413

Přihonský summarises the passage dealing with existence; cf. KrV, B628-630. Kant’s

point is that existence is not a real predicate, so we do not add anything to a thing by

positing that it is. Existence is a second-order predicate or property of a concept; cf. my

note 412, above. Our awareness of the existence of an object is a different matter and must

be grounded in experience, since “for objects of pure thinking there is no means whatever

for cognizing their existence, because it would have to be cognized entirely a priori, but

our consciousness of all existence [. . .] belongs entirely and without exception to the

unity of experience [. . .].” (B629)

Kant’s last statement is directed against Leibniz’s attempt to prove the possibility of God

by showing that impossibility requires a contradiction between a reality and its negation

and that such a contradiction cannot occur in the case of God or the most real being; cf.

Philosophical Writings, trans. and ed. G. H. R. Parkinson (London: Dent, 1973), 4:295-

296; 7:261. Kant argues against Leibniz that the concept of a highest being cannot

generate a contradiction “because the mark of synthetic cognitions always has to be

sought only in experience, to which, however, the object of an idea can never belong-the

famous Leibniz was far from having achieved what he flattered himself he had done,

namely, gaining insight a priori into the possibility of such a sublime ideal being.” (B630)

New Anti-Kant 187

Comment. 1. When Kant defined absolute necessity as existence from

mere concepts, he came close to ascertaining a clear concept of necessity,if only he had been aware that nothing follows from concepts as such butonly from conceptual truths. Just as he defined the necessary as thatwhose nonexistence is self-contradictory, he should have rememberedthat contradiction can really only occur between dictions,414 that is,propositions, so that we should really say that an [146] object isnecessary if the proposition: “this object is not” contradicts a pureconceptual truth.

2. A judgment or proposition is called necessary only insofar as it is amere conceptual truth, such as the judgment that a triangle has threeangles.

3. But it is false that from the unconditioned necessity of a judgmentcan only follow a conditioned necessity of the predicate, namely, onlyunder the condition that the object is really there. If the subject-representation in a proposition has no object whatsoever, then the entireproposition is false.415 If there are no triangles, then the proposition thatall triangles have three angles is false.

4. In addition, it is false that a contradiction is not produced when thesubject of a judgment is cancelled, that is, when the proposition: thissubject does not exist, is put forward. So it is, for example, indeed acontradiction to say: there is no triangle or ‘there is no truth’, and so on.Kant says: “where then is the contradiction supposed to come from?Externally there is nothing that would contradict it, for the thing is notsupposed to be externally necessary. Nothing internally either, for bycancelling the thing itself, you have at the same time cancelledeverything internal.” (p. 458)416 – Here it is clear that Kant does not havea right notion of what a contradiction really consists in.[147] Thecontradiction we commit by cancelling a truly existing thing A arises

414

“Diction” translates Spruch, but both expressions refer to words rather than statements.415

Přihonský bases his claim on Bolzano’s objectuality constraint; cf. WLII, § 148, and my

notes 141 and 142, above. Cf. also WLI, §§ 25, 30. A proposition is true iff A really has band does not depend on human thought. Bolzano claims that the canonical form of

propositions is “A has b” and that all propositions are either true or false. This claim is

problematic; cf. Lukasiewicz (1913), who criticizes Bolzano’s failure to accept

indeterminate propositions in Die logischen Grundlagen der Wahrscheinlichkeits-rechnung, pp. 67-69. Cf. also Dubislav (1931, p. 224), who comments that Přihonský’s

assertion is valid only if we agree with Bolzano that an “all” assertion is applicable only

once we have established that the proposition is objectual. In Logicism the proposition

“all triangles have three angles” is interpreted as follows: “it is the case for all x: ‘x is a

triangle’ implies ‘x has three angles’”, and the statement in quotation marks is true if the

sentential function ‘x is a triangle’ is “always false”.416

Cf. KrV, B623.


neither from outside nor from inside, that is, by means of the proposition:A is not or ‘the representation: A has no objectuality’, since thiscontradiction arises merely because a truth of the form: A is, or ‘the

representation: A has objectuality’, exists.5. So the proposition: there is no God, is contradictory indeed. It

contradicts the truth that there is a God.

6. Those who confer reality onto the concept of God and then want toinfer from it that God necessarily has reality, would obviously draw afalse conclusion, for we cannot immediately infer from the fact that acertain property a is thought in the subject-concept M = something whichhas the properties a, b, c, . . . that the proposition: M has a is true. For tothis inference also belongs the premiss: the representation of somethingwhich has the property a, b, c, . . . is an objectual representation. HenceKant claims with good reason that we cannot infer the existence of Godfrom the fact that we already have the concept of reality in the concept ofGod.

7. However, he went too far in asserting that we already commit acontradiction whenever we bring existence into the concept of a thing.[148] On the contrary, there are countless real concepts containing theconcept of reality, even that of substance, that is, something real which isnot an attribute, such as the concepts: man, animal, plant, body, and soon.

8. So it is false that every existential proposition is synthetic, if wetake it as a proposition of the form: A has reality or ‘A is’.417 Indeed,usually when we state the existence of an object and pronounce theproposition: there is an A, we actually have in mind the following

417

Prihonksy and Bolzano argue against Kant that some existential propositions are

analytic (cf. WLII, § 142, and my note 403, NAK pt. 2, above). Cf. E. Morscher (1997):

“Bolzano’s Method of Variation: Three Puzzles”, pp. 139-166. Morscher points out that

Přihonský suggests replacing the representations [angel], [man], [animal], and [plant] “in

order to test the analyticity of a sentence” (ibid., p. 157). Nonetheless, Morscher admits

that Bolzano would agree with Kant that existence is a second-order predicate (ibid., pp.

159-160); cf. also his earlier paper “Ist Existenz ein Prädikat?” (1974), pp. 120-132, and

his analysis of existential propositions in his introduction to the new edition of Přihonský

(2003), p. LXIV. Cf. also Scholz (1931, pp. 224-225), who comments that Kant’s main

question, whether it makes sense to speak of existence as a property of individuals (and

not only as a property of properties), has not yet been clarified. Modern logic usually

denies this (usually with Kant), at the price of a concept of existence which does not

imply having to name existence, i.e., the very thing we are used to in assertions about

existence outside mathematical usage.

New Anti-Kant 189

proposition: the representation ‘A has objectuality’; and this propositionis certainly synthetic.

9. In addition, it is false that being or reality are not predicates of athing but merely its position. Reality is as much an attribute of certainthings, for example, of man, animals, and so on, as nonreality is anattribute of certain others, for example, truths as such. We say: Caius hasreality, just as rightly as we say: a proposition as such—has—no reality.

10. It is true that this reality is not always a new property which is notthought in the subject-representation, as in the proposition: “man issomething real”, since reality is already contained in the concept man.But it does not follow from this that such propositions are always true,and even less that they must always be false, [149] for whether they aretrue or false depends on whether the subject-representation is objectual ornot.

11. But in what sense is it supposed to be true that a hundred realdollars do not contain the least bit more than a hundred possibledollars?418 A hundred real dollars are something quite different than ahundred merely possible dollars; the former are something that exists,something corporeal, and so on, which the latter are surely not. Butperhaps Kant merely wanted to say that the representation of a hundredreal dollars does not contain anything more and nothing other than therepresentation of a hundred possible dollars. But this is not true either,precisely because the objects of these two representations are sodifferent, for a hundred real dollars are certainly not the same as ahundred merely possible dollars; their representations must also bedifferent, because the same representations also have the same objects.

12. The locution that we must go out beyond the concept of an objectto provide it with existence is equally strange, even if (as is no doubtintended here) only to ascertain that it has existence. This is true only ifit is supposed to mean that we cannot infer from the mere form of aconcept (see 6 above) whether an object corresponds to it or not.

13. We certainly accept that we cannot infer that the concept of abeing which unites [150] all realities in itself does not contain a

418

Here Přihonský is right, for Kant’s example does not serve to illustrate his point: if my

bank statement says that there are a hundred dollars in my account, (unfortunately) this

sum is not increased as I withdraw it, so existence certainly does not add anything in this

sense. But Kant continues by saying that my financial condition is better if I have a

hundred dollars than if I merely have a concept of them (i.e., of their possibility), so he

admits that a hundred dollars which actually exist are more than a hundred possible

dollars that I am thinking of; cf. KrV, B627. Cf. also H. Scholz (1931, p. 224), who

comments that we should begin by asking whether it is at all possible to compare a

hundred thought dollars and a hundred real dollars.


contradiction, merely from the fact that realities do not contain anegation. But in our view, we can judge whether certain realities, forexample, understanding, will, and power, are compatible with each other,partly from experience and partly from mere concepts. And if the conceptof God’s omni-perfection is taken in the way we have suggested above,surely a sufficiently strict proof can be provided that there is also anobject corresponding to this concept, that is, an omni-perfect being.419

From the fact that such a being cannot become an object of experiencedoes not follow that we cannot ascertain the existence of this being. Wecan see that Kant’s entire reasoning is based once again on the falseproposition that we cannot make synthetic judgments about nonsensory

things.The cosmological proof of God’s existence is also deemed impossible

and supposedly goes as follows:

“if something exists, then an absolutely necessary being must also exist. Now I

myself, at least, exist; therefore an absolutely necessary being exists. [. . .] The

necessary can be determined only in one single way, so consequently it must be

thoroughly determined through its concept. Now only one single concept of a

thing is possible that thoroughly determines a thing a priori, namely, that of the

most real being. Thus the concept of the most real being is necessary. So there

exists a most real being.” (p. 465)420 [151] First and foremost, this proof takes as

its ground “that it stands in need of the ontological proof because it presupposes

that an absolutely necessary being is an ens realissimum, from which it would

follow, vice versa, that the ens realissimum is absolutely necessary, as the

419

Kant does not use the word allvollkommen-he uses the following names for God: “most

real being” (ens realissimum), “highest being” (ens summum), and “highest of all beings”

(ens entium); cf. KrV, B604, B607.420

Cf. KrV, B632-633. Přihonský summarises this passage, the last sentence of which

reads: “Thus the concept of the most real being is the only single one through which a

necessary being can be thought, i.e., there necessarily exists a highest being.”

In Kant’s view, the ontological argument stands in need of a cosmological argument, but

unlike the former (which is entirely a priori and moves “from the highest reality to the

necessity of existence”) (B632), the cosmological argument starts from experience and

moves from the major premiss “if something exists, an absolutely necessary being must

also exist” and the minor premiss “I exist” to the conclusion: “an absolutely necessary

being exists” (B604/632). Both the ontological and cosmological arguments erroneously

hypostatize necessity as a formal principle of thought into a “material condition of

existence” (B648), thereby converting regulative principles into constitutive ones. In

addition, the cosmological argument contains “a whole nest of dialectical assumptions”

(B638), such as extending the category of causality to the nonsensory domain, inferring a

first cause from an infinite series, and confusing the logical and transcendental

possibilities of a most real being (ens realissimum) (ibid.).

New Anti-Kant 191

ontological proof asserts. Hence the supposed experience on which the

cosmological proof is grounded is superfluous and mistaken” (pp. 466–467).421

In addition, this proof supposedly contains several other mistakes:

a) “applying the transcendental principle of inferring from the contingent to a

cause, to a nonsensory object”; b) “inferring from the impossibility of an infinite

series of causes to a first cause”; c) “assuming the fact that one cannot

comprehend anything further to be the completion of a concept”; and d) “inferring

a transcendental possibility, since we cannot find a logical contradiction in the

negation of all realities” (pp. 468–469).422

Comment. 1. First, we do not understand how Kant can assert that theappeal to experience in the cosmological proof is dispensable because inits course he borrows something from the ontological proof. In examiningthe ontological proof, Kant asserted that it failed because it did not showthat a real object corresponds to the objects: an absolutely necessary

being and an ens realissimum. But precisely this defect was supposed[152] to be rectified by appealing to the cosmological proof. – Not aword is said to criticize the inference from absolute necessity to omni-perfection423 or vice versa. We do not learn until the examination of thecosmological proof that this inference is false and that it determines theentire nature of the ontological proof.

2. (For reasons familiar to the reader) we cannot accept that theinference from the contingent to a cause may not be applied tononsensory objects. And how else if not from this inference can Kantassume that there are certain things-in-themselves underlying sensoryappearances?424

421

Cf. KrV, B634-635. Kant’s point is that the cosmological argument is fallacious, for

while it is supposedly grounded in experience, it assumes a proposition asserted in the

ontological argument, namely, “that the concept of a being of the highest reality

completely suffices for the concept of an absolute necessity in existence, [. . .]” (B635).

But absolute necessity is a necessity from pure concepts, and therefore “we have to

abandon all experience at once and seek among pure concepts for one that might contain

the conditions for the possibility of an absolutely necessary being.” (B635-636) But how

do we prove that there is just one such being without recurring to experience? In that case,

the cosmological argument grounds the ontological argument rather than vice versa; cf. J.

Barnes (1972), The Ontological Argument, op. cit, pp. 77-79.422

Přihonský summarises the passage at KrV, B637-638.423

Allvollkommenheit, but Kant does not use this word; cf. my note 420, above.424

Přihonský attacks Kant’s view that things-in-themselves are “the true correlate of

sensibility” (KrV, B45) and the underlying assumption that there must be a correlate

which can be thought, even if we cannot know it. But Kant does not say (or assume)

anywhere that there is a causal relation between things-in-themselves and appearances.

The notion of correlate implies only that there is a relation of complementarity,

reciprocity, or one-to-one correspondence between things-in-themselves and appearances.


3. We do not assert in any way that an infinite series of causes isimpossible. On the contrary, we accept the existence of countless suchseries without hesitation. For each present state425 of a finite thing is atleast partially grounded in a preceding state, and so forth, without end.But this in no way contradicts the assumption that all these infinite seriesare also grounded in God, a being whose existence has no further ground.

4. We do not want to argue about the invalid inference Kant criticizesabove (see c above), since we do not know for certain whether weunderstand him rightly. In addition, it hardly matters whether one or theother previously constructed [153] proof of God’s existence is entirelywithout mistakes or not, as long as it has not been proved that it isimpossible to find a better one, and in any case, Kant has not shown this.

5. We think that the possibility of a thing cannot be directly inferred ifno formal contradiction (a and not a) appears in the composition of itsconcept, and for this reason we consider it necessary to understand theconcept of God’s omni-perfection differently from what has usually beenthe case.

Under the heading “Discovery and explanation of the dialectical

illusion in all transcendental proofs of the existence of a necessary

being” (p. 471),426 Kant claims that this illusion arises by “an inevitablesubreption”427 when we represent this merely regulative principle ofreason to ourselves as constitutive.

Nonetheless the distinction between a regulative principle and aconstitutive principle is easily understood, such as that between thefollowing propositions which are possible here: “search for the ground ofany change” and “everything contingent <(variable)> is grounded in anecessary <(invariable)> being”. But in our opinion there is less dangerof confusing those two propositions in the proof of God’s existence thanof using the merely regulative proposition as a constitutive one. [154]

This could hardly happen to anyone, even though Kant declares that thiserror is inevitable. Instead, whoever uses the proposition: “everythingcontingent is grounded in a necessary being” in proving God’s existenceonly does so because he has good reasons to be convinced of its truth.

We can for the most part accept Kant’s objections to thephysicotheological proof which infers the existence of an infinitely wise

425

Zustand.426

Cf. KrV, B642.427

Kant says that the ideal of the highest being is a regulative principle: “to regard all

combination in the world as if it arose from an all-sufficient necessary cause, [. . .].”

However, at the same time we inevitably represent this formal principle as constitutive,

“by means of a transcendental subreption” (KrV, B647).

New Anti-Kant 193

and powerful being from the efficacy of our world-order: namely, thatfrom the latter we can only infer the existence of a very wise andpowerful being and that—to infer the existence of an infinite being—certain premisses are required from the ontological and cosmologicalproofs. – It does not follow from this that the physicotheologicalinferences are completely useless for proving God’s existence, but wemust not expect them to achieve what is not in their nature. Unlike whatis frequently believed, their true relation to the ontological inferencesdoes not consist in providing a preparation for the latter but, on thecontrary, in confirming the result of the ontological proof fromexperience. [155] The “Critique of all theology from speculativeprinciples of reason” (p. 483)428 contains nothing new but only repeatsthe unproved principle that we make neither affirmative nor negativejudgments about God.429

In an “Appendix to the transcendental dialectic” (p. 490),430 Kantexplains that there are three regulative principles (maxims) of reason:431

1) the principle of similarity (by which reason demands that wesummarise every given manifold under a higher generic concept); 2) theprinciple of variety (by which we should still find differences betweengiven things); 3) the principle of affinity (by which we should find acontinuous transition (an intermediary) between species); and that theseprinciples are often falsely considered as constitutive principles.432

428

Cf. KrV, B659.429

Kant’s view is, roughly, that we should have faith in God rather than demonstrate his

existence, that we can only make judgments of probability about the latter, and that

transcendental theology is “up to now only problematic”; cf. KrV, B669-670. Elsewhere

he claims that “it is absolutely necessary that one should convince oneself that God exists;

that his existence should be demonstrated is not so necessary” (“The Only Possible

Argument in Support of a Demonstration of the Existence of God”, in The CambridgeEdition of the Works of Immanuel Kant, Theoretical Philosophy 1755-1770 [1992], p.

163). And in a well-known passage in the first Critique, Kant admits that in respect to the

concepts of God, freedom, and immortality he “had to deny knowledge in order to make

room for faith” (KrV, B, preface xxx). It is beyond my competence to judge Kant’s

critique of (speculative) theology and his views on practical faith in God.430

Cf. KrV, B670.431

“I call all subjective principles that are taken not from the constitution of the object but

from the interest of reason in regard to a certain possible perfection of the cognition of

this object, maxims of reason. [. . .] If merely regulative principles are considered as

constitutive, then as objective principles they can be in conflict; but if one considers them

merely as maxims, then [. . .] it is merely a different interest of reason that causes a

separation between ways of thinking.” (KrV, B694)432

Kant claims that the following principles are regulative: the analogies of experience and

the postulates of empirical thought which provide rules for organizing experience but not

for constituting appearances (unlike the axioms of intuition and the anticipations of


Here we can only agree with our philosopher, although we do notbelieve that a particular natural order is presupposed in these principles.

Excerpts from Kant’s doctrine of the final aim of the natural dialecticof human reason (p. 508).433

[156 “The ideas of pure reason have a very good vocation, and it is merely their

misuse which brings about a deceptive illusion. [. . .] These ideas are mere

schemata for which no object is directly given, not even hypothetically, but they

serve only to represent other objects to us indirectly, in accord with their

systematic unity, by means of their relation to them.434 [. . .] They do not have any

corresponding objects, but this does not impede us from using them as regulativeprinciples in that we derive the object of experience, as it were, from the imaginedobject of this idea as its ground.435 [. . .] We should only consider them as

perception), the cosmological ideas, the principle of reason (which regresses from the

conditioned to the unconditioned), and the ideas of reason (God, the world, and the soul).

All of the above should not be used constitutively, since they would then be given an

illusory experience and would come into conflict with themselves.

In addition, Kant names the following three principles as maxims of reason: (1) the logical

principle of genera or sameness of kind (which postulates identity) is grounded on the

transcendental law of similarity (by which “sameness of kind is necessarily presupposed

in the manifold of a possible experience” [B682]); (2) the logical principle of species or

variety (which requires variety in things falling under the same genus) is grounded on the

transcendental law of specification (Kant takes the opposite view of Ockham’s famous

principle entia praeter necessitatem non esse multiplicanda as entium varietates nontemere esse minuendas [B684]); and (3) the logical principle of affinity or continuity(which arises by uniting the first two, for “all manifolds are akin to one another, because

they are all collectively descended [. . .] from a single highest genus” [B686]).433

Cf. KrV, B697.434

Cf. KrV, B698. Prihonksy’s summary of this passage conflates ideas and schemata, but

Kant distinguishes between them. The schemata serve to represent the objects of these

ideas and for this reason no object is given directly. In Přihonský’s vermittelst derBeziehung auf sie, “they” (sie) refers to ideas. Kant refers to the schemata when he says

vermittelst der Beziehung auf diese Idee, “by means of their relation to this idea”.435

Cf. KrV, B698-699. For example, Kant says that we use the idea of God as a regulative

principle or necessary maxim of reason by arguing from the hypothesis: suppose that if p,

then q. He considers the things in the world “as if they had obtained their existence from a

highest intelligence.” (B699); i.e., he assumes that there is an ideal being or God which

has the power to produce things in the world, as a possible explanation of our experience

of these things. His purpose is to show that the three kinds of transcendental ideas

(psychological, cosmological, and theological) can be used as necessary maxims even

though they cannot be directly referred to any object corresponding to them. Similarly to

the schemata which mediate between the concepts of the understanding and the objects

falling under them, the ideas of reason have analoga to the things they refer to; and

therefore, we can assume or postulate them as hypotheses and derive the object of

experience from them by analogy: we consider objects of experience as if they were

related to an idea of reason. In other words, we formulate a hypothesis, or give a reason

and explain an observation, by deriving an observational statement from this reason. So

Kant’s use of the explanatory nature of hypothesis can be considered a forerunner of C. S.

New Anti-Kant 195

chemata or analoga of real things, and we should always be aware that they are

our means for thinking an indeterminate something of which we have no concept

at all of how it is in itself. We do not assume such ideal entities or beings in theidea per se, but only as regulative principles of the world of sense. We only

assume them so as to extend the empirical use of our reason as far as possible

without extending it beyond the boundary of experience.436 [. . .] Hence we

attribute only such properties as are analogous to the concepts of the

understanding in its empirical use to the indeterminate something we think in

these ideas.437 Even [157] the concepts reality, substance, causality, and necessitywhich we ascribe to these entities lose all significance and are empty titles for

concepts without any content if we venture with them beyond the domain of the

world of sense.438 So these entities of reason are mere ideas and are not reallyused as grounds, but only problematically.439 Particularly:

1. If we assume in the first idea a soul or simple independent intelligence, we

should do this merely to consider all representations as in a single subject and to

derive all our powers from a single basic power, and so on; for all this can only

happen by means of such a schema as if it were a real entity. But we should not

assert that such a simple substance is the real ground of all our representations,

and so on. Nor should we accept hypotheses on the origin, and so on, of the

soul.440

2. We should also use the second idea of the world in general only to proceed in

explaining the given appearances as if the series in itself were infinite; but if

reason itself is considered as the determining cause (of freedom), and thus of

practical principles, we should proceed as if we did not have before us an object

of sense but one of pure understanding, where the [158] series of states can be

regarded as if it began absolutely (through an intelligible cause). All this proves

that the cosmological ideas are merely regulative principles.441

3. Nor do we have the least reason, concerning the third idea, to assume such a

being absolutely. We only assume it in order to consider all connections in the

world according to principles of a systematic unity.”442

But several mistakes arise if we use these ideas constitutively rather thanmerely regulatively, such as

Peirce’s notion of abduction; cf. Peirce (1876), “Deduction, Induction, and Hypothesis”,

in Selected Philosophical Writings, vol. 1 (Bloomington: Indiana University Press, 1992),

pp. 186-199.436

Cf. KrV, B702.437

Cf. KrV, B703.438

Cf. KrV, B705-707. Here Kant refers to God as the ideal being to which these concepts

are attributed.439

Cf. KrV, B709.440

Cf. KrV, B710-711.441

Cf. KrV, B712-713.442

Cf. KrV, B714.


a) lazy reason, if we consider our investigation into nature as absolutelycomplete, rather than seek its cause in the laws of the mechanismof matter.443

b) perverted reason, if we determine a divine purposeanthropomorphically and impose it on nature, rather than seek forit.444

In the end Kant raises several questions:1. “Is there anything different from the world which contains the

ground of the world order? Answer: Without a doubt, for there must be atranscendental ground.”445

2. “Is this being a substance, of the greatest perfection, necessary, andso on. Answer: This question has no significance at all, for all thecategories are of none but an empirical use.”446

3. “May we not at least think this being in accord with [159] ananalogy to objects of experience? Answer: By all means, but only as anobject in the idea.”447

4. “Can we nevertheless assume a wise and omnipotent creator?Answer: Without any doubt, we must presuppose such a being.”448

5. “But then do we extend our cognition beyond the field ofexperience?” Answer: By no means, for we have only presupposed asomething of which we have no concept at all of what it is in itself.”449

6. “Can we make use of this assumption? Answer: Yes, that was whythings were grounded on this idea. You may even attribute to this beingliking, disliking, and desire.”450

Comment. 1. Although Kant has used so many words for describingthe way in which he wants the ideas of pure reason to be used, we do notknow precisely and clearly what he requires. He talks of a (regulative)use of those ideas for the mere purpose of bringing a systematic unity toour cognition of nature. Obviously, this is only possible throughinferences. But inferences cannot be derived from mere concepts, onlyfrom propositions, so Kant must have considered these ideas as certainpropositions, presumably the following propositions: there is a soul, aworld order, a God. [160] However, he also says that one should not

443

Cf. KrV, B718-719.444

Cf. KrV, B720-721.445

Cf. KrV, B724.446

Cf. KrV, B724.447

Cf. KrV, B724-725.448

Cf. KrV, B725.449

Cf. KrV, B726.450

Cf. KrV, B728.

New Anti-Kant 197

attribute any ideality to these ideas, so it seems that he did not want us toconsider these propositions as proved truths but to use them neverthelessas premisses for deriving certain inferences about systematic unity. – Ifthis was really all he meant, he could have expressed himself much morebriefly and clearly!451

451

For Kant, an idea is a concept of reason which has no relation to an object of experience

(cf. KrV, B376-377; JL, § 3). Ideas are thus distinct from concepts and transcend

experience. Just as he had derived the categories from the forms of judgment in the

Analytic, in the Dialectic he derives the ideas from the forms of syllogisms. This is why

Přihonský accuses him of conflating concepts and ideas with propositions and inferences.

He follows in Bolzano’s footsteps, since this objection belongs to the latter’s more general

criticism of Kant’s view that ideas, propositions, and truths are mental entities and,

consequently, that logical laws are the forms of thought. In Bolzano’s view, logical laws

are not forms of thought, in the same sense that mathematical laws, such as the law of

transitivity, are not forms of thought. Cf. WLI, § 16.2-3, §§ 36-37, § 50.2. Přihonský and

Bolzano’s point is correct, at least within their theory. To what extent their criticism is

applicable to Kant is another question, for Kant’s aims (and, consequently, his

distinctions) are very different from theirs: (1) He takes the word idea in a sense that lies

somewhere between Plato’s notion of ideas as forms and the modern view that ideas are

something which occurs in someone’s mind. He agrees with Plato that ideas are forms in

the sense of types or ideal patterns but rejects the Platonic belief that ideas exist in a

world of their own (hence his objection that Plato hypostatized the ideas). Thus Kant

could reply to Přihonský that if ideas are Platonic forms, they can also be principles. On

the other hand, Kant agrees with the modern view that ideas have no independent

existence, and he rejects the claim that ideas are directly related to our experiential

cognition. For this reason, Kant disagrees with the British empiricists’ view that all

objects of consciousness are ideas, regardless of whether they are perceptions, images,

recollections, or concepts, and (2) he distinguishes between the understanding and reason

and claims that the first is the faculty of concepts and judgments, whereas the second is

the faculty of ideas and inferences. Consequently, (3) he distinguishes between concepts

(or categories) and ideas; only the former bear a relation to possible objects of experience.

(4) The understanding is the faculty of thinking and judging, and judgments are “the

function of unity among our representations” (B94). - Thus the specific feature of

categories is that they have a relation to an object only when they are unified in a

judgment (cf. my note 232). Reason, on the other hand, is the faculty of making

inferences; their function is to relate a particular judgment to a universal condition: in the

conclusion of a syllogism a predicate is restricted to an object “after first having thought

it in the major premiss in its whole extension under a given condition” (B379). - Thus the

specific feature of an idea is analogous to that of a conclusion in a syllogism: an idea

relates an unconditioned totality of conditions that are necessary for a given conditioned

state. Hence it can refer to the totality of all possible experience as a regulative principle.

According to Kant, the problem with principles of reason is that since they are synthetic

(because the subject-concept is not contained in the predicate-concept; cf. JL, § 53) and apriori, reason is unrestricted by experience and extends itself beyond its legitimate limits.

Cf. KrV, B382. So the activity of reason is restricted to making inferences (cf. KrV,

B169); but these inferences are valid only if reason is confined within the boundaries of

experience. Cf. KrV, B382-385.


2. Assuming that this is really Kant’s opinion, we further recall thatalthough it is true that false or yet unproved propositions canoccasionally be used with certainty as premisses in inferences of aparticular kind, but unlike Kant we do not believe that this is the best or

only way to obtain these conclusions. Indeed, why should we not attemptto derive all the powers we come across in our self from a singlegrounding power and achieve this without assuming the substantialsimplicity of this self? Why not seek and find for every state in nature apreceding one as its condition into infinity, without presupposing that theseries of these conditions is indeed infinite? And so on.

3. Further, what does Kant mean by saying that the concepts reality,substance, causality, necessity, and so on, lose all significance andbecome empty titles [161] for concepts without any content if we venturebeyond the domain of the world of sense? – Is a concept modified evenslightly if we apply it as a predicate to some object, even aninappropriate object? Does the concept of quadrangularity lose itssignificance if we assert of a circle that it is square, or is not rather thefalsity of this proposition grounded on precisely the fact that the conceptmaintains its significance?

4. We cannot not see by any means how the idea of God can lead tothe mistakes found by Kant if it is used constitutively and not merelyregulatively, that is, if we consider the proposition: there is a God, as trueand proved, and thus accept all inferences resulting from it withoutthinking. Or through which minor premiss we infer from the proposition:there is a God, the conclusion that we consider our investigation ofnature as absolutely complete, rather than seek its cause in the laws ofthe mechanism of matter if we are capable of doing otherwise, andcontinue our search? As for seeking out the purposes of nature, it iscertainly true that the assumption of God’s existence gives us thecertainty that every order and every event in the universe which is notabsolutely necessary, that is, a mere consequence of purely theoreticalconceptual truths, such as mathematical truths, and so on, has a purposeconsisting in the greatest possible happiness of [162] living beings. Butin a teleological investigation we shall not content ourselves with thissimple general affirmation, for there we ask for an empirical proof of agiven natural order or an event which does not appear absolutelynecessary to us, or a particular visibly beneficial effect which itproduces. We can explain the obliqueness of the ecliptic teleologically byproving first of all that among all the pure conceptual truths we areacquainted with (the laws of mechanics), there is not a single one fromwhich this obliqueness follows with absolute necessity, but then show

New Anti-Kant 199

how this order gives rise to the beneficial effect that a much larger part ofthe earth’s surface becomes habitable by man and animals than would bethe case if the earth’s axis were standing vertically on the ecliptic.Anyone can see that such proofs have to be carried out in exactly thesame way, regardless of whether we presuppose God’s existence ascertain or probable or do not presuppose it at all.

5. We already mentioned that in our view Kant is inconsistent inanswering the first of the above questions affirmatively, precisely“because there must be a transcendent ground”.452 Why should theconcept of ground be the only one that extends beyond the domain of theexperiential world and thus give evidence [163] of the existence of anonsensory cause of that world? As little as we may determine “thissomething which is different from the world and contains its ground”, wenonetheless apply some of our concepts to it, for example, something assuch, something different from the world, and finally a ground of theworld. The question is, why are we allowed to apply only these conceptsand no others in this transcendent way?453 And how can Kant say that wehave no concept at all of what this being is in itself, since we call it “abeing different from the world which grounds the world”?454 – We objectthat Kant goes too far in wanting to allow us to attribute desire anddislike to this being.455

452

Cf. KrV, B724, but Kant says that there must be a transcendental ground, i.e., a ground

which can be the condition of a possible experience, as opposed to a transcendent ground

which would be beyond the boundary of experience.453

Kant’s point is not that these concepts are the only ones that can be applied

transcendentally (and not transcendently) but that they are used to answer the first

question (cf. KrV, B724).454

Kant asks whether there is “anything different from the world which contains the

ground of the world order?” and his answer is “without a doubt, for there must be a

transcendental ground.” (KrV, B724) In other words, he wants to explain how this being

can be a transcendental ground, namely, by being different from the world and containing

its ground.455

Kant says that we are not justified in assuming a being above nature with those

properties, but we may use the idea of such a being as grounding the world-order as a

regulative principle, as if the being represented in this idea had those properties. This is

why he says that “we are also justified in thinking of the world-cause in the idea not only

according to a subtle anthromorphism [. . .], but also in ascribing to that being infinite

perfection far transcending what we could justify on the basis of our empirical

acquaintance with the world-order. For the regulative law of systematic unity would have

us study nature as if systematic and purposive unity together with the greatest possible

manifolds were to be encountered everywhere to infinity.” (KrV, B728) In other words,

our idea of this being goes beyond our possible experience of it. Whether or not this view

is correct is for theologians to decide.


Kant defines the transcendental doctrine of method (pp. 535ff)456 as“the determination of the formal conditions of a complete system of purereason” and divides this second main part of the critique into fourprincipal sections: a discipline, a canon, an architectonic, and a history

of pure reason, without explaining how these subsections belong here.[164] According to p. 538, the discipline of pure reason is a special andnegative legislation established from the nature of reason and the objectsof its pure use in a system of caution and self-examination.457 However,these prescriptions should not refer to the content (since it wassufficiently discussed earlier on) but only to the method. – The fortunateprogress of the mathematical sciences could give rise to hopes that inphilosophy we would also obtain this apodictic certainty if only weimitated the mathematical method precisely enough. But Kant assures usthat this hope is futile. Philosophy is a rational cognition from mere

concepts, and mathematical cognition is a cognition from theconstruction of concepts through pure intuitions.458 Mathematics canfollow this method of construction because its objects (space, time, andquantities as such) allow it,459 or (as Kant thinks) mathematics has theseobjects because they are the only ones to which its method ofconstruction through pure intuitions can be applied. But this method isnot applicable to objects dealt with by philosophy because there are nopure intuitions of them (p. 539).460 – Mathematical thoroughness is based

456

Cf. KrV, B735-736.457

Cf. KrV, B740-741.458

Cf. KrV, B741. Kant says that philosophical cognition (and not philosophy) is “rational

cognition from concepts and mathematical cognition, that from the construction of

concepts. But to construct a concept means to exhibit a priori the intuition corresponding

to it.” (ibid.) Cf. on the problem of mathematical cognition and the construction of

concepts: on the procedure of introducing limitations, see my note 192; on Kant’s

“mathematical prejudice” and the schematism, see my notes 203-208; and on the

construction of quantity concepts, see my note 264.459

I use “allow it” for translating es so verstatten.460

Cf. KrV, B741. Přihonský’s presentation of this issue is inaccurate: in Kant’s view, “the

essential difference between these two kinds of rational cognition [. . .] does not rest on

the difference in their matter, or objects.” (B743) Mathematical cognition “considers the

universal in the particular, indeed even in the individual, yet nonetheless a priori” (ibid.).

Therefore the object corresponding to the concept [triangle] is exhibited “either through

mere imagination, in pure intuition or on paper, or in empirical intuition but in both cases

completely a priori, without having had to borrow the pattern for it from any experience.

The individual drawn figure is empirical and nevertheless serves to express the concept

without damage to its universality, [. . .]” (B742-743). Philosophical cognition, on the

other hand, considers the particular in the universal and deals solely with general concepts

(and thus deals only with general principles and procedures, unlike mathematical

cognition, which constructs particular objects). Thus, for example, it does not search for

New Anti-Kant 201

essentially on definitions, axioms, and demonstrations, [165] but inphilosophy none of these three elements can be applied (p. 548).461

1. According to Kant (p. 549),462 to define means to present originally

the exhaustive concept of a thing within its boundaries (i.e., such that thispresentation does not stand in need of a proof), but a) no empirical

concept can be defined in this manner because we can never be certainwhether the given marks are sufficient for differentiation; and b) neithercan a concept given a priori be defined in this way, for example,substance, cause, right, equity, and so on, for we can never be certain thatwe are clearly aware of all the marks we represent in a concept, since itmay contain many obscure ones.463 Thus remain only arbitrarily thought

concepts, although for them we must know what we want to think. Butagain we cannot know a priori whether such an arbitrarily composedconcept has an object if the synthesis contained in the concept cannot beconstructed a priori. This is only possible in mathematics, and thus onlymathematics has real definitions.464 In philosophy, however, we should a)not imitate mathematics in putting the definitions first but ratherconclude with them, because it is only at the end of a treatise that we can

an intuition of the concept [reality], because we cannot derive an intuition of [reality]

outside experience. Contrary to Přihonský’s view of Kant, the latter claims that as well as

mathematics, philosophy deals with magnitudes, e.g., with totality or infinity, so they

sometimes have a common object. What is different is “the manner of dealing with it

through reason” (B743).461

Cf. KrV, B754-755. Kant says that none of these three elements can be achieved or

imitated by philosophy “in the sense in which the mathematician takes them” (ibid.).462

Cf. KrV, B755.463

Cf. KrV, B755. Kant says that an empirical concept can only be explicated because in it

we “have only some marks of a certain kind of objects of the senses”, and therefore “it is

never certain whether by means of the word that designates the same object one does not

sometimes think more of these marks but another time, fewer of them.” Kant uses an

Aristotelian model of definition (cf. Aristotle, Topics VI, 13; Posterior Analytics II, 13;

Metaphysics VII, 12, 1037b12, 1038a1ff.), namely, that a definition is a form of analysis

by genus and specific difference. First, a definition names the attributes a certain thing has

in common with other things; and second, it names the feature that singles a thing out

from others. On this account, a definition should contain the essential attributes of a thing,

tell us what is and what is not included in a concept, give rules for membership or

belonging to a certain concept, and hence clarify the boundaries of a concept. But Kant

argues that this account of definition (or form of analysis) works neither for empirical

concepts, because we cannot provide sufficient criteria for differentiating a concept from

other species, nor for concepts given a priori, because we can never be certain that their

specific marks (or attributes) are presented with sufficient clarity and distinction.464

Kant says that mathematics begins with definitions, that is, mathematical concepts are

first given through the definition and then constructed through intuitions; cf. KrV, B759.

Cf. Dubislav (1931), Die Definition, ch. 2, “Die Definition als Begriffsbestimmung”, p. 7


tell with any reliability whether we have written it rightly,465 and weshould b) not [166] forget that philosophical definitions cannot attain theinfallibility of mathematical definitions.

2. Axioms (p. 552)466 are synthetic a priori principles, insofar as theyare immediately certain. But one concept cannot be synthetically andimmediately combined with another (i.e., unless it is already contained init as a component), since a third mediating cognition is necessary for this.So in philosophy, which is a rational cognition from mere concepts, noprinciple can be found that deserves the name of an axiom. Mathematics,on the contrary, can provide axioms, because by means of theconstruction of concepts in the intuition of an object, it can connect thepredicates of the latter a priori and immediately. Thus we should not putforward axioms in philosophy but first justify all principles through athorough deduction.

3. Demonstrations (p. 553)467 are proofs which not only show thatsomething is but that it cannot be otherwise (apodictic), insofar as it isintuitive (i.e., self-evident). Proofs from mere concepts cannot beintuitively certain. Thus philosophy cannot provide demonstrations butmust leave them to mathematics which constructs its concepts inintuition.468

Comment. 1. In our view, our philosopher has wreaked, andcontinues to wreak, immeasurable damage in the domain of philosophythrough the assertions that occurred to him in this passage.[167] For thephrase in philosophical investigations nothing can be precisely defined

and strictly demonstrated was passionately grasped and maintained evenby those who did not accept any of his other assertions. From then on itbecame (what Kant certainly did not want) a prevailing practice inGerman philosophy to revile strict definition and demonstration aspedantic and—without having made oneself understood—to expect thereader to guess from the mere context of the discussion which conceptsone related to one’s expressions or which reasons had led to one’sassertions and which reasons one would still have to give. Yet of allsciences, philosophy requires the most careful definition of concepts and

465

Přihonský’s summary is slightly off the point: unlike mathematics, philosophy begins

with concepts, and philosophical definitions are analyses of given concepts. Since they are

given, these concepts are, at first, exposed incompletely “so that we can often infer much

from some marks that we have drawn from an as yet uncompleted analysis before we have

arrived at a complete exposition, i.e., a definition;” (B758).466

Cf. KrV, B760.467

Cf. KrV, B762.468

Kant says that the latter is a cognition of the construction of concepts through or bymeans of intuition, not in intuition; cf. my note 459, above.

New Anti-Kant 203

circumspect proofs, from the constant consideration of all the illusoryprinciples opposed to it. It even got to the point that a certain schoolasserted that concepts should transform themselves during aphilosophical presentation and that propositions should cancel each otherout, since no proposition could be completely true by itself. Thus thispractice or rather, malpractice, deserves the strongest rebuke, and wecannot find words enough to warn against it.

2. The distinction Kant assumes between [168] philosophy andmathematics, by defining the first one as a cognition from pure conceptsand the second one as a cognition from the construction of concepts inpure intuition, is completely false and misleading.469 Although it is true,as we have already mentioned elsewhere,470 that the geometer resorts alltoo frequently to a support in his proofs which we could call an appealmerely to that which appears to the eye, in our view it is by no meansnecessary for him to do this. It is possible to derive all geometrical truthsfrom the right definition of space, without even once permitting oneselfan inference which has no other justification than that this is what we aretold by that which appears to the eye. Examples of this procedure, whichat least show how quite a few geometrical truths can be discoveredthrough mere inferences without any appeal to that which appears to theeye, are provided by well-known analytic geometry. The aim ofBolzano’s mathematical treatises, written and partly edited more thanthirty years ago, was to instigate such a method and draw the attention ofthe scholarly world to it. But his written estate*471 contains much more,[169] even the conception of a complete work on such an entirely new

469

Kant does not define philosophy and mathematics as two kind of cognitions but

distinguishes between two methods and two kinds of cognition: philosophical cognition

(and the philosophical method) and mathematical cognition (and the mathematical

method); cf. KrV, B741. Kant also says that the latter is a cognition by the construction of

concepts through or by means of intuition, not in intuition; cf. my note 459. Cf. Dubislav

(1930), “Zur Wissenschaftstheorie der Geometrie”, in Blätter für Deutsche Philosophie 4,

pp. 368-381, and in Die Definition (1931), ch. 2, op. cit. Cf. also Couturat (1905), Laphilosophie des mathématiques de Kant (appendix to Les principes des mathématiques,

1979, pp. 235-308). Both authors criticize the basis of Kant’s distinctions between

philosophical and mathematical cognitions and between the philosophical and the

mathematical method-namely, the distinction between analytic and synthetic definitions.470

Cf. NAK, pp. 70-71, and my notes 203-208.

*471

Bolzano died on 18 December 1848. Shortly before his death, he successfully

completed one of the most astute treatises: “On the Paradoxes in Mathematics and in

Philosophy”, which will soon appear in print. [Přihonský’s note.]

Přihonský refers to the Paradoxes of the Infinite, which he edited and which appeared

posthumously in 1851.


presentation of mathematics. All it needs is a skillful hand, capable ofarranging the material systematically.

3. The concepts which Kant reveals on this occasion as the essence ofdefinition and of other logical doctrines are very imperfect. Here heobviously confuses the mere explication of what is conveyed in a word,that is, the exchange through which we produce the same concept that werelate to a certain word in the soul of the reader and make him cognizethe fact that it is this concept we wish to denote by that word, and thedefinition of a concept, that is, the account of whether this concept issimple or complex and, in the latter case, of which parts it is composedand how it is composed of them. The former is an inevitable exchange onevery word about whose meaning the readers are somewhat in doubt. It isvery incorrect to believe that a philosopher can outwit this explication,since he needs it the most. And it is false that a philosopher lacks thenecessary means for it and is thus permitted to expect his [170] readers toguess painstakingly from the context of the discussion which concepts hehas related to each word, if luck will have it. Further, it is wrong thatthere is no other way to gain such an explication than by defining aconcept: various other ways of achieving an explication are described inBolzano’s Theory of Science, § 668.472 When Kant says that a definitionis a detailed concept and goes on to say that we often fail in ourdefinitions by passing over certain marks (components) which are onlyobscurely thought in the concept, apparently this applies only toexplanations. But then he should not have required that definitions shouldnot need proofs. For the explanation of a given concept, that is, theassertion that a certain concept we are already familiar with fromsomewhere else is completely simple or arises from the combination ofsuch and such parts, is nearly always in need of a separate proof.473 And

472

In WLVII, § 668.2, Bolzano defines “explications” (Verständigungen) as propositions

that are established with the intention of acquainting our readers with the sense in which

we use certain signs: x is a Verständigung if and only if x can convey the sense in which

we use certain signs. An explication transmits information between an author and a reader

iff the reader believes that the author wants him to believe that p; cf. also WLVII, §§ 644,

653, 662, 672. So an explication concerns the interpretive relation between linguistic and

mental events: it is a statement in which a word is used to define a certain thing in such a

way that the reader’s understanding of it is modified. Hence explications are stipulative

definitions or conventions which use familiar signs in a new sense. Cf. A. Kasabova

(forthcoming article), “Bolzano’s semiotic method of explication”. Přihonský opposes

Bolzano’s notion of explication to Kant’s (critique of) definition as analysis by genus and

difference.473

Kant says that the concept of an object “as it is given, can contain many obscure

representations, which we pass by in our analysis though we always use them in

application”, and this is why the “exhaustiveness of the analysis of my concept is always

New Anti-Kant 205

philosophy, as well as mathematics, has a considerable number of suchgiven concepts as equality, similarity, opposition, magnitude, number,sum, product, difference, quotient, power, point, distance, direction,angle, extension, line, area, body, and so on. Mathematicians have stillnot sorted out the explanation of these concepts. In addition, it is falsethat the definition [171] of these concepts does not belong tomathematics at all or that it cannot add anything to its perfection. Tomention just one example: the right explanation of the concept ofsimilarity provides the easiest proof of three very important principleswhich could hardly be proved in previous terms, by means of manyhundred inferences, that similar lines, areas, and bodies behave as doother lines, areas, and bodies which are derived from them in a similarway (the areas of two triangles, e.g., as do the squares of certain sideswith a common denominator, etc.). Besides, Kant’s assertion that not asingle given concept can be explained with certainty, because therealways remains a possibility that we will pass over an obscurely thoughtmark in it, is surely an exaggeration. Should not the presence of such acomponent be finally revealed through precise and often repeatedobservations and by applying the concept?474 For example, is it not

doubtful” (KrV, B756-757). Kant’s point is that strictly speaking, concepts given a prioricannot be defined because I can never be certain that the distinct representation of a (still

confused) given concept is adequate to the object, and for this reason he prefers to use the

expression “exposition” rather than “definition”. Přihonský’s objection missed this point.

On Kant’s critical view of definition as a form of analysis with regard to empirical and apriori concepts, cf. my notes 464 and 466, above.474

Přihonský’s objection recalls Bolzano’s justification of the demonstrative sciences and

his claim that conceptual truths can be tested either through experience or by repeatedly

testing their derivation. In a well-known anti-Kantian passage in the WL, Bolzano asserts:

“If we have not tested the truth of a proposition either by experiment or by repeatedchecking of its derivation, we do not give it unqualified assent, if we are at all sensible, nomatter what critical philosophy may say about the infallibility of pure intuition. . . . Doesnot experience teach us that we make mistakes in mathematical judgments, and that we

make these mistakes more easily, the more we trust what philosophy calls by the high-

sounding name of pure intuition? The geometrician, who thought that a pair of solids have

the same content if they are bounded by similar and equal sides, made such a gross

mistake certainly only because he trusted his intuition, i.e., mere appearance, too much.

The only reason why we are so certain that the rules barbara, celarent, etc., are valid isbecause they have been confirmed in thousands of arguments in which we have appliedthem. This also is the true reason why we are so confident, in mathematics, that factors in

a different order give the same product or that the forces on a lever are in equilibrium

when they stand in the inverse relation of their distances from the fulcrum, etc. But that 2

= 1.414, . . . that the content of a sphere is exactly two-thirds of the circumscribed

cylinder, that in each body there are three free axes of revolution, etc., we assert mainly

because they follow from propositions of the first kind by arguments which others have

conducted hundreds of times and found valid; another factor is that in all these matters we


certain enough that the concept something is a simple concept? – Butassuming we cannot make certain that the concepts we relate to the words(given as examples by Kant) substance, cause, and so on, are indeed thesame concepts we put forward in a certain explanation, it is enough if we

do not have the slightest advantage if the thing turns out to be otherwise. That the reason

for our confidence really lies in these circumstances can be seen most clearly because our

confidence rises and falls as these circumstances dictate.” (WLIII, § 315.4, my italics)

Bolzano’s claim has provoked Husserl’s and Stove’s accusation that he is an empiricist

who holds that we believe logical generalizations because they are true observation-

statements. Husserl was probably the first to point out the epistemological fallacy in

Bolzano’s theory: “One must not judge without a more exact knowledge of Bolzano’s

work; one must not read superficially or look around in Bolzano for quotations in order to

attribute my thoughts to specific like-sounding propositions in his work. No one has taken

notice of the fact that Bolzano's epistemology rests upon the foundation of an extreme

empiricism. [. . .] there is here no difference between Bolzano and his contemporary, John

Stuart Mill.” Husserl (1913), Introduction to the Logical Investigations, trans. Ph. Bossert

and C. H. Peters (1975), p. 48. David Stove, The Rationality of Induction, part 2 (1986),

pp. 160-162. I think that this empiricism-accusation is only justified in part, for Bolzano

argues on the contrary that we cannot generalize from observation or empirical intuition:

we can come to know, by observation, that a single line can be extended, but we cannot

infer from this the general truth that all straight lines can be extended. For this

generalization we need to appeal to the principle of similarity. So we do not learn general

truths through intuition but through conceptual principles: “we can only infer such a

general truth from what the appearance of only a single line teaches us if we assume the

further truth that all straight lines are similar to one another, and thus that every inner

attribute representable by means of concepts found in one of them must also reside in

every other. This proposition, however, no intuition can teach us.” (ML, § 14)

Bolzano’s actual claim about our empirical knowledge of logical and mathematical truths

is much weaker: he claims that we can learn or become acquainted with logical truths by

experience, and we can confirm such truths by testing them empirically. For example, we

can learn the syllogistic form of barbara, that “x is F and all F are G” implies “x is G”, for

all x and all F and all G, by testing it against experience. We discover in all observed

instances that if both the premisses of the barbara form are true, then the conclusion is

true, and there are no instances of barbara where the premisses are true and the conclusion

is false. This implies that our verification not only grounds our confidence in logical rules

and mathematical propositions but also the justification for the applicability of these rules.

And here the accusation has a point, for Bolzano’s attempt to combine epistemological

and logical aspects leads to an inconsistency in his account of justification. However, the

point here is not so much that Bolzano is an empiricist but that such a combination

inevitably results in problems, at least as long as it includes formal logic. Yet Bolzano’s

notion of justification is probably as adequate a foundation for the demonstrative sciences

as you could get in the nineteenth century, and its shortcomings may be due to the

problem of justification itself, rather than to Bolzano’s account. Cf. WLII, §§ 155, 197.3;

cf. Mark Textor, “Logically Analytic Propositions A Posteriori?”, (2001): p. 1; cf. also A.

Von Duhn, “Bolzano’s Account of Justification”, (2002): pp. 21-34 and A. Von Duhn,

“Bolzano’s Distinction between Confirming (Gewissmachen) and Justifying (Begründen)

Propositions”, forthcoming.

New Anti-Kant 207

are convinced that the relation between the representations we indicatedcan present a concept worth putting forward. Then we rightly [172]

demand that a word (with a similar sense, for instance from ordinaryusage) be ascribed as a sign to this concept. So for the word substance

someone could at least form the secondary representation of somethingpersistent which carries the properties we attribute to a substance. Wecan rightfully assert that these secondary representations should beeliminated, or at least separated from the concept that we teach inrelation to this word, as not belonging to the subject. We entirely rejectthe claim that an empirical concept cannot be defined at all because somethink of these marks and others of those marks as attached to objects suchas water or gold, and because we cannot be certain whether these marksare sufficient for making distinctions. For although it is true that differentpeople think of different marks as attached to the same words, such aswater, gold, and so on, what is attached to these words really aredifferent concepts, and it is necessary and anything but impractical toabolish this difference between concepts or at least to convey anexplication to the reader of which concepts we relate to these words. Butthe assertion that the marks contained in a concept are not sufficient fordistinguishing its object is self-contradictory. For if the marks we putforward in our concept, for example, of gold, are also to be found inplatinum, then this is [173] precisely why platinum, and not gold, is theobject of our concept. A new error is revealed when Kant continues bysaying that such words are only an indication, and not a concept of the

thing. Can we indicate a thing without first having a concept of it or atleast a representation?475 – Our readers are already familiar with ourobjection to Kant’s claim that we cannot be certain that an arbitrarilyformed concept has objectivity unless we obtain a corresponding

475

Kant, of all people, would agree with Přihonský that we cannot designate a thing

without first having a concept of it: he claims that philosophy begins with concepts; cf.

KrV, B759. A designation, however, is a word containing only a few marks of a thing, and

therefore it is not a concept of a thing (since its extension has no secure markers), and its

putative definition is only a determination of the word. Přihonský rejects Kant’s claim that

empirical concepts cannot be defined but only explicated because we can never be certain

“whether by means of the word that designates the same object one does not sometimes

think more of these marks but another time fewer of them” (B759). Kant says that there is

little point in defining an empirical concept, for when such a concept and its properties are

under discussion, one will not stop at what is intended by the word but rather advance to

experiments, and therefore the word and the marks attached to it only designate the thing.

A designation, however, is too inaccurate to be a concept (since we can never be certain of

the specific difference which singles out that thing), and a definition would be invalid,

because we cannot provide sufficient criteria for differentiating this putative concept

(which is only a designation) from others.


intuition. In mathematics, as well, it is not through intuitions that weascertain the reality of the concepts: infinitely great, infinite line,polygon, √2, and a hundred others.476 As to the place where definitionsshould be established, the same rules that apply to mathematics alsoapply to philosophy. In both sciences, explications of what is conveyed in

a word should precede their usage, whereas explanations of a conceptcan occasionally be omitted or added later on, but obviously theiromission in philosophy is a worse evil than in mathematics. Finally, theclaim that a mathematician can never make mistakes in his definitions,even in those that are only arbitrary, should also [174] be refuted byexperience itself. For how often do not mathematicians put forwardconcepts which subsequently prove to be very unsuitable and evencontradictory or irrelevant? To take an example, we need only recall theconcept of a tangential contact.477

4. If it were true that no concept could be immediately combined withanother (which was not already contained in it as a part), that is, that nosynthetic judgment could be immediately self-evident, then cognition insynthetic judgments would be impossible anyway, since each judgmentwould first presuppose another judgment from which it was derived, toinfinity. Bolzano (Theory of Science, § 300) rightly teaches that there areand there should be quite a few immediate judgments, empirical as wellas purely conceptual ones.478 On the contrary, judgments which Kant

476

Cf. my note 475, above.477

“Tangential contact” translates Berührungswinkel.478

Přihonský objects to Kant’s account of axioms (cf. KrV, B761) and bases his objections

on Bolzano’s views.

Kant claims that axioms are intuitive principles or immediately certain a priori judgments

that can be exhibited in intuition and from which other judgments are proved, “but they

themselves cannot be subordinated to any other.” (JL, §§ 34-35) He distinguishes between

intuitive and discursive principles (axioms [mathematical principles] and acromata

[philosophical principles]), in order to distinguish between mathematical and

philosophical methods of proof (cf. also KrV, B762-763). Unlike axioms, philosophical

principles have no recourse to a construction through intuition and “can never be certain

from mere concepts” (B761). Bolzano and his school firmly reject this distinction between

the sciences.

Bolzano claims that axioms are unprovable propositions composed of simple concepts, but

it does not follow from their unprovability that they are self-evident (Beyträge, II, § 19).

They are synthetic propositions because, since they are composed of simple concepts, they

do not have a component which can be varied arbitrarily whilst conserving the original

truth value of the proposition (as do analytic propositions); cf. WLII, § 148, § 197. He

rejects the post-Euclidean view on axioms also held by Kant (cf. KrV, B760) that basic

conceptual truths are unprovable because they are intuitively self-evident. In 1810 he

already argues that the characteristic feature of an axiom does not lie in its intuitive nature

(Anschaulichkeit). This property, he says, “is hardly suitable to provide a firm basis for

New Anti-Kant 209

the classification of all truths into two classes, that is, into axioms and theorems”. He

gives three reasons for the unsuitability of intuitiveness. First, this property allows of

differences in degree, so we will not be able to determine precisely what degree would be

sufficient for an axiom. Second, the intuitive nature depends on fortuitous circumstances,

such as our education or experience, which bring us to recognize it more or less

frequently. Third, the degree of intuitiveness differs from one person to the next, and what

is obvious to one may appear obscure to another. Cf. Beyträge, II, § 10; “Appendix on the

Kantian Construction of Concepts through Intuitions” (Anhang über die Kantische Lehreüber die Construction der Begriffe durch Anschauungen) (1810) (appendix to Beyträge),

op. cit.; WLIII, § 305.5, § 314.3. Bolzano’s axioms are basic truths from which the other

propositions in a theorem are derived (cf. Beyträge, II, § 13). Contemporary logicians

would add that they accept several axiomatics, whereas Bolzano accepts only one.

Bolzano has three main claims concerning the immediacy of conceptual judgments in

general and the nature of axioms in particular: (1) Conceptual truths, unlike perceptual

truths, cannot be known (wissen) immediately or be self-evident, (cf. WLIII, § 300.13),

although we can cognize them (erkennen) without inferring them from other truths, but

then we do not have a clear Erkenntnis. We have a clear cognition of a truth if and only if

we are aware of its ground. Cf. WLIII, §§ 300.9, 316.3.

(2) Conceptual truths may be subjectively plausible (einleuchtend) and even certain, but

we need to prove their objective grounds. Such intuitive self-evidence of a proposition is

no proof of its truth. In the Religionswissenschaft (1834, RWI, § 3), Bolzano says that it is

easy to convince someone of the truth that it is colder in winter than in summer, merely by

referring to das blosse Gefühl. Likewise, he adds, it is easy to convince anyone, even the

most stupid person, of the truth that a straight line is the shortest distance between two

points, merely by asking him to span a thread between two points and letting him see that,

as he stretches the thread, the distance between the two points becomes shorter as the

thread takes the position of a straight line. But is this how we make him understand the

reason (Grund) why the straight line is the shortest? Surely not, for all we have done is to

confirm the axiom, without providing a justification. Similarly, the drawing of a triangle

confirms that the sum of three angles is equal to 180°, but it does not prove the theorem.

Cf. WLIII, §§ 305, 314; Anti-Euclid, pp. 208-209. Frege has a similar claim in “Logic in

Mathematics” (1914), 1997, p. 310. Their point is that we can be immediately aware of

conceptual truths, but we are not immediately aware of their objective grounds. Cf. WLIII,

§§ 313, 315.6, 316, note 1.

(3) Bolzano claims (Anti-Euklid, p. 209) that in a scientific presentation, the insight

(Einleuchten) of a true proposition is not immediate but must be established by means of

derivations from simpler truths. We must be able to perceive the correctness of such an

inference without seeing the figure to which it refers. In other words, clear propositions, if

they are conceptual, are not necessarily intuitively self-evident, but they are clear to us

because they are inferred from other, more basic truths; cf. WLIII, §§ 300ff., § 315.14. Cf.

also Appendix (1810), § 10, where he claims that one can very easily test the results of

mathematics by intuition and experience and that the axiom of the straight line being the

shortest between two points is proved by everyone by innumerable experiments long

before it can be proved by deduction. He adds that the well-known obviousness of

mathematics gradually disappears when experience is lacking. So a proposition is more

likely to be self-evident if it is inferred from others or tested empirically.

But although Bolzano claims that we have immediate knowledge of the primary truths of

reason, or conceptual truths, he does not specify the nature of this knowledge, except to


considers as mediated through mere intuition but not mediated throughany other judgments are surely altogether mediated, although it may betrue that we rarely, or never, become clearly aware of the premisses fromwhich we derive them. For how are we supposed to explain the way inwhich a judgment of the form: A is B, where A and B are pure concepts,is mediated through a certain intuition, that is, a mere simplerepresentation which has a single object, from whose properties weshould also think of a connection between a representation [175] x andthe representations A and B?479 It is said that the representation x canmediate the judgment: A is B if, on the one hand, we are aware that therepresentation x is subordinate to the representation A and if, on the otherhand, we can find the mark B in it. Granted that this is sufficient, are notthe perceptions x is an A and x is a B, real judgments?480 So if its

say that the principle of sufficient reason is the first expression of our awakening reason

to always inquire about the ground; cf. WLIII, § 311.2. Bolzano qualifies our capacity to

come to know a truth or make a true judgment as a “feeling for truth” (WLIII, § 316.3,

316, note 1) through which we can come to know a truth without coming to know its

ground. And he assumes that there must be primary propositions, whose truth we

recognize “from them themselves” (Beyträge, II, § 19; WLIII, § 300.9) or by

understanding the meaning of their components, for otherwise we could neither make any

grounded inferences nor show that some statements cannot be grounded. Cf. WLIII, §§

300.9, 316, note 1. Cf. Von Duhn (2002), op. cit.479

But in Kant’s view, intuitions are not simple; cf. my notes 111-115 (NAK, part I) and

Přihonský’s discussion of intuitions, pp. 15-17 (ibid.).480

Přihonský uses a Bolzanian claim from the Athanasia (pp. 54-55), according to which

judgings such as “I have a representation of something red” are perceptions asserting the

presence of a representation. We may ask whether perceptions can be assertions, but

according to this view in the Athanasia, judgings are seeings because they express our

immediate awareness of an intuition, and seeings are judgings, because we are

immediately aware of the truth or falsity of our perceptual judgments. In addition (cf.

WLIII, § 300.12), perceptual truths can be intuitively self-evident, although Bolzano

restricts the self-evidence to immediate judgments where something is predicated of the

first-person subject.

But on Přihonský’s own view, intuitions (or perceptions-he does not distinguish between

them) provide reliable information about the external world. And we can directly infer the

existence of the (represented) object from the existence of its representation, because an

intuition only has one object; cf. NAK, pp. 49-50, and my notes 160-162. So he claims

that our perceptions are true judgments, and this claim is problematic, to say the least, for

how does he account for perceptual illusions or hallucinations? Cf. my note 607, below.

Cf. also Bolzano’s genetic explanation of recollections as judgings: a recollection consists

of a judging and an internal relation between an intuition and its trace. Bolzano also says

that the trace we observe grounds our intuition. If we can represent the trace as a

grounding or foundation, a cognitive act or judging is involved in the formation of a

recollection; cf. WLIII, §§ 283.4, 284.2. But in the Athanasia, Bolzano claims that

recollections have a perceptual base: we notice a perception of the trace left by a past

representation; cf. Athanasia, sec. 5, pp. 151ff.; also WLIII, § 286, note.

New Anti-Kant 211

mediation occurs in such a way, can we tell whether the judgment: A is Bis not mediated through other judgments? In addition, we alreadymentioned that we do not see how these two judgments are sufficient fordeducing the universal judgment: all A are B.481

5. We concede that philosophical proofs rarely or never have a highdegree of reliability and visibility, such as certain mathematical proofsdo, but we believe that this distinction is based merely on a difference indegrees and that there are particular philosophical principles which arenot only equal to, but even surpass, mathematics in this respect. Is notmany a logical principle, for example, in the doctrine of inferences (eventhough it surely cannot be proved from intuition) proved more clearlythan many a mathematical truth, for example, the truth of the irrationalrelation between the diameter and the circumference?

[176] Because the Critique of Pure Reason teaches that we do not infact know anything at all in its pure and speculative use, one wonders ifwe are not at least permitted to make hypotheses? (p. 577).482 And theCritique replies: “not as propositions which we justifiably assume with adegree of probability, since for a hypothesis of such a kind its possibility

must be proved, as well as its adequacy for explaining what is given.”483

But Kant continues:1. We cannot assert this possibility of a single object with any new

and not empirically given property, for example, of an explication that iscapable of intuiting without sense, or a duration that is not in time, or our

soul as a simple substance, and so on.484

Anyone who needs a mere idea

481

Herleitung, which is how Bolzano calls his deductio (or justification) of axioms; cf.

Beyträge, II, § 21; cf. also Von Duhn (2002), op. cit.482

Cf. KrV, B797, but Kant says that through the critique of our reason we finally know

that we cannot in fact know anything at all in its pure and speculative use. Přihonský fails

to mention Kant’s important claim that a hypothesis is an opinion which is “connected as

a ground of explanation with that which is actually given and consequently certain”

(B798). This explanatory nature of hypothesis was also exploited by Bolzano when he

claimed that in an explanation, a statement p is true on the grounds of a second statement

q, because q explains why p is the case. Cf. WLII, §§ 168, 201.3-4; BolzanosWissenschaftslehre in einer Selbstanzeige, p. 82. Cf. Von Duhn (2002), op. cit. On the

explanatory nature of hypothesis or abduction, cf. C. S. Peirce (1876), “Deduction,

Induction, and Hypothesis”, op. cit., pp. 186-199.483

This quote does not appear in the Critique, nor is there a corresponding statement.

Přihonský probably summarises the third section on “The discipline of pure reason with

regard to hypotheses” (cf. KrV, B797).484

Cf. KrV, B798-799. Kant’s point is that we can only make explanatory hypotheses

about possible objects of experience, “using the conditions of possible experience as

conditions of the possibility of things, but it is by no means possible for [our reason] [. . .]

to create new ones, independent of these conditions, for concepts of this sort [. . .] would


of reason for the explanation of a thing in nature attempts to explainsomething he does not adequately understand by means of somethingabout which he understands nothing at all. Even the wildest hypothesis,as long as it is physical, is more tolerable than a hyperphysical

hypothesis, that is, the appeal to a divine author (p. 578).485

2. A hypothesis which is inadequate and in need of auxiliaryhypotheses arouses the suspicion of being a mere invention. Such is thehypothesis of an unlimitedly perfect cause of the world which requiresstill further hypotheses for explaining the world’s many evils andimperfections. [177] Such is the hypothesis of the simple self-sufficiencyof our soul if we are to explain phaenomena of growth and decay, and so

on (p. 580).486

In addition, it is an absurd proposal to make the reality oftranscendental ideas merely probable, just as if one thought to prove aproposition of geometry as merely probable. Pure reason can cognize

everything only a priori and necessarily, or not at all (p. 581).487

However, although in the merely speculative questions of pure reason nohypotheses are allowed to ground propositions, they are nonethelessentirely admissible for defending them, that is, in polemical use, whenideas are contested which are presupposed in any issue that is not

speculative (ibid.). If, for example, we object that the growth and declineof our mental powers seems to contradict the immaterial nature of thesoul, you may put forward the hypothesis that our body is nothing but thefundamental appearance to which the entire faculty of sensibility and

therewith all thinking are related (p. 583).488

To the objection of thecontingency of conception, you may put forward the hypothesis that alllife is really only intelligible, not subject to temporal changes at all, andthat this [178] life is nothing but a mere appearance (sensible

representation) of the purely spiritual life (p. 584).489

Nonetheless, we

[. . .] be without any object.” (ibid.) We may think of the soul as simple, but the

assumption that the soul is a simple substance would be unprovable, as well as haphazard,

since “there can be no insight at all into the possibility of a simple appearance.” (B800)485

Cf. KrV, B800-801. Here Kant refers to transcendental hypotheses and rejects the use

of speculative reason for explaining nature by a transcendental ground of explanation,

such as an idea of reason or a hyperphysical ground.486

Cf. KrV, B802. Kant says that to make a (transcendental) hypothesis we must also

adequately determine a priori the consequences which are given, and for this we may

require auxiliary hypotheses. However, since each of these auxiliary hypotheses requires

the same justification as the underlying thought (which is inadequately grounded), they

can give no reliable testimony.487

Cf. KrV, B803.488

Cf. KrV, B806.489

Cf. KrV, B808.

New Anti-Kant 213

should not take these hypotheses as our true opinions but should abandonthem as soon as we have finished off the dogmatic self-conceit of our

opponent (p. 585).490

Comment. 1. It is true that proposing a hypothesis always requiresthe assumed object to have at least problematic possibility (i.e., that thisassumption does not contradict any truth we are familiar with). But Kantwrongly asserts that this assumption is cognized as possible in all

regards (absolutely possible), that is, we must first assure ourselves thatthere is no truth or even that there is no pure conceptual truth

contradicted by our assumption before we may venture to advance it witha greater or lesser degree of probability, depending on the nature of thecircumstances. For example, I can assume with a high degree ofprobability that the hundredth decimal place of the number π is a nine,because this is what I have found in mathematicians’ works on thissubject, although I don’t really know whether I am assuming an absoluteimpossibility. However, since Kant admits and even asserts specificallyand repeatedly that the assumption of a transcendental idea does notcontradict any truths we are familiar with [179] (or that the impossibilityof God or the soul is unprovable), the first reason for which God’sexistence or the existence of the soul as a simple substance cannot evenbe considered as hypotheses is thereby completely eliminated.491

2. The concept of a duration that is not in time is self-contradictory,because by a duration we understand nothing other than a being that isvalid throughout a certain period. However, such a duration is a veryreprehensible hypothesis.

3. We cannot accept Kant’s accusation (our readers already knowwhy) that whenever we use a transcendental idea (e.g., of God or thesoul) for the explanation of a thing in nature, we explain something wedo not adequately understand by means of something about which weunderstand nothing at all.

4. Nonetheless, there is something true in the assertion that thewildest hypothesis, as long as it is physical, is more tolerable than ahyperphysical hypothesis. Of course an appeal to God which occurs insuch a way that we wish to prove the nearest cause of a certain

490

Cf. KrV, B809.491

Cf. KrV, B799. Kant says that concepts of reason or transcendental ideas neither have

objects “in any sort of experience” nor “designate objects that are invented”. They are

thought-entities, “the possibility of which is not demonstrable, and which thus cannot be

used to ground the explanation of actual appearances through a hypothesis” (ibid.).

Therefore they are merely thought problematically in order to ground regulative

principles.


appearance as an immediate effect of God is reprehensible without anyproblems, and any hypothesis which places that nearest cause in certain[180] natural forces is more acceptable, however improbable it may be.Although an appearance may be inexplicable by means of known forces,we are not justified in claiming that it originates as an immediate effectof God, because we do not cognize all natural forces. Hence we reject theexplanation usually offered by theologians that a revelation or a miracleis an immediate instruction or effect of God. However, we may assumean effect intended by God, for example, an instruction we are supposed toreceive through it, without deciding anything about the physical cause ofan event. For further details on this, cf. the Theory of Science (§ 379note) and the Theory of Religion (vol. 1, § 174ff.).

5. Kant’s use of expressions reveals that he may have been aware ofthe exaggeration in the assertion that any hypothesis which is insufficientfor explaining a phaenomenon and is in need of certain auxiliaryhypotheses (further assumptions), arouses suspicions such that we shouldhave to abandon it and that any reliable explanation would consist of asingle and simple assumption. If my room is filled by a sudden lightwhich disappears as quickly as it appeared, and immediately afterwards Ihear a slowly fading crash, is it not a very probable [181] hypothesis thatthis was a thunderclap? A few moments later I see a neighbouring towerin flames: should I not be allowed to add to my previous hypothesis thenew auxiliary hypothesis that the lightning has struck this tower?Another second and the entire tower explodes in front of my eyes with aterrible bang: may I not venture the third hypothesis that this tower wasused for storing gunpowder which has now caught fire? –

6. In addition, it is not even true that the assumptions of a perfectbeing492 and of the soul as a simple substance are entangled bydifficulties whose removal requires our resorting to new hypotheses. Themany evils that we find in the world, the phaenomena of growth, and thedecline of our mental powers, can be sufficiently explained without anyhypothesis that we are not justified or even required to make, forexample, that every created being has and can have only a limitednumber of powers, that it gradually increases in perfection, and that thereis an interaction between these powers, and so on. Readers will notexpect that we elucidate all this in detail but will be content with areferral to Bolzano’s Theory of Religion (vol. 1, p. 286) and theAthanasia (2nd ed., pp. 84ff.). Then we may [182] judge whether the

492

Allvollkommen.

New Anti-Kant 215

hypotheses Kant finally accepts for polemical (apologetic) give us moresatisfaction than what is said in the above-mentioned books.

7. It is simply not clear that it is an absurd proposal to maketranscendental ideas or pure conceptual truths in general merely probable

because pure reason can cognize everything only a priori andnecessarily, or not at all. For Kant himself claims that God’s existencecannot be asserted or denied a priori, so we have precisely that casewhere probability inferences are appropriate, and does he not say(p. 529)493 explicitly: however much purposiveness you encounter in theworld, so much confirmation you have for the rightness of your idea (of ahighest intelligence) and (p. 477)494 that the proof of the purposiveness ofthe world always deserves to be named with respect and increases thebelief in a highest author to an irresistible conviction, and (p. 618)495 thatthe question (of God’s existence) is merely a problem for speculation forsomeone who does not recognize a moral law, and in that case it can besupported by strong reasons from analogy [183] but not with grounds towhich even the most obstinate scepticism must yield. Surely aconfirmation with degrees is nothing other than a probabilitystatement,496 and thus the Königsbergian sage admits explicitly thatwhich he calls absurd elsewhere (p. 581).497 However, it is not absurd tooccasionally accept merely probable proofs even in geometry and in theother mathematical sciences. Who would not immediately recall thebeautiful principles in the doctrine of prime numbers, which Fermatdiscovered through experiment and established as probable by examplealone?498 What is true is only that in this science we may not be satisfiedwith mere proofs of probability but must search for others, since we canassume in advance that a diligent search will finally result in a definiteproof for or against the truth of the proposition. But we have alreadypointed out that in a theological doctrine exactly the opposite should be

493

Cf. KrV, B727.494

Cf. KrV, B651.495

Cf. KrV, B857.496

I use “probability statement” for translating Schluss der Wahrscheinlichkeit. Schlusstranslates as “conclusion” or “inference”, but a conclusion drawn from a hypothetical

statement is not a probability conclusion, and an inference from a hypothesis is not a

probability inference but a measure of the probability we can allot to the support-relation

between the hypothesis and the total evidence at our disposal, i.e., [probability] qualifies

the evidence or the argument (the degrees of support and confirmation).497

Cf. KrV, B803.498

Cf. WLIII, 391.4, where Bolzano refers to Fermat’s discovery of the properties of prime

numbers by first observing a certain property in individual things and then searching for

the most general concept of those things which have the property in question.


certain in advance, so we ask once more: why should we reject groundsof probability when they are available?

On the nature of the proofs of pure reason Kant teaches (pp. 586ff.)499

as follows:1. The first rule is to attempt no transcendental proofs without having

first considered from whence we can derive the principles (premisses) forthese proofs and with what right one can [184] expect success ininferences from them (p. 588).500 But if such ostensible proofs are alreadygiven, then oppose the non liquet of your mature power of judgment totheir deceptive conviction, and even if you cannot yet penetrate theirdeception (p. 589),501 you do not need to concern yourself with thedevelopment and refutation of each groundless illusion, since you candispose of the entire heap of these inexhaustible tricks of dialectic atonce in the court of a critical reason, which demands laws (ibid.).

2. For each transcendental proposition only a single proof can befound. For every transcendental proposition proceeds merely from oneconcept and states the synthetic condition of the possibility of the objectaccording to this concept. Thus there can only be a single ground ofproof, because outside this concept there is nothing further by which theobject can be determined. If I am to draw an inference not from conceptsbut from the intuition corresponding to a concept, this intuition willprovide me with manifold material for synthetic propositions that I canconnect in more than one way, thus allowing me to reach the sameproposition by different paths (p. 589).502

3. The transcendental proofs of pure reason must never be apagogic

but always ostensive, since only the latter combine the conviction of truthsimultaneously with insight into its sources. [185] The apagogicalmethod of proof can be allowed only in those sciences where it isimpossible to substitute that which is subjective in our representations forthat which is objective, namely, the cognition of what is in the object.Where the latter is prevalent, frequently the opposite of a certainproposition either simply contradicts the subjective conditions of thoughtbut not the object, or the two propositions contradict each other only

499

Cf. KrV, B810.500

Cf. KrV, B814.501

Cf. KrV, B814-815. At this point Kant adds: “you still have a perfect right to demand

the deduction of the principles that are used in them, which, if they are supposed to have

arisen from pure reason, will never be provided for you.”502

Cf. KrV, B815.

New Anti-Kant 217

under a subjective condition (that is falsely held to be objective), inwhich case both propositions can be false.503

Comment. 1. It is an excellent piece of advice that before attemptinga proof, or more generally, before attempting the examination of aproposition in general, we should check first whether it is evidentbeforehand that we lack the necessary cognitions for this examination,and each thinker should take this advice to heart. However, this is notalways evident beforehand, and consequently it would certainly be askingtoo much that we should not begin the examination of a propositionbefore we can expect to complete it successfully. We definitely reject thesuggestion that a learned non liquet should counter each proof of atranscendental proposition without heeding it, and that all such inferencesshould be disposed of in heaps without even considering them. For [186]

Kant says (p. 566):504

“When I hear that an uncommon mind has demonstrated away the freedom of the

human will, the hope of a future life, and the existence of God, I am eager to read

the book, for I expect that his talent will advance my insights.” – Should this be

invalid just for the opposite case?

2. We do not assert that transcendental propositions can always onlybe proved in a single way, if by that proof we understand a mereconfirmation but not an objective justification.505 In our view, theground from which Kant derives this does not prove anything,since all proofs, even mathematical proofs, can proceed from pureconcepts, and even those proofs referring to what appears to the

eye come about quite differently from the way Kant shows. And ifevery proposition allows only a single proof from pure concepts,why does Kant allow two kinds of proofs and distinguish betweenapagogical and ostensive proofs, and not criticize the former fornot offering any certainty, but only for not offering any insightinto the source of truth?506

503

Cf. KrV, B819.504

Cf. KrV, B781.505

On this Bolzanian distinction cf. Von Duhn (2002), op. cit. In Přihonský’s view, Kant’s

proofs are mere confirmations; cf. on Kant’s distinction between direct (ostensive) and

indirect (apagogical) proofs, my note 506, below.506

Kant allows only one kind of transcendental proof of pure reason, namely, ostensive

proof, because it directly gives certainty, since the conclusion is drawn from a single

premiss. Ostensive proofs show that a proposition is true, whereas apagogical proofs are

refutations showing that the opposite is not true (cf. Vienna Logic, trans. Michael Young,

1992, p. 893 [German], p. 339 [English trans.]). Apagogical proofs are inappropriate for

transcendental purposes because although they can produce certainty, they cannot produce

comprehensibility (Begrifflichkeit) of the relation between a truth and its grounds.


3. If an apagogical proof is considered such that the truth of aproposition is inferred by proving [187] that the contradictoryopposite507 results in an inconsistency (and we conclude fromKant’s remarks on p. 591508 that he does not relate any otherconcept to this name), then we assert (based on what Bolzano hasshown in the Theory of Science, § 530) that all apagogical proofscan be avoided and converted into direct proofs by making a fewsimple changes. But we doubt very much whether it is true ofevery direct or nonapagogical proof that it reveals an insight intothe sources of truth (i.e., the why of a proof). The proof of the

Apparently, Kant distinguishes between certainty and understanding: unlike certainty,

understanding requires our cognition of the explanation, and since understanding

immediately proceeds from transcendental proofs, he accepts only ostensive proofs and

rejects apagogical proofs as unsuitable (because understanding does not immediately

proceed from them). That is why he says that “the transcendental proofs of pure reason

must never be apagogical but always ostensive, since only the latter combine with the

conviction of truth simultaneously the insight into its sources.” (KrV, B819) However,

apagogical (indirect and negative) proofs are acceptable in other sciences, including

mathematics, where they serve to prove the falsehood of the opposite (using modus

tollendo tollens). But in philosophy “that can easily deceive, for one can just as well err

oneself in the matter, even if one proves that the other errs-both can quite easily be

wrong” (Dohna-Wundlacken Logic, p. 749 [German]; Lectures on Logic [English trans.],

p. 483). So apagogical proofs show mistakes and prove propositions by reductio adabsurdum, but they do not prove a proposition “in connection with its grounds. That is an

ostensive proof; [. . .].” (Blomberg Logic, § 196) In ostensive proofs, the truth of a

cognition can be shown from the relation to the grounds, and in this sense they are genetic

proofs which offer insight into the source of the truth. In other words, we know that p on

the grounds of q where q explains why p is the case, so in an ostensive proof we

understand a proposition through its explanation. Thus Kant’s notion of proof is close to

Aristotle’s οτι (cf. the latter’s famous distinction between οτι and διοτι in PosteriorAnalytics, A13, 78a-b [1975] and J. Barnes’ commentary, p. 155) and, as Bolzano points

out: “the Greeks required that we only say of something that we know it, if we have come

to know it from its ground.” Cf. WLIII, § 316.4 and note 1. Bolzano rejects this

epistemological distinction (between true judgments whose ground we know or do not

know). He distinguishes between the objective or logical grounding of truths and their

subjective or epistemological grounding. Cf. WLII, § 198; cf. also ML, § 14; WLIV, §

525. But this distinction raises another epistemological problem: how do we become

acquainted with the grounds of a cognition other than through intuition or insight? How

do we know whether a true judgment is justified or not? Bolzano was aware of this

problem and adds that if our distinction is between two kinds of cognitions, the words

grasping (Begreifen) or insight (Einsehen) would be more appropriate than knowing

(Wissen), to qualify such Erkenntnisse as clear insights; cf. WLIII, § 316.4. However, the

epistemological question remains throughout his attempts to combine logic and

epistemology in the Theory of Science. Cf. Von Duhn (2002), op. cit.507

Contradictorisches Gegentheil.508

Cf. KrV, B818-819.

New Anti-Kant 219

first theorem of Euclid’s Elements (the possibility of anequilateral triangle) is a direct proof, but would anyone claim thatthis proof offers any insight about why an equilateral triangle ispossible, that is, why for every two given points a and b, there areone or infinitely many points c, so that the distance ac and bc issimilar to the distance ab? Could anyone imagine that the groundof the presence of such a third point lies in the fact that twocircles on the same plane cut each other at a and b with the radiusab? Is there anyone who does not see that exactly the reverse isthe case, since the first is a consequence of the second?509 – By acanon of pure reason, Kant explains (p. 595),510

“I understand the collection of a priori principles [188] of the right use of certain

cognitive faculties in general. [. . .] Now according to the proofs that have

previously been given, all synthetic cognition of pure reason in its speculative useis entirely impossible. Thus there is no canon for its speculative use at all. [. . .]

Consequently, if there is to be any legitimate use of pure reason and a canon of it

at all, this will concern only the practical use of reason, which we will now

investigate.”

Comment. Surely anyone will notice that after Kant unconditionally

denied our reason the faculty of a synthetic cognition of nonsensoryobjects, in most places, even where this question is discussed mostlaboriously, he also gave reasons which, if they were proofs at all, wouldunconditionally show the impossibility of such a cognition.511

509

Bolzano uses this example for explaining that some scientific proofs are not grounded

proofs (justifications) but only confirmations (certifications): if we reverse the order of

propositions in a proof, we do not obtain a grounded proof but merely a confirmation (an

ungrounded proof). Geometers inverted the order between the propositions in the proof of

the first theorem in Euclid’s Elements and thus did not state the correct ground of the truth

to be demonstrated. The truth that an equilateral triangle is possible contains the objective

ground of, or justifies the truth that, two circles cut each other, “but if one only pays

attention to the mere knowing, then the converse relation might quite easily occur”. Two

circles cut each other or have a point in common, because for every two points a and b,

there must be a third, c, which has the same distance ab from both a and b, so that ca = cb= ab. We can easily mistake a confirmation for a justification, however, by inverting the

order of the propositions in the demonstration. We know that an equilateral triangle is

possible by confirming this on the grounds of the truth that two circles cut each other. Cf.

WLIV, § 525; Anti-Euklid, pp. 210-211. Cf. Von Duhn (2002), op. ci

t. Cf. also Mancosu (1999), “Bolzano and Cournot on Mathematical Explanation, pp. 436-

437. Cf. also Scholz (1931, p. 226), who comments that since to this date we have no clear

concept of indirect proof, the presupposition for such a discussion is missing.510

Cf. KrV, B824.511

The canon of pure reason “does not serve for expansion, as an organon, but as a

discipline serves for the determination of boundaries; and rather than discovering truth, it

has only the silent merit of guarding against errors.” (KrV, B823) The canon of pure


Nonetheless he adds the limiting appositive: “in its speculative use”, inthe present passage we have emphasized, as well as on several earlieroccasions. This is all the more disconcerting since that expression, whichis on the whole obscure, is nowhere explained in detail. The speculative

use of pure reason is really supposed to mean the application of reasonfor discovering speculative truths; the practical use, its application fordiscovering practical truths. Let us [189] see whether our assumptionsare confirmed or contradicted in what follows.

Kant says (in the first paragraph on the ultimate end of the pure use ofreason):

“The final aim to which the speculation of reason in its transcendental use is

ultimately directed concerns three objects: the freedom of the will, the

immortality of the soul, and the existence of God” (p. 596).512END BLOCK

QUOTE Then he attempts to show that for all these three the merely speculativeinterest of reason is very small, since one would not be able to use the discoveries

that might be made in the investigation of nature. Thus their importance can really

only concern the practical.513 Everything is practical that is possible through

freedom, and freedom in a practical sense is the faculty which can be determined

independently of sensory impulses through motives of reason. This practical

freedom is proved through experience,514 since we often overcome sensory desires

and, for example, decide something that promises to be useful only in a remote

way. The question about transcendental freedom515 (i.e., the question whether

reason can initiate a series of appearances without all determining causes from the

world of the senses) has no practical interest. So only two questions remain with

[190] regard to which a canon must be possible: Is there a God? Is there a future

life?

Comment. It still remains unclear what Kant means by the practical

use of pure reason.516 We accept, however, that the decision concerningthe three questions about freedom, immortality, and God has nospeculative interest insofar as it has no influence on our investigation of

nature. We also accept that the question of there being a freedom in the

reason thus gives the principles for the correct use of pure reason, whereas the canon of

pure understanding (the transcendental analytic) gives the principles for the correct use of

the understanding, the cognitive faculty which alone is capable of true synthetic a prioricognitions. “But where no correct use of a cognitive power is possible there is no canon.”

(B824)512

Cf. KrV, B826.513

Cf. KrV, B828.514

Kant writes: “Practical freedom can be proved through experience.” (KrV, B830)515

Kant does not actually use the expression “transcendental freedom” but refers to “the

transcendental sense of the concept” of freedom; cf. KrV, B829.516

Kant’s point is that the canon of pure reason can be applied only in the context of the

practical use of reason. In addition, the canon of pure reason deals with two questions

concerning practical interest: “is there a God?” and “is there a future life?”

New Anti-Kant 221

sense of Kant’s transcendental freedom has no practical significance,that everything depends only on the existence of the so-called practical

freedom, that this is sufficiently proved by experience, and that there thusremain only two contentious objects which at the same time have thehighest practical significance: God and immortality.

In the second paragraph: On the ideal of the highest good as a

determining ground of the ultimate end of pure reason, Kant says that allthe speculative and practical interest of reason is united in threequestions: 1) What can I know? 2) What should I do? 3) What may I

hope? The first question is merely speculative, the second [191] questionis merely practical, and the third question is simultaneously theoreticaland practical (pp. 601–602).517 – There is a moral law (as everyone’sconscience testifies) which commands absolutely and not onlyhypothetically that something ought to happen. This moral law requiresus to do that through which we may become worthy of being happy

(p. 603).518 – And now the question arises: if I behave so as to become

worthy of happiness, may I also hope to partake of it?519 I (Kant) say thatjust as the moral principles are necessary according to reason in itspractical use, it is equally necessary to assume according to reason in itstheoretical use that everyone has cause to hope for happiness in the samemeasure as he has made himself worthy of it. But this may be hoped foronly if a highest reason commands according to moral laws (that is, anideal of the highest good which at the same time grounds nature as itscause) and if a future world is presupposed for us, since in our presentlife we cannot find such a connection between virtue and happiness. ThusGod and a future life are two presuppositions that are not to be separatedfrom the obligation which pure reason imposes on us (pp. 604–605).520

517

Cf. KrV, B832-834.518

Cf. KrV, B836.519

Kant writes: “now if I behave so as not to be unworthy of happiness, how may I hope

thereby to partake of it?” (KrV, B837).520

Cf. KrV, B838-839. Kant’s point is that the ideal of the highest good (or the highest

reason) brings together happiness and morality: “how the consequences of [moral] actions

will relate to happiness is determined neither by the nature of the things in the world, nor

by the causality of the actions themselves and their relation to morality; and the necessary

connection of one’s hope for happiness with the unremitting effort to make oneself worthy

of happiness cannot be cognized through reason if it is grounded merely in nature but may

be hoped for only if it is at the same time grounded on a highest reason [. . .]. I call the

idea of such an intelligence, in which the morally most perfect will [. . .] is the cause of

all happiness, insofar as it stands in exact relation with morality (as the worthiness to be

happy) the ideal of the highest good. [. . .] Now since we must necessarily represent

ourselves through reason as belonging to such a world [. . .], we must assume the moral


From these (Kant thinks) follow God’s qualities, for example,omnipotence, so that the whole of nature and its relation to morality inthe world is subject [192] to him (p. 608),521 and so on. So it is purereason alone, but only in its practical use, which has the merit ofconnecting to our highest interest a cognition that mere speculation canonly imagine but never validate. Thereby reason turns this cognition notinto a demonstrated dogma but into an absolutely necessarypresupposition for its most essential purposes (p. 610).522

Comment. 1. We can now see that by cognitions which pure reasonobtains through its practical use, Kant does not understand, as wepresumed, certain practical propositions but the same theoretical

propositions about God and our soul which he claimed to be unattainablein their speculative use. A cognition is obtained through the practical use

of reason if it is derived from a practical premiss, although it is atheoretical truth as such. It is unattainable by a purely speculative use ofreason if there is no way it can be derived from purely theoreticalpremisses. Here the question is, above all, how such an assertion wouldfit with the ground that was previously offered for explaining theimpossibility of a synthetic cognition of nonsensory objects. Thisimpossibility was merely based on the fact that we [193] have nointuitions of such objects. Will this ground be removed by adding apractical premiss? Certainly not, so why is it that a cognition becomespossible, as it were, by means of this substitute?

2. It is true that moral laws are not hypothetical in the sense that theyare not bound by the condition: “as long as we feel like it”. Nonethelessthere is another condition they are subject to, that commands us to want

and to attempt only that which we consider as conducive to general well-

being.523 So a moral law depends on the presupposed problematic

world to be a consequence of our conduct in the sensible world; and since the latter does

not offer such a connection to us, we must assume the former to be a future world for us.”521

Cf. KrV, B843. Here Kant refers to the standpoint of moral unity and argues that if we

assess the cause that alone can provide us with the appropriate effect (i.e., an obligating

force as a necessary law), then there must be a single supreme will (i.e., a unity of

purposes), and this will must be omnipotent.522

Cf. KrV, B846.523

Přihonský’s interpretation of Kant’s moral law is problematic, for a moral law can tell

us what we ought to do, but not what we want to do. At KrV, B835, Kant says that moral

laws determine the use of the freedom of a rational being in general without regard to

empirical motives, i.e., happiness. These laws are expressed by an “ought” and thereby

indicate the relation between an objective law of reason and a will that is not necessarily

determined by this law (because of its subjective constitution); cf. Grounding for theMetaphysics of Morals, p. 413 (academy edition). So moral laws (or imperatives)

recommend actions to the human will (against the latter’s empirical inclinations) in order

New Anti-Kant 223

possibility of an action and on our insight into its expected success. Ofcourse Kant would never admit this claim (especially the last part), butwe think that mere common sense (to which Kant refers, after all,because of the existence of a moral law) decides entirely in our favour.For if common sense is supposed to determine the good or evil nature ofan action, it tends to consider the advantages or disadvantages which canreasonably be expected from such an action, that is, their influence on theweal or woe of living beings. This shows clearly that common senserecognizes the law of promoting general well-being or universal

happiness [194] as the basis and the condition of all moral actions or asthe supreme moral law. Although in the Critique of Practical Reason

Kant rejects all principles of happiness, his reasons really only concernthe principle of personal happiness,524 which is certainly a false and

to achieve the highest possible good (i.e., happiness). Thus, if you want to achieve x(happiness), then you should do y. If this end is presupposed as possible, then the law

which recommends that we achieve this end by following a certain course of action is a

hypothetical and problematic imperative. If this end is presupposed as actual for all

rational beings, the law which recommends it is a hypothetical and assertoric imperative

(ibid., p. 415). However, an apodictic moral law (or categorical imperative) is one that

can be turned into a universal rule of human conduct: “act only according to that maxim

through which you can at the same time will that it should become a universal law” (G, p.

421). This moral law is not hypothetical because it does not impose a conditional

necessity of action. Rather than telling us to do x in order to achieve a certain end, it

eliminates this necessity in terms of means and ends and tells us to do x if and only if x is

universalizable; i.e., before you do x, consider what would happen if everyone did it. It is

another question which maxims could (not) or should (not) be made into universal laws of

human conduct.

Concerning Přihonský’s view (or at least his interpretation of Kant) that the moral law can

tell us what we want to do, it is probably based on Bolzano’s view that deontic

propositions must necessarily be resolutions (Willensentschliessungen) made by a rational

being. Yet Bolzano holds that “[t]he concept of obligation in my sense applies only to

actions, or rather to mere resolutions of rational beings, [. . .]” (WLII, § 144). So he not

only relates the concept of sollen to the concept of wollen, but also his notion of

obligation is determined by his notion of the will, contrary to Kant, for whom the will is

determined by the notion of obligation. But Kant and Bolzano have different conceptions

of the will. Whilst Kant holds that the will is related to the faculty of desire (higher or

lower, depending on whether the will is determined by moral or empirical principles (cf.

KpV, I, bk. 1, §§ 7-8), Bolzano’s notion of will is related to the notion of attention. It

seems that, for Bolzano, the will has a cognitive feature, as a mental act which influences

the formation of representations, and is linked to attention and perception (cf. WLIII, §

286.5). On this point the two authors’ intentions are completely askew, for Kant would

object that “[h]owever many natural grounds or sensible stimuli there may be that impel

me to will, they cannot produce the ought but only a willing that is [. . .] always

conditioned, whereas the ought pronounced by reason opposes this conditioned willing

with [. . .] prohibition and authorization.” (KrV, B576)524

Selbstbeglückung.


reprehensible principle. However, Kant raises no valid objections to theprinciple of universal happiness.525 For example, he asserts that it isunsuitable as a law of practical will because the prescriptions derivedfrom it rest on mere experiential data and are thus shaky, since theydepend on each person’s variable concepts of happiness. For this reasonthese prescriptions cannot be universally valid for all times and foreveryone, as is required of moral laws (Critique of Practical Reason, vol.4, p. 318).526 – But we reply: It is true and undeniable that in order todecide whether some actions are conducive to universal happiness or not,various insights are presupposed that mainly rest on experience. Henceone or several experiential premisses are required to derive a moralprescription from the law of universal happiness. But this fact alone doesnot render that principle either [195] false or useless, unless we findexperience and its judgments reprehensible as such.527 Even from Kant’sown principle it is difficult to determine what is dutiful in each individualcase, without drawing on experience! To meet his requirement, that is, toalways choose that action which can be established as a universal law, isit not necessary to know which action such an attribute befits?528 But this

525

Although Kant does not consider happiness as a principle for determining moral action,

he considers it as indispensable for the highest good (which consists in a worthiness to be

happy, or morality, as well as in actual happiness); cf. KrV, B832. The imperatives

concerning the pursuit of happiness are hypothetical counsels of prudence, depending on

what we count as belonging to our happiness. Eudaimonists do their duty “only by means

of the happiness they anticipate”, whereas duty commands the will directly and is not

directed towards any end outside itself (cf. Metaphysics of Morals, p. 377).526

Cf. Critique of Practical Reason, pt. I, bk. 1, ch. 1, § 8, theorem 4, remark 2, where

Kant distinguishes between the principle of happiness and the moral law as follows: the

principle of happiness can indeed give maxims, but these maxims are inappropriate for

being laws of the will, even if their object were universal happiness, since the cognition of

happiness is merely empirical and depends on each person’s variable opinion. Hence it

can give general rules (that hold for the most part) but not universal rules (that hold

necessarily and are universally valid). Therefore no practical laws can be based on the

principle of happiness, for “it does not prescribe the same rules to all rational beings, even

though all rules stand under the same name, i.e., happiness. The moral law, however, is

thought as objectively necessary precisely because it holds for everyone who has reason

and will.” (ibid.)527

Kant does not hold the principle of happiness to be false or useless, although he claims

that, because of its empirical base, it cannot be made into a universally valid law; cf. my

notes 164-166, above.528

Cf. Kant’s categorical imperative: “act only according to that maxim through which you

can at the same time will that it should become a universal law” (G, p. 421), i.e., choose

to act on a maxim which you can turn into a universal rule of conduct; cf. also my note

164, above. Přihonský rightly points out that we lack criteria for determining precisely

which maxims are appropriate for universalization. In this respect, Kant’s theory raises (at

least) two questions: (1) which rules of conduct that may suit one or several individuals

New Anti-Kant 225

cannot be determined merely by considering a certain action, at least notalways. There is hardly an action (unless it be commanded by thesupreme moral law) that is universally obligatory for everyone and underall circumstances, since an action is a duty always only for those whofind themselves in the same situation.529 Thus we must be familiar withthe situation and the particular circumstances in which a case occurswhen all those who find themselves in it must obey the same law. Buthow should we establish this if not through experience? We shall soonsee that Kant’s attempt to decide the question a priori and from acontradiction into which we are led if we act against a law of duty isdisastrous in the highest degree.530 Kant unwisely concludes that by

can (or cannot) be made into universal rules of conduct? and (2) which rules of conduct

that may suit one or several individuals should (or should not) be made into universal

rules of conduct?529

An action is not necessarily a duty always only for those who find themselves in the

same situation: we might argue that an action is a duty for an individual x if and only if it

is what x (i.e., an agent who has and exercises certain moral features) would do in certain

circumstances y, if and only if he is motivated to carry out that action because it is a right

action. Our account of duty and our criterion for identifying certain actions as duties may

thus depend on (a) our motivation and (b) our own ideas of what a right action is (i.e., our

moral sentiment), as well as on the situation we find ourselves in. For example, if I am

looking for a reason to explain the fact that I am looking after my parents, do I say that I

act out of a sense of duty because I find myself in those circumstances or because I think

it is the right thing to do? We could even go one step further and ask whether acting from

a moral sentiment (of what is right) is a sufficient reason for acting from duty, without

referring to our (familiarity with) particular circumstances. Thus in the Grounding for theMetaphysics of Morals Kant attributes “genuine moral worth” to the third philanthropist’s

actions because “he does good not from inclination but from duty.” (academy edition, sec.

1, pp. 398-399) (the third philanthropist is the impartial one). To act from inclination is

acratic, whereas to act from duty is to act by choice (and hence by reason). Thus he

qualifies duty as an action that combines reason and the moral sentiment to do what is

right, e.g., “to help others where one can”. Admittedly, this interpretation of the Kantian

passage may be stretching things a bit, but it is consistent with Kant’s view that the ideal

of the highest good is a combination of happiness and morality; cf. KrV, B838-839, and

my note 521, above.530

Přihonský’s objection is a bit off the point: Kant claims that there are cases where the

universalization of a maxim is inconsistent and necessarily contradicts itself and that these

cases involve actions contrary to duty, such as lying, cheating, or stealing (cf. G, pp. 397-

398). For example, I can by no means will a universal law of lying, for by such a law there

could really be no promises at all, and “my maxim would be bound to annul itself as soon

as it was made a universal law” (G, pp. 402-403). Thus he does not claim that we are led

into contradiction by deviating from the law of duty but that as a result of universalizing a

maxim, such as “I may lie whenever it serves my interest”, the rule would self-destruct. If

lying were a universal rule of human conduct, it would lead us into a contradiction:

nobody would bother making promises and other statements, since no one would believe

them. Who would act on a rule that he does not believe to be true?


presupposing the principle of universal well-being, moral laws are onshaky ground since different persons have different concepts ofhappiness and its object. Inadequate concepts of what [196] truly makesus happy may only be corrected. Besides, people usually err only whenthey are blinded by passion.531 – In addition, the assertion is false thatlaws of duty are of such a kind that each of them is necessarily valid atall times and for everyone.532 On the contrary, there are only a few morallaws which hold universally and without exception.

At this point our readers may be interested to hear whether and howthe principle of universal well-being can be derived from clear premisses,a principle which common sense has recognized with much certainty,according to an innate feeling for truth.533 – Bolzano attempted such adeduction in the Theory of Religion (vol. 1, § 87) and we believe that hesucceeded. He starts out from two concepts that are irrefutably right,namely, that all moral obligation must be grounded by an action thatbinds unconditionally, without exception, and that the action must also bepossible. So if we examine all the ways of acting (of finite beings) inturn, among them we will surely find that which is possible as such, that

531

Here Přihonský seems to agree with Kant’s point that we should act from duty rather

than from inclination, since moral actions proceed from reason and not from weakness of

the will; cf. G, sec. 1, pp. 398-399, and my note 529, above. Besides, we do not just err

because we are blinded by passion: an erroneous act may be a consequence of

manipulation or false information; for example, someone may be sentenced for a crime he

has not committed because the evidence points to the contrary, or because he was set up.532

Cf. my notes 526 and 529, above. Kant claims that the moral law is universally valid

(i.e., it is a universalizable maxim) and holds necessarily for everyone who has reason and

will (cf. Critique of Practical Reason, § 8, remark 2). He uses the singular case when

referring to this law (the categorical imperative). Rather than referring to laws of duty,

Kant distinguishes between two functions of duty: (a) acting from a sense of duty or from

respect for the law (i.e., morality proper), or acting because we believe that the action is

right, and (b) acting according to duty (i.e., legality). We can act according to duty with

maxims determined by an inclination, but a moral action follows only from duty, i.e.,

acting from maxims conformable to the law (cf. KpV, pt. 1, bk. 1, ch. 3, p. 82).533

Bolzano claims that we have a “feeling for truth” (Gefühl für die Wahrheit) (WLIII, §

316.3, § 316, note 1) through which we can come to know a truth without coming to know

its ground; i.e., he claims that we can have a direct insight into perceptual as well as

conceptual truths. Since moral principles are conceptual truths, we can therefore cognize

them with certainty through a feeling for truth, except for the grounding principle or

supreme moral law which we must show to be deducible from other truths in order to

prove its objective ground (as Přihonský does). For although we can be immediately aware

of conceptual truths, we are not immediately aware of their objective grounds, and we

have a clear cognition of a truth if and only if we are aware of its ground. In other words,

conceptual truths may be subjectively plausible (einleuchtend) and even certain, but we

still need to prove their objective grounds. Cf. WLIII, §§ 300.9, 305, 314, 316.3; Anti-Euklid, pp. 208-209; Beyträge, II, § 21. Cf. Von Duhn (2002), op. cit.

New Anti-Kant 227

is, for human beings, and has unconditional validity, without exception.The latter property will serve as a mark by which we can easily cognizethe supreme moral law.534 All thinkable [197] actions which are possiblefor us apply to mere changes that we cause partly in inanimate beingsand partly in animate beings. As far as the latter are concerned, ouractions can only consist in effects on their faculties of cognition, desire,

or sensation. There are, of course, a number of commands concerning theinanimate world. But anyone can see that these are only conditional, forif we ought to carry out or not carry out certain changes in the inanimateworld, then this should happen only because of the influence that thesechanges have on living beings. Therefore the action demanded by thesupreme moral law and which is the ultimate ground of all moralobligations can only be among the effects on living beings. But evenamongst the effects on the cognitive faculty, there is not a single onewhich is commanded unconditionally. For that we should strive topromote the cognition of truth and keep error under control (which is themost likely candidate for an unconditional rule) has a reason, which isevident from the fact that we are ordered not to neglect the more usefultruths whilst we are spreading less useful ones. Therefore the spreadingof truth is demanded only because of the benefits it can create.Concerning the effects on the faculty of desire of living beings, there isprobably only one duty without any exceptions, [198] to act in such away that everyone should want what is morally good and not what ismorally evil. But this rule cannot be considered as the supreme morallaw, because it is a mere tautology, since we call morally good or badprecisely that which we should or should not do. And to derive any dutyfrom this proposition, we would have to presuppose another proposition,namely, the determination that this or that is morally good or bad. Only asingle kind of effect remains to be considered, that which refers to thefaculty of sensation. There is no other place for an unconditionalobligation if not here. But there are two ways in which we can have aneffect on the faculty of sensations of living beings: we can arouse in themeither pleasant or unpleasant sensations. Since nobody would want to saythat the arousal of unpleasant sensations is an unconditional command, itmust therefore be the arousal of pleasant sensations. – Although here wemust confront the objection that even this rule has exceptions, since thereare quite a few pleasant sensations which we deny both ourselves andothers and quite a few unpleasant ones which we must bring about. But atthe same time, it becomes clear after a bit of thinking that this

534

Bolzano’s account of the supreme moral law commands the greatest possible promotion

of general well-being (i.e., that we produce the greatest possible good); cf. WLIV, § 395.


exception—as far as there is an obligation—can only be illusory. Forduty only forbids a pleasant sensation if it cannot be granted withoutcreating even more unpleasant sensations [199] for oneself or others.Likewise, it demands unpleasant sensations only if they are a means forpreventing an even greater unhappiness. And therefore, even in the caseof such illusory exceptions, we should basically act in such a way as toproduce the greatest possible amount of pleasant sensations (pleasure)

or happiness. But now we may ask whether the supreme moral law onlydemands that every one of us should merely take care of increasing theirown happiness or try to promote everybody’s happiness (as much aspossible). If it were the former, there would be no difference betweenduty and wish, and the difference between virtue and vice would consistonly in a wise or unwise choice of the means for the same ends. Certainlythe supreme moral law demands that we do not merely promote our ownhappiness, but the happiness of all, as much as we can. Thus we couldcall it the principle of the promotion of general well-being. It is mostunjust to confuse the latter with the principle of eudaimonism or personal

happiness535 by which we understand the error contested above, namely,to care exclusively for one’s own happiness. – However, it is quite rightto act in complete accord with the principle of general well-being if, incase several actions are possible for us, we always only [200] choose thatthrough which the amount of producible pleasure becomes the greatest,no matter in which beings. Understandably, what matters here is not thekind of pleasure or happiness that is produced but the degree, which iswhy this law is rightly called the law of supreme pleasure and themaximum principle.*536 – Other formulations of the supreme moral laware either mere tautologies, such as “obey reason”, “live according tonature” (Stoics),537 “strive towards perfection” (Plato, Leibniz), “act

535

The ancient Greek notion of eudaimonia has been translated as “happiness”,

“flourishing”, or “well-being”, none of which correspond to Přihonský’s

Selbstbeglückung, which I translate as “personal happiness”. Contrary to Přihonský’s,

Kant’s, and other modern views, eudaimonia was considered a virtue (in the sense of a

happiness worth having) rather than a (selfish) pleasure. On the differences between

modern conceptions of happiness and ancient notions of eudaimonia, cf. J. Annas, “Virtue

and Eudaimonism”, (1998): pp. 37-55.

*536

Cf. G. Th. Fechner, “On the Highest Good”, Leipzig, 1846.537

Přihonský refers to the Stoic claim that commitment to virtue is grounded in human

nature. This statement is not a tautology, since it remains to be shown how this

commitment is related to a person’s moral and psychological identity. Besides, the Stoics

(especially Zeno) held that cosmic and human nature are (part-whole) related principles of

ethics: human nature is considered as a part of nature as a whole (the rational structure of

the world), and since cosmic nature is rational, human nature must also be rational. So the

New Anti-Kant 229

according to the maxim of your will, so that you could want rationallythat it should become a universal law” (Kant),538 and so on. – Or they donot give the real (objective) ground of our duties, as “obey God’s will”,“emulate God”, and so on.

We shall consider only Kant’s principle more closely. It says: “actonly according to that maxim through which you can at the same timewill that it should become a universal law” (Grounding for the

Metaphysics of Morals, vol. 4, p. 43, and Critique of Practical Reason,vol. 4, p. 130).539 We defined it as a tautology, for since the words“universal law” mean nothing other than what everyone ought to do, sowe should also grasp this principle in the following expression: “whatyou ought to do, should be constituted in a way that everyone ought to doit”.540 But no fruitful truth, let alone all practical truths or duties, can bederived from such a tautological proposition; it is therefore impossiblethat this is the supreme moral law. In addition, it is true only if it isexpressed with a certain limitation, roughly like this: “what you ought todo, under certain circumstances in a certain situation, should beconstituted in a way that [201] everyone ought to do it who happens to bein the same situation under the same circumstances”. But this statementpresupposes the practical truth: “in this or that situation you should dothis or that”, and that is why it cannot be acknowledged as the supremelaw. – Kant thinks that this tautological principle yields all duties, andattempts to show this by several examples (cf. Grounding for the

Metaphysics of Morals, vol. 4, pp. 44–45, and Critique of Practical

Reason, vol. 4, p. 126);541 but it will do if we examine the deduction inthe Critique of Practical Reason through which he tries to prove the dutythat one should not keep entrusted goods, on the grounds that theopposing maxim would be self-contradictory if it were established as auniversal law—for then nobody would entrust anything to anybody. Ifthis were right, the proposition expressing the permission to keepentrusted goods would certainly be false and exactly the opposite [202]

Stoic view of living according to nature is a lot more complex than Přihonský lets on; cf.

B. Inwood and P. Donini, “Stoic Ethics”, (1999), pp. 675-738.538

The formulation of the categorical imperative is “act only according to that maxim

through which you can at the same time will that it should become a universal law” (G, p.

421) and “act in such a way that the maxim of your will could always also hold as a

principle establishing a universal legislation” (KpV, § 7); cf. my notes 526, 529, 531, and

533, above.539

Cf. G, p. 421, and KpV, pt. 1, bk. 1, ch. 1, § 7, academy edition, vol. 5, p. 30.540

“Dasjenige, was du sollst, soll so beschaffen sein, dass es alle sollen.” For grammatical

and semantic reasons I have added [do].541

Cf. G, pp. 422-423; KpV, § 4, remark to theorem III, pp. 27-28.


proposition would be true.542 But no deontic proposition is constituted insuch a way that its contradictory opposite should really be called a self-contradictory proposition. The contradiction Kant has in mind here ismerely illusory. It is not even true that, as he claims, there would be noentrusted goods if there were no obligation to return them. For if it weremerely not obligatory to keep them, some people could still hope to getback the goods they entrusted to others. Then, assuming that there wereindeed no more entrusted goods, the rule: “you should not keep entrustedgoods” would not contain an internal contradiction because it only statesthat we should not keep a good if it has been entrusted to us; but it doesnot demand that we not keep an entrusted good if it has not beenentrusted to us. The rule would be self-contradictory only in the lattercase. In addition, we could prove the falsest propositions as obligationsaccording to this rule: for example, the proposition “as far as possible,you should ravage everything around you”. No matter how much effortpeople put into acting according to this rule, the world would continue toexist for a long time. But assuming that there were nothing left to ravage,this proposition would still not contain any internal contradiction since itdemands that we ravage only as long as there is something to be ravaged.But enough of that. [203] 3. In the same section (On the ideal of the

highest good), Kant declares it a necessity that we may hope that

everyone attains happiness in accordance with his virtue, without furtherexplaining why this is necessary. He merely says (p. 606):543 “it isnecessary that our entire way of life be subordinated to moral maxims;

542

Kant tries to show that the universalization of this maxim would be inconsistent

because it would destroy itself, i.e., that the same maxim, if taken as a universal law,

would necessarily contradict itself: e.g., before deciding to keep entrusted goods, I should

first consider what would happen if everyone did it. Entrusting goods (and doing business)

would be impossible. Kant shows that the practical result of such an action contradicts the

maxim an individual might want to act on, but this does not imply that the statement is

false and its opposite is true, or vice versa. Cf. my note 531, above.

The Bolzano-Přihonskian account of normative statements is problematic because

according to them, all propositions are either true or false. Deontic propositions, such as

“you should do A”, are practical truths with a theoretical base: we can only have an

obligation to do A if A is not impossible, so the practical truth that we should do A is

based on the theoretical truth that A is possible. And A is logically possible iff it is based

on the principle of noncontradiction; cf. WLII, § 200. However, normative statements or

statements expressing a wish or desire are neither true nor false, but in Bolzano’s account

there is no room for indeterminate propositions; cf. Von Duhn (2001), vol. 5, pp. 3-12,

and Kasabova (2002), “Is Logic a Theoretical or Practical Discipline? Kant and/or

Bolzano”, pp. 319-333.543

Cf. KrV, B841-842. The last part of the sentence reads “[. . .] an efficient cause which

determines for the conduct according to this law an outcome corresponding precisely to

our highest ends, whether in this or in another life.”

New Anti-Kant 231

but it would at the same time be impossible for this to happen if reasondid not connect with the moral law an efficient cause which determinesan outcome corresponding precisely to the conduct conforming to thislaw.” In the contrary case he says (p. 607)544 that “the splendid ideas ofmorality are indeed objects of approbation and admiration but notincentives for resolve and realization.” From this it follows that Kantheld it to be a proper impossibility to comply with the demands of morallaw unless one could hope for a reward, although he denies this in severalother writings and although this is wrong as such. – Elsewhere hetherefore draws an essentially different conclusion (e.g., Critique of

Practical Reason, vol. IV, p. 257)545 which roughly goes as follows: “thefact that there are certain obligations is undeniable, since we are aware ofthem in our inmost self or are aware that there is something we ought to

do. But what we ought to do must also be possible, that is, we must beentirely convinced that whatever practical reason [204] demands ought tohappen, is possible. Practical reason requires the realization of the

highest good. Hence the realization of the highest good must be possiblefor us. But this could not be the case if our will were not free, our soulnot immortal, and the world and its order not ultimately grounded in anomnipotent, omniscient, and holy being. Thus it is certain that our will isfree, that our soul is immortal, and that there must be an omnipotent,omniscient, and holy God.” Kant calls this kind of inference a demand orpostulate of practical reason.

Although Kant contradicts his own theory by granting reason all of asudden the prerogative to make right synthetic judgments—apart fromthis, is his argument otherwise correct?546 We think that it is not and

544

Cf. KrV, B841.545

Cf. KpV, pt. 2, bk. 2, ch. 2, sec. 4-6 (approximately). Kant says that there are three

postulates of practical reason: immortality of the soul, freedom, and the existence of God.

These are the necessary conditions of obedience of a human being to the moral law which

determines its will, and they must be postulated to establish our respect for this law and

thus the possibility of its realization. “The first derives from the practically necessary

condition of a duration adequate to the perfect fulfillment of the moral law. The second

comes from the necessary presupposition of independence from the world of sense and of

the capacity of determining man’s will by the law of an intelligible world, i.e., the law of

freedom itself; the third arises from the necessary condition of such an intelligible world

by which it may be the highest good, through the presupposition of the highest

independent good, i.e., the existence of God.” (KpV, 2.2.6., “On the postulates of pure

practical reason”, p. 133)546

The reason in question here is pure practical reason (the consciousness of the moral

law) to which pure theoretical or speculative reason is subordinated; cf. KpV, 2.2.2.3.,

“On the primacy of the pure practical reason in its association with speculative reason”,

pp. 120-122; cf. also KrV, B575, where Kant says that the limits of the speculative use of


deserves to be accused of circularity and of presupposing that which it issupposed to prove. For if, following Kant, we consider the highest goodas a virtue and a happiness in accord with the attained degree of virtue(Critique of Practical Reason, p. 245),547 it is obvious that we can onlysay that the highest good is demanded by reason and should be achieved,if we have proved [205] that this highest good is something which is

possible.548 For we have no obligation to achieve something impossible.But if we attempt to prove the possibility of the highest good, that is, toshow that there is a happiness in the world corresponding to each andevery degree of virtue, then we will see that this proof can be carried outonly by presupposing the existence of God, immortality, and freedom (cf.Bolzano’s Theory of Religion, vol. 1, § 167).549

4. In his outline of moral theology, Kant extols (p. 608)550 its peculiar

advantage over speculative theology, that it inexorably leads to theconcept of a single, most perfect and rational primordial being. He thenshows, by way of examples, how God’s unity, omnipotence, omniscience,omnipresence, and eternity result from his theory and concludes(p. 611)551 with the remark that “certain actions are not obligatorybecause they are God’s commands but, on the contrary, can be regardedas divine commands only because we hold them to be obligatory.” –

reason when determining “what is” (or what exists) do not hold for the practical use of

reason when determining what “ought to be”. Reason has causality (since it produces

imperatives for ruling our actions) but only in its practical use: “This “ought” expresses a

possible action, the ground of which is nothing other than a mere concept, whereas the

ground of a merely natural action must always be an appearance.” (ibid.) The concept

grounding a possible action is “the concept of a single original being as the highest good”

(B846).547

Cf. KpV, 1.2.2. “Inasmuch as virtue and happiness together constitute the possession of

the highest good for one person, and happiness in exact proportion to morality (as the

worth of a person and his worthiness to be happy) constitutes that of a possible world, the

highest good means the whole and perfect good, wherein virtue is always the supreme

good, being the condition which has no condition superior to it, while happiness [. . .]

always presupposes conduct in accord with the moral law as its condition.” Cf. also KrV,

B841-842.548

Kant postulates a supreme original good on which he grounds the reality of the highest

good, which is a combination of happiness and morality (the worthiness to be happy); cf.

KrV, B842.549

It seems that Kant and Bolzano agree on this point, for these are, precisely, Kant’s three

postulates of practical reason; cf. my note 546, above.550

Cf. KrV, B842-843.551

Cf. KrV, B847. Kant says that “we will not hold actions to be obligatory because they

are God’s commands but will regard them as divine commands because we are internally

obligated to them.” Kant’s argument is directed against “divine command ethics”, i.e., that

moral laws are derived from God’s commands.

New Anti-Kant 233

Although we fully agree with the last claim, we strongly doubt whetherwe can derive a clear concept and establish a convincing proof of God’sperfections from that practical postulate without the support of severalsynthetic propositions referring to a nonsensory object. [206] Whereinshould we posit an omnipotent being? Only in that God can do everythingwhich is not impossible as such. But how should we judge if something isimpossible as such without cognizing certain synthetic truths that do notrefer to mere objects of experience?

Excerpt from the third section “On opining, knowing, and believing”(p. 611).552 (According to Kant) opining is taking something to be truewith the consciousness that its grounds are objectively (i.e., for everyone)and subjectively (for ourselves) insufficient. Believing is takingsomething to be true on grounds we hold to be subjectively sufficient butobjectively insufficient. Knowing is taking something to be true ongrounds we hold to be subjectively and objectively sufficient.553 – In thetranscendental use of reason, opining is too little but knowing is toomuch. In a merely speculative regard, we cannot judge here at all, but ina practical relation, we can believe. Once a proposed end is carried out,that is, when we ought to act, it is necessary that we consider certainmeans as suitable for attaining this end. This taking to be true is onlysubjectively necessary but not objectively necessary if we alone do notknow of any other means; it is [207] thus a belief which is onlycontingent in our own view. Kant calls such contingent beliefs whichground the use of the means to certain actions pragmatic beliefs.554 –

But even if we cannot undertake anything in relation to an object andthe taking to be true is therefore merely theoretical, we can still oftenconceive of an undertaking if there is a means for determining the matter;thus there is in merely theoretical judgments an analogue of practicaljudgments, and we can call this taking to be true a doctrinal belief

“The doctrine of the existence of God and immortality belongs to doctrinal belief.

For the presupposition of a wise author of the world is a condition for the

552

Cf. KrV, B848. Přihonský refers to the title of the third section of “The canon of pure

reason.”553

Cf. KrV, B850. In addition, Kant distinguishes between conviction, or taking something

to be true on grounds which are held to be objectively sufficient, and persuasion, or

taking something to be true on grounds which are held to be only subjectively sufficient

(i.e., “if it has its ground only in the subjective constitution of the subject”). Such

judgments are only valid for ourselves, whereas judgments of conviction are valid for

everyone (cf. B848).554

Cf. KrV, B852. Kant also explains that we can determine the degree of strength of

pragmatic beliefs (i.e., whether they are mere persuasions or firm beliefs) by examining

the interest that is at stake, e.g., in the paradigmatic case of betting; cf. KrV, B852-853.


contingent but not inconsiderable aim, that of having a guide for the investigation

of nature. In addition, the outcome of my experiments so often confirms the

usefulness of this presupposition that I would say too little if I called my taking it

to be true merely an opining. Instead, even in this theoretical relation it can be

said that I firmly believe in God. In regard to God’s wisdom and in view of the

excellent equipment of human nature and the shortness of life which is so

inappropriate to it, we can likewise find sufficient ground for a doctrinal belief in

a future life.” (pp. 615–616)555–

But there is something unstable about this doctrinal [208] belief; it isentirely otherwise for moral belief. For there it is absolutely necessary

that something must happen, namely, that I comply in all points withmoral laws.556 And, according to all my insight, only a single condition ispossible under which this end can be attained, namely, that there is a Godand a future world. I also know with complete certainty that nobodyknows of any other conditions.557 So I will inexorably believe in theexistence of God and an eternal life. Hence no one can say he knows thatthere is a God and not even that it is morally certain that there is a God,but he can claim that he is morally certain that there is a God (p. 617).558

Comment. 1. Kant’s attempted definitions of the words opining,

believing, and knowing correspond neither to usage nor scientific ends.That which someone has through immediate perception, for example, apain he now feels, he cognizes from a ground which is sufficientlyperfect for himself but which is incommunicable to others and is thus notobjective. Hence we should say that he only believes he feels this pain,which is completely against usage. The definitions given in the Theory of

Science are far more right (§ 321)!559 –

555

Cf. KrV, B854-855.556

Kant uses the singular noun “moral law” (KrV, B856).557

Kant’s sentence continues, “that lead to this same unity of ends under the moral law”

(KrV, B856).558

Cf. KrV, B856-857.559

Bolzano’s account of believing and knowing is a variant of the view that knowing is the

highest degree of belief, the belief we hold with the highest degree of certainty, such that

x knows that p if and only if x believes that p is true and has good reasons for believing p.

The Bolzanian variant expounded in § 321 is this: I know that p if and only if I have

cognized the truth of p and understand it, so that my confidence in the explanation of p is

indestructible. I believe that p if and only if I have cognized the truth of p and understand

it, but my confidence in the explanation of p can be destroyed. Bolzano says that we know

something once we have scientific proof of it. For example, we know that the Pythagorean

theorem is true, “once we have been exposed to its proofs. For now we can cognize the

truth of this proposition in a way that assures us that we could not persuade ourselves of

its falsity even if we tried.” The condition sine qua non for my knowing that p is that even

if I wanted to, I could not judge that ¬ p (i.e., it involves making a second judgment qabout p). In addition, Bolzano claims that knowing is will-independent, whereas believing

New Anti-Kant 235

depends on the strength of our will, to hold the grounds for p (cf. also WLIII, p. 293). For

example, he says: “I believe that the heavenly bodies are populated, since I have decisive

reasons to form this judgment, although I do not doubt that, if I wanted, it would be

possible for me to persuade myself of the contrary.” When I believe that p, I am not

certain of the grounds for the truth of p, and consequently, I can be persuaded to believe

that ¬ p is true by concentrating my attention on true or apparent reasons against p. My

certainty of the truth of p depends on the intensity of the conviction with which I judge

that p is true. However, I believe that p if and only if the degree of my confidence in p is

less than certain, so that it is liable to destruction. Cf. Von Duhn (2002), op. cit.

Přihonský already mentions this Bolzanian position at the beginning of the NAK, p. 27;

cf. also my note 130. However, Bolzano improved his account of knowing in MiscellaneaPhilosophica by dissociating knowing from certainty (so that wissen is no longer

reducible to Gewissheit except in statements expressed in the first person singular). He

says that x may know that p without being certain that p. For example, a nervous examinee

may know the correct answers but may not feel fully confident. It seems that Bolzano

wanted to improve his distinction between knowing and believing, contrasting knowing

with error, whereas in WLIII, § 321, he had only contrasted knowing with ignorance (you

can’t have knowledge about things you ignore, though you can have beliefs about them).

Here his point is that my conviction or degree of certainty is neither sufficient, nor

necessary, for knowing. For example, I cannot know that World War II was fought in the

1980s, and my conviction that it was, is mistaken. On the other hand, my conviction (as a

nervous examinee) that I will fail my exam is unnecessary; cf. Philosopische Tagebücher1827-1844, II, BBGA, vol. 18.2, pp. 65-66. I thank Mark Textor for pointing out this

passage to me.

In addition, in WLIII, § 321, Bolzano objects to Kant’s notions of “objectively sufficient

ground” and “subjectively sufficient ground”. In Bolzano’s view, Kant’s grounds for

taking a judgment to be true are only subjective and not objective at all, since they are

always “derived from the nature and interest of the person making the judgment,

regardless of whether or not he followed correct rules of thought.” (§ 321) But Kant’s

account of opining, believing, and knowing is also a variant of the view that knowing is

the belief we hold with the highest degree of certainty. Hence he distinguishes between

opining, believing, and knowing as three modes of holding something to be true. The

latter are in turn determined as having either subjectively sufficient grounds (valid for

myself) or objectively sufficient grounds (valid for everyone): knowing has both

subjectively and objectively sufficient grounds, believing has only subjectively sufficient

grounds, and opining has neither (cf. KrV, B848; JL intro. IX. D, Dohna-WundlackenLogic, p. 732, my note 554, above). If we read the respective texts, we can easily perceive

the similarities between their accounts: “believing” is to hold something to be true on

grounds that are less than certain, i.e., only subjectively sufficient, or what Kant calls a

consciousness of the possibility of the opposite, or as Bolzano puts it, if I am not certain

of the grounds for the truth of p and consequently I can be persuaded to believe that ¬ p is

true, by concentrating my attention on true or apparent reasons against p. “Knowing” is to

hold something to be true on objective grounds. But how do we become acquainted with

the grounds of a cognition except through intuition or insight? Bolzano was well aware of

this epistemological problem, since he adds that in this case the words grasping

(Begreifen) or insight (Einsehen) would be more appropriate than knowing (Wissen) in

order to qualify such Erkenntnisse as clear insights (deutliche Einsichten); cf. WLIII, §

316.4 and note 1. Cf. on this epistemological problem in Bolzano’s logic, Von Duhn


2. It is a new mistake to claim that belief can only occur in a practical

relation (i.e., where we should act). [209] Since Kant cannot deny thatbelief also takes place in doctrines which do not influence our actions,for example, in regard to the question of whether planets are inhabited,he seeks to explain this by the assumption that in such cases we at leastimagine that our actions are grounded on taking something to be true, forexample, that we can place a bet on them. Is there anyone who does notsee how forced this explanation is?560 –

3. But the worst thing is that the case our philosopher presupposeswith such certainty of the so-called pragmatic belief, that on its accounthe even fabricates an analogon of doctrinal belief, does not really occuranywhere in the cold light of day. Our aim to attain a certain end cannotprovide a ground for believing that this or that means is the most suitableor the only one suitable for this end. Belief, as any other taking to be truewhich is not an immediate judging, arises only from grounds that rest onother judgings, and the mere will only mediately influences our judging,insofar as it determines us to direct our attention to these grounds.561 Butnot to mention that our wish to attain a certain end cannot immediatelyproduce the belief that a certain means is the most suitable one forattaining it, this wish cannot, at least if we are reasonable, [210] evencorrupt our judgment about the suitability of this end. But someone whoeagerly wishes to attain a certain end also wishes all the more eagerly tofind out the right means and will take care not to give a preference to onerather than the other in an obstinate way. Kant’s own example refutes

(2002), op. cit. Once again, Bolzano’s account is very close to Kant’s view that knowing

is taking something to be true on objectively sufficient grounds where “objective grounds”

are understood as those which are immediately obvious to us. Cf. on Bolzano’s unintended

proximity to Kant, Gerhardt Gotthardt (1909), Bolzanos Lehre vom “Satz an sich” in ihrermethodologischen Bedeutung, p. 31.

We can also safely reject Přihonský’s view that Kant’s definitions of opining, believing,

and knowing are against usage, for I hardly have scientific knowledge (wissen) of my

pain: I cognize that I have a headache if and only if my headache is immediately obvious

to me, and this ground is subjectively sufficient (for me), so that I hold my pain to be true

with a sufficient degree of certainty, although I can hardly communicate it to others or

know it to be universally and objectively necessary (i.e., holding for everyone).560

But Kant says that betting is a touchstone for determining whether our beliefs are firm

or mere persuasions; cf. KrV, B852-853, and my note 555, above.561

Cf. my note 523 on Kant’s and Bolzano’s contrasting notions of the will. Besides, Kant

does not say that pragmatic belief is grounded on our aim to attain a certain end, but that

it is grounded on our taking something (that is theoretically insufficient) to be true. Vice

versa, he claims that pragmatic belief grounds “the actual use of the means to certain

actions” (KrV, B852).

New Anti-Kant 237

him: “A doctor”, he says on p. 614,562 “must do something for a sickperson who is in danger, but he is not familiar with the illness. Heobserves the symptoms and then judges, because he does not know anybetter, that it is consumption. His belief is merely contingent even in his

own judgment; someone else might perhaps do better.” – So it is that adoctor should not convince himself he is dealing with consumption if it isstill uncertain whether this is the case. Instead, he should always recallthat there could be another judgment which is even more probable. Thuswe can see from this example that even though we occasionally expressourselves in this way, it is nonetheless so untrue to say “we must decide

so that we can act” that it is not unusual for us to make provisions in ourconduct for a case we do not even consider to be probable but which wedo not hold to be entirely impossible.563

4. If it were indeed the case, as Kant says, that belief in God andeternal life are necessary for virtue (but we hold that [211] this doublebelief is only conducive to virtue), it would then be our duty to direct theentire attention of our mind to the grounds supporting God andimmortality in order to strengthen this belief as far as possible. But westrongly doubt whether these efforts can bring about such a firm andunshakeable belief as Kant describes moral belief to be, if the contrary istrue, namely, that theoretical or speculative reason cannot offer anydecisive grounds for these two truths. For surely the weight of thesegrounds is not increased by mere will, unless it were to happen throughsome kind of self-deception.564

562

Cf. KrV, B852.563

Maybe Kant means that before acting on a maxim, we should first consider the

consequences of its universalization and then decide (since we have the capability of

choice [Willkür]) whether to act in accord with the moral law or not.564

Cf. my note 546, above, on Kant’s postulates of practical reason which determine the

will (for they are the necessary conditions for obeying the moral law). Virtue, for Kant, is

the supreme good or highest condition for good (cf. my notes 548-549, above), whereas he

considers virtues in the sense of such moral features as moderation, courage, and patience

as the means for the end of a good will (e.g., in his discussion of “duties of virtue”, MM,

p. 383; G, pp. 398-399; cf. also my note 528, above). He does not say that belief in God

and eternal life are necessary for virtue. Instead, moral belief (which is at issue here)

offers certainty on the basis of the three postulates of practical reason (the existence of

freedom, immortality, and God). This certainty is grounded on the inability to renounce

these postulates without becoming contemptible in our own eyes; cf. KrV, B856. And if

there is any self-deception, it could only arise from the combination of my moral belief

and my moral disposition, i.e., that my conviction is subjectively grounded on my own

moral certainty that there is a God (cf. KrV, B857). This is why Kant claims that “since

the moral precept is thus at the same time my maxim (as reason commands it ought to be),

I will inexorably believe in the existence of God and an eternal life.” (B856)


5. Of course things would be different if the necessity to obey themoral law which Kant mentions here were to occur in the strictest senseof the word. For that which is necessary in the strictest sense of the wordalso has reality. Thus all of us would indeed obey the moral law, andsince this (as Kant says) is not possible in any other way than bybelieving in God and immortality, this belief must be found in all of us(without our being able to explain the way in which it was broughtabout).

6. Finally, the distinction between the two propositions “it is morallycertain that there is a God” and “I am morally certain that there is a God”[212] seems to us akin to an empty subtlety. That of which I amcompletely certain, I also consider, precisely for that reason, assomething which is certain as such.565 But if the second expression isonly supposed to refer to something which is generally communicable

(since it is certain for all human beings), we accept that there areincommunicable certainties, but in our view the doctrines discussed heredo not belong to truths of this kind. For assuming that the certainty ofGod’s existence and our immortality merely emerges from ourconscience in the way taught by Kant, does not Kant say quite rightlythat this conscience is shared by all human beings?566 So why should wenot expect the same certainty that we have of those two truths to hold forall truths, in exactly the same way that we expect the same geometricprinciples we cognize as truths to hold for all others, since we maypresuppose the same representation of space for them?567

The architectonic of pure reason is the art of systems or sciences(according to p. 619).568 A system, however, is the unity of the manifold

565

Kant’s point is that my conviction is subjectively grounded on my own moral certainty

that there is a God and for this reason it is more precise to say: “I am morally certain that

there is a God”; cf. KrV, B829, B857.566

Kant does not use the word conscience (Gewissen) in this passage. Conscience, for him,

is our “inner judge” which warns (or threatens) us before we do a certain action, although

he does say that conscience can be considered as a support for the postulate of the

existence of God as an omnipotent moral being; MM, p. 439 (which implies that his idea

of God is rather Orwellian).567

Kant could reply that the certainty we have of mathematical or theoretical truths is

logical (and knowing can be communicated if it concerns objects of reason), whereas the

certainty we have of practical truths is moral, i.e., it depends on our moral disposition or

sentiment (Gesinnung), and here knowledge must be denied “in order to make room for

faith,” since “all practical extension of pure reason is impossible” (KrV, preface to B,

xxx). Nevertheless, it is not clear why moral truths cannot be communicated, even if our

own moral sentiment is incommunicable.568

Cf. KrV, B860. Kant’s architectonic refers to systematizing the sciences on the basis of

the rational concept or idea of the whole of a science, as well as to the outline of this idea

New Anti-Kant 239

cognitions under one idea, one rational concept of the form of a whole,insofar as this concept determines[213] a priori the domain of themanifold as well as the position of the parts with respect to one another.– Here Kant merely wanted to outline the architectonic of pure reason,that is, of a priori cognitions. All rational cognition is either cognitionfrom concepts or cognition from the construction of concepts. The first iscalled philosophical cognition and the second mathematical cognition.The system of all philosophical cognition is philosophy according to thescholastic concept, but according to the cosmopolitan concept (i.e., theone pertaining to that which interests everybody), we can definephilosophy as the science of the relation of all cognition to the essential

ends of human reason.569 Now, the highest essential end is none otherthan the entire vocation570 of man and the philosophy of it is called moral

philosophy. Philosophy has (Kant continues) two objects: nature andfreedom. The philosophy of nature pertains to everything that is; that offreedom or morals pertains to everything that should be. All philosophyis either cognition from pure concepts, that is, pure philosophy, orrational cognition from empirical principles, that is, empirical

philosophy. The philosophy of pure reason is either propaedeutic

(preparation), which investigates the faculty of reason in regard to all a

priori cognitions,571 and is called critique, or the system of pure reason,the whole philosophical cognition from pure reason and systematicconnection, and is [214] called metaphysics. This is divided into themetaphysics of the speculative and the practical use of pure reason and istherefore either the metaphysics of nature, also called metaphysics in thenarrower sense, or the metaphysics of morals, also called pure

morality.572 Metaphysics in the narrower sense consists of transcendental

philosophy, which considers only the understanding and reason itself inthe system of all concepts and principles that are related to objects as

itself (cf. also JL, VI, p. 49; § 3, p. 93). He considers that all a priori cognitions belong to

a possible system and this (architectonic) system is philosophy or the legislation of human

reason. This encyclopedic notion of philosophy as a system was rejected by Bolzano and

his school, who consider philosophy as a science of the objective connection between all

those truths into whose ultimate grounds we try to penetrate as far as possible; cf. “Was istPhilosophie” (posthumous manuscript), (1969), pp. 26, 30. But this organization of truths

in terms of ground-consequence is also a way of establishing a system! Cf. also Berg

(1997, op. cit.) who considers Bolzano to be a prescient encyclopedist.569

Cf. KrV, B866-867.570

Bestimmung.571

At KrV, B869, Kant refers to “all pure a priori cognition”, using the singular.572

Cf. KrV, B869-870.


such without assuming objects that are given573 (ontology), and thephysiology of pure reason, which considers the sum total574 of givenobjects, though only rationally. However, the use of reason in thisrational study of nature575 is either immanent, that is, it pertains to natureas far as its cognition can be applied in experience, or transcendent, thatis, it pertains to that connection of objects of experience which surpassesall experience. This transcendental physiology576 has either an inner

connection to its object or an outer one: the former is the transcendental

cognition of the world; the latter is the transcendental cognition of God.Immanent physiology considers the concept of all objects of the sensesbut only according to a priori conditions under which they can be givento us as such.577 It has two sorts of objects, 1) those of outer sense orcorporeal nature and 2) those of inner sense or [215] the soul, and thushas two sciences: rational physics and psychology. Accordingly, thewhole system of metaphysics consists of four main parts: 1) ontology, 2)rational physiology, 3) rational cosmology, and 4) rational theology; andrational physiology is divided once more into rational physics andpsychology.

Comment. 1. It is certainly true that a system or a science is acollection of cognitions determined by an idea, according to its form aswell as according to its content; nevertheless the properties of this ideashould be indicated more precisely. For not every collection of cognitionsordered according to an idea presents a science. After all, Kant evenconcludes the architectonic with the following beautiful and memorablewords:

“that, as mere speculation, it <(metaphysics)> serves more to prevent errors than

to amplify cognition does no damage to its value but gives it all the more dignity

and authority through its office as censor, which secures the general order and

concord, indeed the well-being of the scientific community, and prevents its bold

and fruitful efforts from straying from the main purpose, that of generalhappiness.”578

Here Kant clearly admits that general happiness must be the mainpurpose of the sciences. [216] Should this not result in a more precisedetermination of the idea according to which the whole of cognitions

573

Kant uses the hypothetical form “objects that would be given” (KrV, B873).574

Inbegriff (ibid.).575Naturbetrachtung (KrV, B874). I follow Kemp Smith instead of Guyer, who translates

“consideration of nature”, or Pluhar, “contemplation of nature”. Cf. also NAK, p. 418.576

Kant refers to transcendent physiology (since it surpasses all experience) (KrV, B874).577

Immanent physiology “considers nature as the sum total of all objects of the senses”,

i.e., as it is given to us (KrV, B874).578

Cf. KrV, B879.

New Anti-Kant 241

must be formed in order to lay a legitimate claim to the name of ascience? This should be compared with Bolzano’s proposals in theTheory of Science (§ 395).579

2. Our readers already know that we do not accept the way in whichKant separates mathematics and philosophy. Nonetheless we separatethese two sciences in a way which seems to us entirely in accord withpresent usage: we define mathematics as a theory of magnitudes andcount the theory of time and space, statics, mechanics, hydraulics, optics,and other similar sciences as mathematical sciences only because allthese sciences more or less deal with calculations (i.e., determinations ofmagnitude).

3. It is a bit of an exaggeration when Kant asserts on this occasionthat only mathematics can be taught but not philosophy, and that inregard to the latter, we must be content with teaching to “philosophize”(p. 623).580 This also holds for his claim that the use of reason inmathematics excludes all deception and error because it occurs in pureintuition (ibid.).581 Thank God mathematicians have never been so vain asto declare themselves infallible; [217] besides, they could easily berefuted by the history of their science. To express exactly what we feel,we should say that in the more than Babylonian confusion reigning to thisday in the domain of philosophy, no one can confidently extol the systemhe considers to be most right by saying that therefore everyone must becontent that he teaches, if not philosophy, then at least, to philosophize.

579

Cf. in WLIV, § 395, Bolzano presents the supreme principle of the Theory of Science:

“in dividing the whole domain of truth into individual sciences and by presenting these

truths in their proper manuals, we must act according to the demands of moral laws, i.e.,

in such a way that we produce the greatest possible good (or bring about the greatest

possible promotion of general well-being).” Cf. also “philosophy is the collection of all

those truths whose ultimate grounds we strive to penetrate inquiringly, so that we may

thereby become wiser and better.” Bolzano, “Was ist Philosophie”, op. cit., p. 26, and my

note 568, above. On p. 30 of the same manuscript Bolzano slightly modifies his definition:

“philosophy is the science of the objective connection between all those truths into whose

ultimate grounds we endeavour to penetrate as far as possible, so that we may become

wiser and better.” Kant ends “The architectonic of pure reason” on a similar note, namely,

that general happiness is the main purpose of metaphysics (or rational cognition from

mere concepts), which “considers reason according to its elements and highest maxims,

which must ground even the possibility of some sciences and the use of them all.” (KrV,

B879)580

Cf. KrV, B866, except that Kant says that of all rational sciences, only mathematics can

be learned and that we can at best only learn to philosophize, whereas in Přihonský’s view

of Kant, mathematics can only be taught (which implies that we might not even be able to

learn it).581

This (obviously problematic) statement refers to Kant’s view that mathematical

cognition is grounded in pure intuition (cf. my notes 128, 190, 201, 262, and 312).


For even if it is more difficult, we do not consider it impossible that someday people will agree on the doctrines of philosophy (especially practicalphilosophy) roughly in the same way they now agree on the doctrines ofmathematics. We do not altogether understand how Kant could haveclaimed that we cannot acquire a merely historical cognition (withoutinsight into their grounds) of mathematical doctrines, as we can ofphilosophical doctrines, since he certainly must have had a merelyhistorical cognition of many mathematical doctrines, for example, inastronomy.582

4. It is also disconcerting that in this enumeration of philosophicalsciences no explicit mention is made of logic, although we gather from p.624 that Kant has counted this science among the philosophicalsciences.583 [218] 5. We think it is a bit forced that rational theology ispresented as a branch of the rational study of nature (as the study ofgiven objects), or as the branch of the study of nature which considers theconnection of the whole of nature with a being beyond nature. How canthe consideration of a being that is beyond nature be defined as a branchof the study of nature?

6. Kant replies to his own objection, “how can I expect an a priori

cognition and thus a metaphysics of objects insofar as they are given toour senses and thus given a posteriori?”,584 as follows: “We take fromexperience nothing more than what is necessary to give ourselves anobject, partly of outer and partly of inner sense <(matter and thinkingbeings)>. But we abstain entirely from empirical principles that mightadd any sort of experience in order to judge something about theseobjects.”585 – We would prefer to say that although the concepts matter

and mind are induced in us by experience, they are not empiricalconcepts at all but pure ones, that is, concepts that do not contain a single

582

Of course Kant does not deny that we can acquire the historical cognition of

mathematical doctrines (i.e., the history of mathematics). Instead, he says that

mathematics is the only rational science which can be learned a priori, whereas

philosophy can only be learned historically. In addition, a mathematical cognition,

“however one has learned it”, can still count subjectively as a rational cognition [and not

only objectively] (KrV, B865).583

Cf. KrV, B867. Kant says that “[t]he mathematician, the naturalist, the logician are only

craftsmen of reason, however eminent the former may be in rational cognitions and

however much progress the latter may have made in philosophical cognition.” The logic

which he considers as having made progress in the philosophical tradition is

transcendental logic, and since this is not a logic accepted by Přihonský, he has little

cause for being disconcerted that Kant makes no explicit mention of it.584

Cf. KrV, B875.585

Cf. KrV, B876. The last sentence reads: “any sort of experience beyond the concept in

order to judge something about these objects.”

New Anti-Kant 243

intuition as a component. This is self-evident for anyone who isacquainted with our definition of the concept intuition, [219] that it is asimple representation which only has a single object. For what simplerepresentation with only a single object could be in the representationsmatter or mind? If Kant overlooked this and, besides these two concepts,also considered various other pure concepts as empirical ones, forexample, the concept of change as well as the concepts of pleasure andpain, and so on, this was only because the concept he combined with theword intuition was lacking the mark of simplicity.586

Kant concludes the Critique with a short “History of pure reason” (pp.633ff).587 Since he himself says that this title stands in his book only toindicate a place in the system that must be filled in the future, we have noreason to dwell on the little he presents here in four pages. He is contentto bring the most substantial changes in metaphysics under threeviewpoints:

1. With regard to the object of all our rational cognitions, he says,some were merely sensual philosophers, others merely intellectual

philosophers. The first asserted that reality and truth are only in theobjects of the senses, whereas the second asserted that they are only inthe intelligible objects which we cognize through a special (intellectual?)intuition not accompanied by any sense.[220] 2. With regard to the origin

of pure cognitions of reasons, the empiricists claim that they are derivedmerely from experience, whereas the noologists assert that their source isin reason.

3. With regard to method the naturalists established the principle that,in the most sublime questions of metaphysics, we can accomplish morethrough common understanding without science than throughspeculation. The observers of a scientific method, however, proceededeither dogmatically or sceptically. “The critical path alone is still open. Ifthe reader has had the courtesy and patience to accompany me on thispath, then he may now judge if it pleases him to add his part to making

586

Kant considers the mere concept of matter as impenetrable lifeless extension and the

concept of a thinking being as the empirical internal representation I think. (KrV, B876).

However, Přihonský wrongly accuses Kant of considering these concepts as empirical:

matter (the determinable in general) is a concept of reflection (these have no reference to

an object) and is contrasted with form; cf. KrV, B323. The I think is not an empirical

concept but the transcendental ground of thinking (cf. Kant’s discussion of transcendental

paralogisms and his distinction between rational and empirical psychology (B402-404 and

cf. also my notes 238, 245, and 336-344). On Bolzano’s and Kant’s notions of intuition,

cf. my notes 111-115 and 160-162 and Přihonský’s discussion in the introduction (NAK,

pp. 15-17) and in part 1 of the NAK, pp. 49-50.587

Cf. KrV, B880.


this footpath into a highway, whether or not that which many centuriescould not accomplish might not be accomplished before the end of thepresent one: namely, to bring human reason to full satisfaction in thatwhich has always, but until now vainly, occupied its thirst forknowledge.”588

Comment. 1. If intellectual intuition is taken to mean nothing otherthan that we have the capacity to also form concepts of objects we do notperceive through any sense and to make many right synthetic judgmentsabout them, then we should not want to deny that [221] there is such anintellectual intuition. But let no one pass off these judgments asimmediate cognitions unless he has first shown that they cannot beinferred judgments.

2. If the empiricists had investigated the nature of our empiricalcognitions more closely, it would soon have become clear to them thatmost of the judgments they considered as immediate experiences orperceptions were inferred judgments, and if they had only examined theway in which the inferring occurred, then they would have found that itincluded premises which were not perceptual judgments themselves butpure conceptual judgments.

3. We agree with the view that the merely common humanunderstanding gives no wrong answers to those questions which are byfar the most important ones in theoretical and practical philosophy, andwe are convinced that statements about which all humans on earth thinkin consensus at least have the highest degree of reliability, although theircontent is such that it injures human sensibility rather than flatters it. Butfrom this it does not follow by any means that we should not agree toinvestigate the premises on which such judgments rest and the way theyshould be demonstrated in systematic connection. Such scientific revisionwill always be of great use. [222]

4. Concerning the critical method, however, up until now it has notsucceeded in producing what its author did not consider as impossible:“to bring human reason to full satisfaction in that which has always, butuntil now vainly, occupied its thirst for knowledge”;589 indeed, we thinkthat, considering Kant’s injunction against reason not to venture into thenonsensory domain, it must be regarded as an attempt that entirely failedto accomplish its purpose.590 Nevertheless we do not hesitate to repeat at

588

Cf. KrV, B883.589

Cf. KrV, B883.590

This injunction concerns the understanding (which must not venture beyond the

boundaries imposed by experience). However, reason is not confined within determinable

boundaries, and its objects are nonsensory: the soul, the world (as a whole), and God; cf.

New Anti-Kant 245

the end of our investigations what we already expressed categorically atthe beginning. Far from denying Kant’s contributions to philosophy, wethink rather that there is not a single philosophical discipline he has notenriched with new and important doctrines owing to his astute comments.Despite the many mistakes made by their author, his critical workscontinue to be a treasure trove for anyone interested in philosophicalknowledge, especially the Critique of Pure Reason, which rightlyacquired its fame.[223]

KrV, B385. Besides, Kant’s critical method offers to investigate “the faculty of reason in

regard to all pure a priori cognition” (KrV, B869), and its ‘antifoundationalist’ (in the

sense that for him our knowledge does not rest on the foundation of indubitable perceptual

beliefs) aim can hardly be considered obsolete or unsuccessful. Do the judgments resting

on our sense experience not result from theoretical assumptions about that very sense

experience (which is a laboratory test of a certain hypothesis within the paradigm of a

certain theory)? Or, as Kant puts it: critical philosophy examines the foundations of our

knowledge and a critical objection “overturns a theory merely by showing that on behalf

of its assertion something has been assumed that is void and merely imagined, thereby

withdrawing from it the presumed foundation [. . .].” (KrV, A389)


5.4. New Anti-Kant

APPENDIX

In our opinion we do our readers a favour by summarising thedistinguished doctrines of the Critique of Pure Reason, dispersed in thesepages according to the rambling order of the Kantian work, in aperspicuous presentation accompanied by some concluding remarks.

The main tendency of the Critique is obviously to assign certainboundaries to our cognition and to restrict it to mere objects ofexperience, and the Critique sets about attaining this aim in the followingway: first of all it investigates the question of under which conditions our

cognitions, that is, our judgments, are valid or, as Kant puts it, thequestion of the origin of our judgments as such. On this question it makesan undoubtedly curious discovery, namely, that our cognition does notconsist of merely analytic judgments but also of synthetic judgments andthat the latter, especially, are of considerable importance for science,[224] since they alone can truly serve to enrich and enlarge it. Hence weunderstand how it is possible that the Critique deals with almost nothingother than the origin of synthetic cognitions. Besides, Kant’s acumendoes not miss the important distinction between synthetic judgments wemust observe here and which grounds the division between a priori and aposteriori judgments or between conceptual and experiential judgments.We prefer the second designation, because it indicates the nature of thesetwo kinds of judgments more precisely and eliminates certain disturbingsecondary representations prompted by the expressions a priori and a

posteriori, which easily generate the opinion that there are judgmentswhich exist in the soul previous to each and every experience or evenperception.591

591

Both Kant and Bolzano distinguish between experience and perception but not in the

same way. Bolzano distinguishes between experiential judgments, or mediate judgments

of intuition, and perceptual judgments, or immediate judgments of intuition (cf. WLIII, §

300.12, and my note 116). Kant holds that experiential judgments, or a posteriorijudgments, are a synthetic connection of perceptions: “the possibility of synthetic aposteriori propositions, i.e., of such as are drawn from experience, [. . .] requires no

special explanation; for experience itself is nothing other than a continual conjoining

(synthesis) of perceptions.” (P, § 5) He distinguishes between perceptual judgments and

experiential judgments by claiming that the first have only subjective validity and assert a

spatiotemporal relation between perceptions within my mental state without reference to

the object, whereas the second have objective and universal validity and assert a

spatiotemporal relation between perceptions in a general consciousness by subsuming

them under a pure concept of the understanding (which unifies them). He gives the

following example: “The concept of cause being such a concept, it therefore determines

New Anti-Kant 247

With regard to synthetic experiential judgments,*592 our philosopherthinks that their validity obviously depends on an intuition. A predicate isascribed to a subject because the judging person has found it in anintuition to which the subject refers.593 However, we believe we haveproved that [225] the origin of those judgments most likely to allow forsuch a definition is somewhat different than Kant pretends. But Kant wasso convinced that his definition was right that he tried to extend it to all

synthetic judgments, even a priori judgments. Thus he established as thesupreme principle of all judgments that each object of our cognition isconditioned by an intuition which we connect with it.594 In everyconceivable way he attempted to eliminate the difficulties he encounteredin the application of this principle. The judgments of pure mathematicshad already caused him some trouble.595 He had to admit that theirvalidity is self-evident for everyone, but no one who makes them canrefer to such intuitions as those Kant stipulated for experientialjudgments. Thus Kant should have already renounced his theory ofintuitions in his consideration of mathematical judgments, but theassumption of certain pure intuitions offered him a solution, albeit a veryuntenable one. For although the usual manner of presenting geometrymore or less appeared to justify his view, it would not fit in any way witharithmetical propositions. Kant, however, had become completely caughtup in the thought that all synthetic judgments were grounded on anintuition. Hence he pursued it most persistently, and although his [226]

proofs, faulty as they are, really only refer to experiential judgments and,amongst a priori judgments, only concern mathematical ones, he stillwould not be put off and concluded, as though he had demonstrated thematter in general: “all synthetic judgments and all a priori judgments, ifthey are to be right, must be grounded on intuition, and where this is notthe case, that is, where there are no intuitions, as in judgments whose

the intuition which is subsumed under it, e.g., that of air, with respect to judging in

general, so that the concept of air serves, with regard to expansion, in the relation of the

antecedent to the consequent in a hypothetical judgment.” (P, § 20) Cf. also my note 127

on Kant’s notion of experience.

*592

Strictly speaking, all experiential judgments are synthetic.593

Přihonský’s interpretation of Kant is not quite correct; cf. my note 591, above, my notes

234-245 on the apperceptive I think and the unity of judgment, and Přihonský’s

commentary, pp. 82-84 (NAK, pt. 1).594

Přihonský’s objection concerns the transcendental apperception; cf. my preceding note

497.595

Cf. Přihonský’s commentary, pp. 69ff., on the “προτον φευδοζ” of the Critique of PureReason and his objections to Kant’s assumption that mathematical judgments are

grounded on pure intuition, and my notes 201ff.


subject represents a nonsensory object, our judgments are unconfirmedper se, and no one is entitled to make them. So man’s most importantobjects, God, freedom, and immortality, are outside the sphere of hiscognition, and if he ventures beyond this sphere and makes syntheticassertions about them, he soon becomes entangled in paralogisms andfalls into antinomies or propositions which are proved as well as theirdirect opposite. In the constitutive usage of theoretical reason, mannecessarily falls prey to such a state of mental confusion, and he can besaved only by practical reason which secures the belief in God, freedom,and immortality through its postulates!”

Who would believe that so many scholars repeated these assertionsand that Kant was praised as the philosopher who first [227] succeeded inindicating the boundaries of our experience with certainty? He talks ofhis efforts in view of this goal on several occasions (e.g., p. 518, p. 572,etc.).596 – But who could measure the limits of the mind?597 Who coulddetermine that man should venture thus far and no farther? Hegel andSchelling were opposed to this, although they did not clearly expound thereasons for their opposition. Precisely because of this other philosophers,and amongst them even Herbart and Beneke, have maintained thisKantian doctrine, albeit with certain modifications. Dr. Theodor Waitz, amost honourable contemporary thinker, presents matters in a somewhatdifferent way and wants to credit our philosopher with the eternal meritof having “first grasped the thought that an insight into the formation598

of our concepts is the only means and at the same time the only possibleguarantee of the validity and objectivity of cognition itself.”*599 It isdebatable whether this thought is really Kantian, as Waitz assumes, and

596

Cf. KrV, B708, B790-791. But the passage in quotation marks is Přihonský’s summary

rather than a quote from the Critique.597

In the Critique of Pure Reason, Kant attempts to determine the boundaries of human

reason, not of the human mind. At KrV, B708, he even says that “the systematic

connection that reason can give to the empirical use of the understanding furthers not only

its extension but also guarantees its correctness”. The empirical use of reason is

strengthened by this systematic unity and opens up “new paths into the infinite (the

undetermined) with which the understanding is not acquainted, [. . .].” He restricts the use

of the understanding to the ground of possible experience and limits the use of reason to

the concepts of the understanding which it unifies under principles (cf. KrV, B382). In

short, Kant attempts to systematize the workings of our mental faculties, and his threefold

distinction between sensibility, understanding, and reason was criticized for

distinguishing between the understanding and sensibility or for distinguishing between the

understanding and reason or for failing to distinguish between acts of thinking and acts of

judging, etc.598

Bildungsgang.*

503 Cf. his Grounding of Psychology, pp. 114-115.

New Anti-Kant 249

“grounds his endeavor to criticize reason”, but it contains at least onemistake. For even if cognition depends on the concepts we have ofobjects, what Waitz infers from this does not follow, [228] namely, thatan insight into the formation of our concepts is the only possibleguarantee of the validity of our cognition. An insight into our conceptformation can certainly not provide a guarantee for the validity of ourjudgments, that is, the truth of our judgments, for in this way we cannever acquire certainty. For this formation can be presented and becognized as rightly presented only through a long series of judgmentsrequiring a new guarantee. But often the cognition of a truth, that is,some judgment we make about some object of our thoughts, is asufficient guarantee of its validity. I take this or that to be true, forexample, that just now I feel a pain or hear a sound, without demanding aproof or guarantee that I am not mistaken. This case occurs in allimmediate judgments, unlike mediate (experiential and conceptual)judgments, which rest on a shorter or longer series of inferences.Nevertheless, various means are at my disposal for becoming convincedof the rightness of these inferences. The most important means areindicated by Bolzano in the Theory of Science, § 300. The Kantian“analysis of the cognitive faculties”, however, is least capable ofachieving this, in whatever way it is carried out. In our opinion,sensibility, understanding, reason, [229] imagination, and whatever elsethe many powers are called which are mentioned in the Critique can bejustifiably put aside. It only causes misunderstandings to speak of powersand faculties when we should talk of representations, propositions, orjudgments themselves.600 The question Kant investigates is this: do ourjudgments correspond to truth? Also, are judgments with a nonsensoryobject true and can their truth be proved? and so on—the answers to allthese questions do not require recourse to the above-mentioned powers.Instead, considering these powers is only meddling in unfamiliar

affairs,601 which is tempting insofar as we know nothing about thesepowers except that they are causes that bring about effects, and byhypostatizing them we arbitrarily separate or oppose them, and so on—inferring something more from this name than what we could have

600

Přihonský repeats Bolzano’s objection that Kant “psychologized” logical laws and

linguistic expressions by failing to distinguish between mental (subjective) acts and ideal

(objective) relations; cf. WLI, §§ 16.2-3, 19, 34.3, 36, 50.2, 65.11.a; WLIII, § 291.1-2. Cf.

also “Ueber den Begriff des Schönen”, p. 99, where Bolzano says that the προτον ψευδοζof modern German philosophy is that the notion of a concept is mixed up with the mental

act (the thought) as well as the object of this act; cf. also my note 323 on Kant’s and

Bolzano’s different notions of objective and subjective representations.601

Přihonský writes Allotrion.


inferred anyway. So, in any case, the Critique has pursued a wrongcourse in deciding the question concerning the validity of our cognition,and to get things straight we must proceed in an entirely different way.We should not begin by investigating and analyzing the human cognitivefaculty or the cognitive faculty of thinking beings as such, for it isnecessary to discuss the nature of truths as such or, to speak moregenerally, of propositions as such, their kinds and component parts, their[230] interrelation, and to discover similar relations between them. Thereis no use in dealing with cognition and its conditions unless we havedone this. Besides we should not forget that the object of cognition is,first of all, only truths and not really existing objects.602 For it is onlyfrom certain truths asserting something about our representations that we

602

Contrary to Přihonský’s view, Kant would agree with the second part of Přihonský’s

claim (since he completely rejects the realist claim that the objects of our cognition are

actually existing objects) but not with the first part, since for him there are no truths as

such (i.e., absolute truths), and even if there were, they would be noumena which we can

think but not cognize. Bolzano and Kant have very different views on truth, and the only

common denominator between them is that they are variants of the correspondence theory

of truth (cf. KrV, A58). For Bolzano, “true” is a logical and grammatical predicate used to

describe entities as having the property of truth. Statements, judgments, beliefs, claims,

assumptions, propositions, sentences, and utterances are considered as truth-bearers, that

is, the things we ordinarily consider as either true or false. In other words, what is stated,

believed, claimed, asserted, etc., can have the property of truth (cf. WLI, § 19ff.). N.B.,

Bolzano has five uses of the word “truth” (cf. WLI, §§ 25, 26, 34). (1) Truth is a property

of a proposition which states something as it is and does not depend on human thought.

(2) A proposition with this property is called a truth as such or an objective truth. (3)

Truth is the property of a judgment which grasps a true proposition as such. True or

correct judgments are cognitions (Erkenntnisse) or cognized truths (and all truths are

cognizable). (4) Truth is the collection of true propositions or true judgments. (5) True is

a synonym for actual. Bolzano also attempts to give proofs that there are true

propositions, one of which is a proof by mathematical induction. He assumes that there is

at least one truth. If there are at least n truths, then there are at least n + 1 truths, and so

there are an infinite number of truths (Cf. WLI, §§ 26, 32; PU, § 13).

For Kant, on the other hand, truth consists in adequatio between a mental act and its

object, and he considers the problem of truth as an epistemological problem. What makes

a proposition true for Kant is our knowledge of its object. (This is the material truth

condition given in transcendental logic, which deals with the sources of our knowledge of

objects. Truth is the agreement of knowledge with its object in the sense that this

agreement is determined by the conditions of experience which are also the conditions of

the objects of experience; cf. KrV, B197, B375. These conditions include both the

appearances given in pure intuition and the categories of the understanding, which “lead

to truth, that is, to the conformity of our concepts with the object” [B671].) In short,

Kant’s truth condition with regard to the cognitive content of a judgment is this: the

judgment “snow is white” is a true cognition if and only if it corresponds to an object as it

appears to me, that is, as it is given in intuition and subsumed under a concept by means

of a schema or rule of the understanding; cf. KrV, B, § 16. Cf. Von Duhn (2001), op. cit.

New Anti-Kant 251

can infer whether one of our representations refers to a real object andwhat other properties this object has; for example, we infer the existenceof sensory objects from perceptual judgments, and so on. But truthsdealing with sensory things are not the only truths we can recognize. Bymeans of concepts we gradually attain an immeasurable domain of truthsopening up to our cognition, which we continue to penetrate without everbeing able to cognize it exhaustively. – The cognition of objects (morerightly: the cognizability of truths)603 has always caused much trouble forphilosophers. The more recent ones often defined this business “as adarkness that covers the deepest secret of philosophy, the enigma of theconnection between the ideal world and the real one, or between theworld as it is, independently of all cognition, and as it is reflected in thebrain as the cognitive organ.”604 [231] To disperse this darknesssomewhat and, as they put it, to go beyond every dualism (including theKantian one between things-in-themselves and appearances), theyresorted to the representation of the subject-object identity. But neitherthe doctrine of identity nor the subsequent dialectical method nor certainother attempts cast the desired light. – So is it true that darkness isessential, as it were, and that it is impossible to go beyond it? ‘Not atall’, we reply, ‘you could easily overcome it if you could only makeyourselves accept the assumption we have repeatedly recommended inthese pages, namely, that “there are infinitely many truths as such and weare capable of grasping a significant amount of them, as well as dividingand ordering the cognized truths in particular sciences”.’605 A furtherdiscussion of this topic would divert us too far from our goal. Hence werefer the reader to the introduction and Bolzano’s Theory of Science, §19ff. We conclude our treatise with a few words on Kant’s distinctionbetween things-in-themselves and their appearances.

There is no clear definition of a thing-in-itself to be found in the Critique.Unlike appearances, such things are assumed [232] in support of certaintheorems in the same way as the pure intuitions of time and space wereassumed in order to justify the doctrine of the necessity of intuition forgrounding synthetic conceptual truths.606 Kant hardly seems to suspect the

603

Cf. WLI, § 26.4, where Bolzano claims that although there are unknown truths, all

truths are cognizable. “The cognizability of an object is the possibility of pronouncing a

true judgmenton it.”604

Here Přihonský presumably parodies or paraphrases a nineteenth century German

idealist.605

Cf. WLI, §§ 1, 2, 16, on the tasks of a Theory of Science.606

Although it is true that Kant could have defined this notion more precisely, he does

provide some explanation: the name thing-in-itself refers to the noumenon as well as to


contradiction in which the Critique is caught by assuming an object that is inany case nonsensory and asserting straight away that we do not cognize

things-in-themselves but only their appearances. In which sense is thissupposed to be true? We could accept this proposition only if we wereallowed to interpret it such that we do not immediately cognize real thingsbut infer their reality and other properties from certain representations(mental appearances).607 But Kant’s expression hardly refers to this sense

transcendent ideas. It is chiefly a boundary concept which limits the use of our

understanding and reason to objects of a possible experience: a thing-in-itself can be

thought (provided that it is not self-contradictory) but not cognized (since it cannot be the

object of a possible experience); cf. KrV, Bxx, A249-250; P, § 29; and Přihonský’s

commentary on the noumena, NAK, pp. 100-101, and my notes 293-294; as well as

Přihonský’s discussion of the transcendent ideas of God, the world, and the soul, pp.

110ff, and my notes 319, 323, 325-326. Přihonský quite correctly points out that the thing-in-itself is an assumption used as the support of a theorem, for Kant assumes that there

must be a thing-in-itself or a correlate of sensibility which is thinkable but not cognizable

(and that nothing intuited in space is a thing-in-itself) and that thus we can only know

things as they appear but not as they are in themselves; cf. KrV, B45, A249-250.607

Cf. my notes 113-114 on Bolzano’s weak causal claim concerning our knowledge of

objects and Přihonský’s surprisingly strong causal claim that intuitions alone provide

reliable information about the world and that we can therefore justifiably infer an object’s

existence from the existence of its representation in our mind, since an intuition has only

one object (and thus causes only one mental representation), NAK (pt. 1), p. 49, and my

note 160. This view is indefensible, if only because it does not account for perceptual

illusions, hallucinations, errors, etc. Besides, it is easily refutable by an anti-realist (whom

Přihonský attacks above, on p. 170), who would argue that we have no reason to think that

our intuitive representations actually correspond to anything in the world, since they rest

on our perceptual faculty. Nor could Přihonský’s claim withstand being put through the

Sceptic’s modes of suspending our judgment. Sextus Empiricus would argue against such

an aetiologist view that we have no reason for inferring the existence of a thing from the

existence of a mental representation (or an appearance in my mind). How a thing appears

to us (or what intuition we have of it) depends on the situation in which it appears, the

subject to whom it appears, or the background against which it appears: for example,

doves’ necks appear different in colour depending on the different ways they turn them

(Pyrroneioi hypotyposeis, I, 120); and hearing similarly grasps the same sound at night as

loud, in the daytime as faint” (Pyrroneioi hypotyposeis, II, 55-56), Sextus Empiricus,

Outlines of Scepticism, trans. J. Annas and J. Barnes (1994); and J. Annas and J. Barnes,

The Modes of Scepticism (1985), pp. 23-25.

Thus it would be more prudent to postulate a causal power ex hypothesi, as a possible

explanation of our having certain sensations (or intuitions), as Bolzano does. For example,

we can only assume that colour properties have the power to produce colour sensations in

us as a possible explanation of our experience of colours. In other words, we formulate a

hypothesis, or give a reason and explain an observation, by deriving an observational

statement from this reason: the reason we have a sensation of red is that there is a red

surface which can cause this sensation. Cf. Kasabova (2004), op. cit. Would Přihonský

also use his previous objection against his master? (namely, the objection to Kant’s

New Anti-Kant 253

since he also absolutely denies us any mediate cognition of things-in-themselves. Nevertheless, we read in the first edition of the Critique that“appearances are nothing but representations” (cf. vol. 2 of the Complete

Works, p. 246).608 However, these words should not have a sceptical meaningbut a truly critical one. So by expressing himself in this way, Kant had topresuppose that certain objects outside ourselves correspond to theseappearances or representations inside ourselves (or at least many of them)and that these real objects, since they may not be [233] things-in-themselves,are properties or qualities of those things, for after all, besides them there isnothing truly real.609 Thus Kant would have granted us the cognition ofproperties or qualities of things by means of representations.610 But it is notclear how he could then deny us the cognition of things-in-themselves. Forthings (and thus also things-in-themselves, if they are things) cannot becognized in any other way than through their properties, neither by humansnor by any other cognizing being. – In the following editions, Kant takes theword “appearances” in an entirely different sense and defines it assynonymous with the words “objects of a possible experience” (cf. vol. 2 ofthe Complete Works, p. 234),611 “empirical objects” (p. 239),612 and beings ofthe senses (p. 243),613 according to which the statement “we do not cognize

procedure of using assumptions (such as the thing-in-itself) as the support of a theorem;

cf. p. 232).608

Cf. KrV, A250.609

Unlike Bolzano, Kant does not assume that there are properties of actual objects which

have the power to cause certain representations in us. Instead, Kant tackles the problem

the other way round and starts not by assuming the causal power of objects but with our

cognitive powers. Thus he would say that we can explain cognitive tasks, such as colour

discrimination, by presupposing cognitive structures, and here Bolzano agrees with him

(i.e., both authors claim that colour discrimination is possible because we are capable of

conceiving colours [Bolzano] or because we can associate our sensations with pleasurable

responses [Kant]). Cf. Kasabova (2004), op. cit., and my notes 160 and 480, above.610

There is one sense in which Přihonský’s statement is correct, namely, that Kant would

accept the indirect reference view held by Bolzano and Přihonský, i.e., that objects are not

immediately given and that our only access to them is through representations; cf. WLI, §

48.3; cf. also his discussion of adherences in the Athanasia, sec. 1, p. 21, where he comes

close to Locke’s claim that colours are secondary qualities which have the power to

produce colour sensations in us; cf. An Essay Concerning Human Understanding, bk.

II.viii., §§ 24-25.611

Cf. KrV, B282-283; cf. also KrV, B298.612

Cf. KrV, B299. Appearances are objects of experience and thus they are empirical

objects. We should note that Kant uses the word Gegenstand to refer to the objects of

experience or appearances and the word Objekt to refer to the manifold of intuitions

united in a concept and related to the I think (then the object of experience becomes an

object for our cognition); cf. KrV, B136-137, B197; cf. also my note 304.613

Cf. KrV, B306. Here Kant opposes appearances to noumena: “if we call certain objects

[Gegenstände], appearances, or beings of sense (phaenomena), because we distinguish the


things-in-themselves but only their appearances” is merely anotherexpression of the well-known and unpleasant result of the Critique, “that wecan only make right judgments about objects of the senses”.614 If this weretrue, it would at once put an end to all metaphysics. But in our view we havesufficiently demonstrated the falsity of the principles from which Kantinferred this assertion.

way we intuit them from their constitution in itself, then it already follows from our

concept that to these we oppose, as it were, [. . .] other possible things, which are not

objects of our senses at all, and call these beings of understanding (noumena).”614

Cf. my note 243, above; cf. KrV, B126, Přihonský’s commentary, p. 85 (pt. 1), and my

notes 255-256; cf. also KrV, B176 (Kant’s doctrine of the schematism which Přihonský

rejects), cf. p. 87 (pt. 1), and KrV, B190 (Kant’s claim that synthetic a priori judgments

are objectively valid if and only if they are dealing with the object of a possible

experience). Cf. also Dubislav (1931, p. 226), who notes that Přihonský’s summary of

Kant’s most important doctrines, as they are developed in the Critique, is unsatisfactory

and refers us to his own pertinent summary: “Ueber Bolzano als Kritiker Kants” (1929),

pp. 357-368).

REFERENCES

Annas J. and Barnes J. (1985). The Modes of Scepticism. Cambridge: Cambridge Univ.

Press.

Aristotle (1975). Posterior Analytics. Oxford: Clarendon Press, ed. by J. Barnes.

Aristotle (1984): The Complete Works of Aristotle (2 vols set), edited by J. Barnes,

Princeton, Princeton University Press, Bollingen Series.

Arnauld A. and Nicole P. (1662). La logique ou l'art de penser, (1683), P. Clair and G.

Girbal, eds. Vrin, Paris, 1981.

Barnes J. (1972). The Ontological Argument. London: Macmillan.

Bauer R. (1966). Der Idealismus und seine Gegner in Oesterreich, Beihefte zumEuphorion, Zeitschrift für Literaturgeschichte, 3. Heft, Carl Winter

Universitätsverlag, Heidelberg.

Bealer G., “A Theory of the A Priori”, vol. 13, Philosophical Perspectives (Oxford:

Blackwell, 1999), pp. 29-55.

Beaney M. (1977). ed. The Frege Reader. Oxford: Blackwell.

Beneke Fr. C. (1837). review of Hartenstein’s: The Problems and Fundamental Doctrinesof General Metaphysics in the Haller Literary Journal.

Beneke Fr. C. (1842). System of Logic. Berlin.

Berg J. (1962). Bolzano’s Logic (Stockholm: Almqvist)

Berg J. (1997). “Bolzano the Prescient Encyclopedist”, “Bolzano and Analytic

Philosophy,” Grazer philosophische Studien, 53: pp. 13-33, Rodopi.

Berkeley G. (1709). A New Theory of Vision. London: Everyman, 1963

Beth E. W. (1953). “Kant’s Einteilung der Urteile in analytische und synthetische”,

Algemeen Nederlands tijdschrift voor wijsbegeerte en psychologie, The Hague, pp.

253-264.

Bolzano B. (1804). Betrachtungen über einige Gegenstände der Elementargeometrie,

Prague: Karl Barth.

Bolzano B. (1810). Beyträge zu einer begründeteren Darstellung der Mathematik, Prague,

Caspar Weidmann.

Bolzano B. (1810). Beyträge zu einer begründeteren Darstellung der Mathematik in

Bolzano’s Early Mathematical Works, Czechoslovak Studies in the History of

Science, Prague, 1981.

Bolzano B. (1827). Athanasia oder Gründe für die Unsterblichkeit der Seele. Ein Buch fürjeden Gebildeten, der hierüber zur Beruhigung gelangen will. Sulzbach, 2nd edition,

1838, reprint 1970.

Bolzano B. (~1830). Anti-Euklid (manuscript), Kazimir Vecerka ed., Sbornik 1967,

Prague, pp. 204-215.

Bolzano B. (1836) Lebensbeschreibung des Dr. B. Bolzano mit einigen seiner

ungedruckten Aufsätze und dem Bildnisse des Verfassers eingeleitet und erläutert von

dem Herausgeber, Sulzbach: J. E. v. Seidel.

Bolzano B. (1837). Wissenschaftslehre, Versuch einer ausführlichen und grösstenteilsneuen Darstellung der Logik mit steter Rücksicht auf deren bisherige Bearbeiter.Hrsg. von mehreren seiner Freunde mit einer Vorrede des Dr. J. Ch. A. Heinroth. 4Bd. Sulzbach, reedited in 4 vols. by W. Schultz, 2. Neudruck der 2. Auflage, Leipzig,

1929; Lizenzausgabe Scientia Verlag, Aalen, 1981 (BBGA I, vols. 11-14).

REFERENCES256

Bolzano B. (1843). “Ueber den Begriff des Schönen”, in Mathematisch-physikalische undphilosophische Schriften 1842-1843, BBGA, vol. 18.

Bolzano B. (1843). “Aufsatz, worin eine von Herrn Exner in seiner Abhandlung ‘UeberNominalismus und Realismus’ angeregte logische Frage beantwortet wird”. BBGA I,

vol. 18, pp. 63-85.

Bolzano B. (1845), Kurzgefasstes Lehrbuch der katholisch-christlichen Religion, als derwahren göttlichen Offenbarung. Bautzen, Weller.

Bolzano B. (1851). Paradoxien des Unendlichen. 1st edition F. Prihonsky ed., Leipzig;

reedited, Meiner, Hamburg, 1921, new ed. B. van Rootselaar, 1975.

Bolzano B. (1851). Drei philosophische Abhandlungen, welche auch von Nicht-

philosophen sehr wohl verstanden werden können, und vier akademische Reden von

allgemein menschlichem Interesse. Aus Dr. Bernard Bolzano’s schriftlichemNachlasse. Leipzig, Reclam. (BBGA, vol. 2A12/3), pp. 43-104.

Bolzano B. (1935). Briefwechsel Bolzano-Exner, ed. E. Winter, B. Bolzano's Schriften,

vol. 4, published by the Königl.-Böhmische Gesellschaft der Wissenschaften, Prague.

Bolzano, B. (1949) Bolzano’s Wissenschaftslehre in einer Selbstanzeige, in E. Winter:

Leben und geistige Entwicklung des Sozialethikers und Mathematikers BernardBolzano (1781-1848), Hallische Monographien, Halle.

Bolzano B. (1950). Paradoxes of the Infinite, trans. D. A. Steele (London: Routledge)

Bolzano B. (1969). Was ist Philosophie (posthumous manuscript), Rodopi.

Bolzano B. (1969-). Bernard Bolzano Gesamtausgabe, hrsg. E. Winter, J. Berg, F.

Kambartel, J. Louzil, B. van Rootselaar; F. Frommann-Holzboog Verlag, Stuttgart

Bad-Cannstatt, comprising an Einleitung and 4 series, I. Schriften, II. Nachlass, III.

Briefwechsel, IV. Dokumente, (BBGA).

Bolzano B. (1972). Theory of Science, partial translation: R. George (London: Blackwell);

Bolzano B. (1973). Theory of Science, partial translation: B. Terrell (Dordrecht: Reidel);

Bolzano B. (1996). Appendix on the Kantian Theory of the Construction of Conceptsthrough Intuitions; and other (partial) translations of Bolzano’s early mathematical

works by S. Russ in W. Ewald, From Kant to Hilbert, vol. 1 (Oxford: Clarendon

Press)

Brentano B. (1924-1925). Psychologie vom empirischen Standpunkt (Leipzig: Meiner.

vol. 1.

Brentano B. (1993) Auszug aus einem Brief an Oskar Kraus vom 31 Mai 1909, EduardWinter: Ausgewählte Schriften aus dem Nachlass, eingeleited und herausgegeben von

E. Morscher. Sankt Augustin: Akademia.

Cantor G. (1883). ”Ueber unendliche, lineare Punktmannichfaltigkeiten”, 5.

Mathematische Annalen 21. Leipzig.

Cantor G. (1883). Grundlagen einer allgemeinen Mannichfaltigkeitslehre. Einmathematisch-philosophischer Versuch in der Lehre des Unendlichen. Leipzig,

Teubner.

Cantor G. (1932). Gesammelte Abhandlungen mathematischen und philosophischenInhalts. Berlin.

Cousin V. (1834). on French and German philosophy, the German translation by Dr.

Hubert Beckers, professor of philosophy at the Lyceum at Dillingen, and the critical

preface by Privy Councillor v. Schelling, Stuttgart and Tübingen near Cotta.

Couturat L. (1905), La philosophie des mathématiques de Kant. appendix to Les principesdes mathématiques. Olms, 1979.

Dedekind F. (1888). Was ist und was sollen die Zahlen?. Braunschweig, Vieweg.

Drobisch (1862), “Locke als Vorläufer Kants”, in Zeitschrift für exacte Philosophie, pp.

1-32.

REFERENCES 257

Dubislav W. (1929). “Ueber Bolzano als Kritiker Kants”. Philosophisches Jahrbuch derGörres-Gesellschaft, 42, pp. 357-368.

Dubislav W. (1930). “Zur Wissenschaftstheorie der Geometrie”. Blätter für DeutschePhilosophie 4, pp. 368-381.

Dubislav W. (1931). Die Definition, Meiner, Leipzig.

Etchemendy J. (1990). The Concept of Logical Consequence. Cambridge: Harvard Univ.

Press.

Fesl M. (1988). Einleitung zu Bolzano’s: Ueber die Perfectibilität des Katholicismus, in

Ganthaler and Morscher: “Ueber Bolzano, Stoppani und die “Perfektibilität” des

Katholizismus”, Forschungsberichte und Mitteilungen des ForschungsintitutsPhilosophie/Technik/ Wirtschaft and der Universität Salzburg 15. Salzburg, pp. 17-75.

Francke F. (1838). Zur Theorie and Kritik der Urtheilskraft, Leipzig: Julius Klinckhardt.

Frege G. (1884) Die Grundlagen der Arithmetik Breslau: W. Koebner.

Frege G. (1884). The Foundations of Arithmetic, trans. J. L. Austen (Oxford: Basil

Blackwell, 1986). 1st edition, Blackwell, 1950.

Friedman M. (1992). Kant and the Exact Sciences, Cambridge MA, Harvard University

Press.

Fries J. (1820) Handbuch der Psychischen Anthropologie, 1-2. Jena: Cröker, 1821.

Hoppenstedt Fr. W. von (1933). Franz Brentanos Kantkritik, doctoral dissertation, Kiel.

Husserl E. (1901). Logical Investigations (Logische Untersuchungen), Tübingen: Max

Niemeyer, 1980; trans. J. N. Findlay (London: Routledge, 1970, reprinted 2002)

Husserl E. (1913). Introduction to the Logical Investigations, trans. Ph. Bossert and C. H.

Peters. The Hague: Martinus Nijhoff, 1975.

Inwood B. and Donini P. (1999). “Stoic Ethics”. The Cambridge History of HellenisticPhilosophy. Cambridge: Cambridge Univ. Press.

Kant I. (1783). Prolegomena. Riga, Hartknoch.

Kant I. (1838). Kant’s Complete Works, vol. 2. Leipzig. Modes and Baumann.

Kant I. (1838). Imm. Kant’s Works, complete edition in ten volumes, carefully revised.

With a preface by Gustav Hartenstein, Prof. of Philosophy at the Univ. of Leipzig;

Modes & Baumann. Leipzig.

Kant I. (1838). Imm. Kant’s Complete Works, edited by Karl Rosenkranz and Friedr. Wilh.

Schubert; Leop. Voss. Leipzig.

Kant I. (1902-1955). Kant's gesammelte Schriften, ed. Berlin Academy of Sciences.

Berlin, Georg Reimer, 23 vols.

Kant I. (1929). Critique of Pure Reason N. Kemp Smith (London: Macmillan, 2nd

impression 1933, 1990).

Kant I. (1956). Critique of Practical Reason, trans. L. White Beck. New York: Macmillan.

Kant I. (1966-1979). Kant's gesammelte Schriften, ed. Berlin Academy of Sciences.

Berlin, Georg Reimer, 9 vols.

Kant I. (1967). Критика на Чистия Разум, превод Цеко Торбов, София, Българска

Академия на Науките.

Kant I. (1980). Grounding for the Metaphysics of Morals, trans. J. Ellington. Indianapolis:

Hackett.

Kant I. (1992). The Jäsche Logic and Other Lectures on Logic, trans. J. M. Young.

Cambridge: Cambridge Univ. Press

Kant I. (1992). The Cambridge Edition of the Works of Immanuel Kant, TheoreticalPhilosophy 1755-1770. Cambridge: Cambridge Univ. Press.

Kant I. (1996). Critique of Pure Reason W. Pluhar. Indianapolis: Hackett.

Kant I. (1998). Critique of Pure Reason P. Guyer & A. Wood. Cambridge: Cambridge

Univ. Press.

REFERENCES258

Kant I. (2002). Prolegomena to any Future Metaphysics that will be able to come forwardas science, trans. G. Hatfield. Cambridge: Cambridge Univ. Press.

Kasabova A. (2002). “Is Logic a Theoretical or Practical Discipline? Kant and/or

Bolzano”. Archiv für Geschichte der Philosophie, de Gruyter, Berlin, NewYork, no. 3,

pp. 319-333.

Kasabova A. (2004). “Colour Sensations and Colour Qualities: Bolzano between Modern

and Contemporary Views”, British Journal for the History of Philosophy 12(2): pp

247-276.

Kasabova A. (forthcoming article), “Bolzano’s semiotic method of explication”, to appear

in the History of Philosophy Quarterly.

Kerry B. (1886). “Ueber Anschauung und ihre psychische Verarbeitung”, 2,

Vierteljahresschrift für wissenschaftliche Philosophie, 10. Jahrgang, Leipzig, 1886,

pp. 438-440.

Künne W. (1997). ““Die Ernte wird erscheinen . . .” Die Geschichte der Bolzano-

Rezeption 1849-1939“. Bolzano und die österreichische Geistesgeschichte, Academia,

St. Augustin (pp. 9-82).

Leibniz G. (1973). Philosophical Writings, trans. and ed. G. H. R. Parkinson. London:

Dent.

Leibniz G.W.F. (1981). New Essays on Human Understanding, trans. P. Remant and J.

Bennett. Cambridge: Cambridge Univ. Press.

Löffler W. (1999). “Bolzanos kosmologischer Gottesbeweis im historischen und

systematischen Vergleich”. Bernard Bolzanos geistiges Erbe für das 21 Jahrhundert.Academia, St. Augustin, pp. 295-316.

Losev A. (1923), “Философия имени” (first published in 1927), reprinted in Бытие, имя,космос, Mысль, Москва, 1993.

Loužil J. (2000), “Bernard Bolzano, Josef Jungmann und die Anfänge der tschechischenNationalbewegung” in Bernard Bolzano und die Politik, Böhlau, Wien.

Lukasiewicz J. (1913). Die logischen Grundlagen der Wahrscheinlichkeitsrechnung.Academy of Sciences, Cracow.

Mancosu P. (1999). “Bolzano and Cournot on Mathematical Explanation”, in

“Mathématique et logique chez Bolzano”, Revue d’histoire des sciences, vol. 52, no.

3/4: pp. 436-437.

Mann S. E. (1957). Czech Historical Grammar (London: Athlone Press)

Marbach G. O. (1838-1844) Lehrbuch der Geschichte der Philosophie, 3 vols. Leipzig:

Otto Wiegand.

Marty A. (1916). Raum und Zeit, published posthumously by J. Eisenmeier, A. Kastil, O.

Kraus, Max Niemeyer, Halle.

Marty A. (1928). “Von den logisch nicht begründeten synsematischen Zeichen”, in O.

Funke: Grundfragen zur Bedeutungslehre, Reisland, Leipzig.

Maund B. (1995) Colours, their Nature and Representation. Cambridge: Cambridge Univ.

Press.

Morscher E. (1974). “Ist Existenz ein Prädikat?”. Zeitschrift für philosophischeForschung, pp. 120-132

Morscher E. (1997). “Bolzano’s Method of Variation: Three Puzzles”, in “Bolzano and

Analytic Philosophy”, Grazer philosophische Studien, vol. 53, pp. 139-166.

Morscher E. (2003). ed. Franz Přihonský, Neuer Antikant, Academia, St. Augustin.

Palàgyi M. (1902). Kant und Bolzano, eine kritische Parallele. Halle, Niemeyer.

Paul H. (1935). Deutsches Wörterbuch, Niemeyer.

Peirce C. S. (1876). “Deduction, Induction, and Hypothesis”, in Selected PhilosophicalWritings, vol. 1 (Bloomington: Indiana University Press, 1992), pp. 186-199.

REFERENCES 259

Přihonský F. (1822). Einträge in Pierer’s Encyclopädisches Wörterbuch derWissenschaften, Künste und Gewerbe, in Verbindung mit mehreren Gelehrten.Herausgegeben von Dr. A. Binzer (Altenburg, Hahn, new edition 1824).

Přihonský F. (1828). Schreiben eines katholischen Geistlichen an den Verfasser der zweyBriefe, durch die jüngst zu Dresden erschienene Schrift: die reine katholische Lehre,veranlasst, Sulzbach: Seidel.

Přihonský F. (1835). “Besprechung von Bolzano’s “Athanasia” (1827)”. JenaischeAllgemeine Literatur-Zeitung 28, no. 91-93, pp. 337-360;

Přihonský F. (1835). “Besprechung von Bolzano’s “Religionswissenschaft””. FreimüthigeBlätter über Theologie und Kirchenthum, 5, Stuttgart, pp. 265-267;

Přihonský F. (1836). Besprechung von Johann Baptist v. Hirscher: Die christliche Moral,

als Lehre von der Verwirklichung des göttlichen Reiches in der Menschheit. Jenaischeallgemeine Literatur-Zeitung 34, vol. 4, no. 214-217, pp. 265-269;

Přihonský F. (1837). Besprechung von Bolzanos “Religjonsbekenntnisse zweier

Vernunftfreunde, nämlich eines protestantischen und eines katholischen Theologen”.

Freimüthige Blätter über Theologie und Kirchenthum 6, Stuttgart, pp. 353-370;

Přihonský F. (1841). “Ueber den Begriff eines katholischen Dogma”, in: FreimüthigeBlätter über Theologie und Kirchenthum 17, Stuttgart.

Přihonský F. (1845). Vorwort des Herausgebers. Bolzano’s Ueber die Perfectibilität desKatholicismus. Streitschriften zweier katholischen Theologen; zugleich ein Beitragzur Aufhellung einiger wichtigen Begriffe aus Bolzano’s Religionswissenschaft.Leipzig, Voss. (published anonymously).

Přihonský F. (1846). Besprechung von Fechners “Ueber das höchste Gut”. Neue Jenaischeallgemeine Literatur-Zeitung 6 (Leipzig, 1847), no. 228-229, pp. 912-915;

Přihonský F. (1847). Die Monadologie. Leipziger Revue, Zeitschrift für Literatur, Kunstund Leben, no. 24, pp. 94-96;

Přihonský F. (1847). Besprechung von Zimmermann “Leibnitz’ Monadologie. NeueJenaische allgemeine Literatur-Zeitung, 7, no. 61, pp. 241-243;

Přihonský F. (1847). Besprechung von Karl Ludolf Menzzer: Naturphilosophie vol. 1.Neue Jenaische allgemeine Literatur-Zeitung, 7, no. 288-289, pp. 1149-1155.

Přihonský F. (1849). “Bolzano”, Bohemia 22 (Prague, semester 1, no. 40-41) and

“Bernard Bolzano, professor of the theory of religion at the University of Prague, born

on 5 Oct. 1781, died on 18 Dec. 1848”. Neuer Nekrolog der Deutschen, 26 (Weimar,

1850), pp. 765-784.

Russell B. (1919). Introduction to Mathematical Philosophy. London: Allen & Unwin.

Russell B. (1937). The Principles of Mathematics, 2nd ed. London: Allen & Unwin.

Russell B. (1983). The Collected Papers 1: Cambridge Essays 1888-1899 (London, Allen

& Unwin.

Scholz H. (1924). “Das Vermächtnis der kantischen Lehre vom Raum und von der Zeit”,

Kant-Studien 29, pp. 21-69, and

Scholz H. (1931). Geschichte der Logik, Junker und Dunnhaupt: Berlin, unveränderte

Neuauflage unter dem Titel Abriß der Geschichte der Logik. Alber: Freiburg i. Br.

1959, 1967.

Schwarz Ph. (1911). Bolzano's Vorstellungstheorie und Kant's Lehre von derRaumanschauung, Inaugural-Dissertation zur Erlangung der Doktorwürde der HohenPhilosophischen Fakultät der Georg-August Universität zu Göttingen. Hofer,

Göttingen.

Sebestik J. (1984), “Bolzano et Brentano, deux sources autrichiennes du Cercle de

Vienne”, Fundamenta Scientiae, vol.5, nos. 3-4, pp.219-235.

REFERENCES260

Sebestik J. (1986) “Le cercle de Vienne et ses sources autrichiennes”, Le Cercle deVienne, doctrines et controverses, J. Sebestik and A. Soulez, eds., Klincksieck, Paris.

Sebestik J. (1986) “La classe universelle et l’auto-appartenance chez Bernard Bolzano”,

Eleutheria. Athènes, pp. 72-85.

Sebestik J. (1992) Logique et mathématique chez Bernard Bolzano. Paris, Vrin.

Sebestik J. (1996) “Etudes Bolzaniennes”, Revue de métaphysique et de morale, 3, 1996,

pp. 437-448.

Sextus Empiricus (1994), Outlines of Scepticism, trans. J. Annas and J. Barnes

(Cambridge: Cambridge Univ. Press.

Sebestik J. “La dispute de Bolzano avec Kant. (Le fragment d’un dialogue sur la

connaissance mathématique” (forthcoming article).

Siebel M. (1996). Der Begriff der Ableitbarkeit bei Bolzano. (Beiträge der Bolzano-

Forschung 7, ed. E. Morscher and O. Neumaier), St. Augustin.

Solovyov V. (1988). Собрание Сочинения, published posthumously (1911-1914),

Просвещение, Петербург, reprinted in Брокхаус и Ефрон, Т.2, Москва, pp. 441-

479.

Stove D. (1986). The Rationality of Induction, part 2. Oxford: Clarendon Press.

Stumpf C. (1891). Psychologie und Erkenntnistheorie. München. Verlag der königl.

Akademie.

Tarski A. (1936). “On the Concept of Logical Consequence”. Logic, Semantics, andMetamathematics: Papers from 1923 to 1938 (Oxford Univ. Press, 1956, reprint,

Indianapolis: Hackett, 1983).

Textor M. (1996). Bolzanos Propositionalismus, de Gruyter, Berlin.

Textor M. (1999). “Bolzano über die Unvergänglichkeit der Seele”. Bernard Bolzanosgeistiges Erbe für das 21. Jahrhundert, Academia, St. Augustin, pp. 269-290.

Textor M. (2001). “Logically Analytic Propositions A Posteriori?”, History of PhilosophyQuarterly 18.

Textor M. “Caius at Noon or How Bolzano Misunderstood the Copula”, forthcoming.

Ulrici H. (1845). History and Criticism of the Principles of Modern Philosophy. Leipzig.

Van Benthem J. (1984). The Variety of Consequence according to Bolzano. Studia Logica,XLIV, Reidel, Groningen, pp. 389-403.

Von Duhn A. (2001). “Theoretical Laws and Normative Rules: Kant and Bolzano’s Views

on Logic” in: Kant und die Berliner Aufklärung, Akten des IX. Internationalen Kant-

Kongresses, De Gruyter, Berlin, 2001, Bd. 5, pp. 3-12.

Von Duhn A. (2002), “Bolzano’s Account of Justification”, in “The Vienna Circle andLogical Empiricism”, The Vienna Circle Institute Yearbook 10, pp. 21-34

Von Duhn A. (2002). “Bolzano’s Distinction between Confirming (Gewissmachen) and

Justifying (Begründen) Propositions”, Quaderni del Dipartimento di Scienze delleCommunicazione. Univ. di Salerno.

Waitz Th. (1846). Foundation of Psychology. Hamburg and Gotha. Perthes.

Winter E. (1933). Bernard Bolzano und sein Kreis. Dargestellt mit erstmaliger

Heranziehung der Nachlässe Bolzanos und seiner Freunde, Leipzig: Jakob Hegner.

Winter E. (1949). Leben und geistige Entwicklung des Sozialethikers und MathematikersBernard Bolzano 1781-1848. Halle, Niemeyer.

Winter E. (1956). Der böhmische Vormärz in Briefen B. Bolzanos an F. Přihonský.Winter E. (1965). Wissenschaft und Religion im Vormärz. Der Briefwechsel Bernard

Bolzanos mit Michael Josef Fesl 1822-1848. (Berlin: Akademie-Verlag).

Winter E. (1968). Romantismus, Restauration und Frühliberalismus im österreichischenVormärz, Wien, Europa Verlag.

REFERENCES 261

Wolff M. (2001). “Geometrie und Erfahrung. Kant und das Problem objektiver Geltungder Euklidischen Geometrie”. Kant und die Berliner Aufklärung. Akten des IX,Internationalen Kant-Kongresses, vol. 1. Berlin, New York, pp. 209-232.

Zimmermann R. (1847). Leibniz’s Monadology. Vienna, Baumüller and Seidel.

Zimmermann R. (1871). Ueber Kant's mathematisches Vorurtheil und dessen Folgen.

Wien, Kaiserliche Akademie der Wissenschaften.

Zimmermann R. (1888-1889). “Philosophie und Philosophen in Oesterreich” in:

Oesterreichisch-Ungarische Revue. Neue Folge 6 (Vienna, pp. 177-198, pp.259-272).

INDEX

A

a posteriori · 65, 87, 89, 90, 93, 99, 101, 105, 107, 244, 248

footnotes: 116, 123, 128, 472, 591.

a priori · 13, 15, 19, 59, 65, 70, 87, 89, 90, 91, 92, 93, 94, 98, 100, 101, 102,

105, 107, 117, 122, 124, 129, 130, 131, 133, 139, 143, 162, 183, 186, 192,

202, 203, 214, 216, 221, 227, 241, 244, 248, 249

footnotes: 117, 123, 128, 129, 145, 148, 152, 154, 157, 159, 164, 165, 171,

189, 202, 203, 206, 214, 220, 236, 237, 244, 251, 254, 258, 259, 268, 284,

320, 330, 399, 413, 420, 458, 473, 478, 486, 511, 568, 571, 590, 614.

analytic · 59, 61, 70, 96, 98, 99, 101, 128, 130, 143, 155, 168, 169, 205, 248

footnotes: 106, 137, 139, 140, 141, 142, 144, 152, 165, 169, 203, 214, 219,

264, 266, 329, 330, 340, 341, 417, 451, 463, 474, 478, 506, 511.

appearance · 35, 41, 45, 50, 62, 104, 109, 137, 139, 145, 152, 153, 155, 160, 163,

171, 174, 176, 184, 214, 215

footnotes: 127, 160, 171, 251, 254, 257, 258, 261, 281, 284, 290, 293, 306,

312, 321, 324, 360, 365, 371, 372, 388, 389, 397, 398, 399, 422, 432, 474,

484, 491, 546, 602, 606, 607, 612, 613.

C

category · 13, 23, 24, 60, 61, 70, 129, 130, 131, 132, 133, 134, 135, 136, 137,

138, 139, 141, 144, 160, 167, 171, 177, 197

footnotes: 131, 165, 214, 220, 225, 233, 237, 240, 241, 243, 251, 254, 257,

258, 261, 264, 284, 288, 291, 292, 293, 294, 304, 318, 325, 365, 398, 420,

451, 602.

cognition · 15, 16, 35, 39, 45, 47, 51, 60, 65, 72, 75, 88, 89, 90, 94, 104, 105,

128, 129, 135, 139, 143, 152, 153, 198, 201, 203, 204, 210, 218, 221, 224,

229, 241, 242, 244, 248, 249, 250, 255

footnotes: 97, 122, 123, 124, 127, 128, 135, 145, 154, 155, 162, 163, 164, 165,

191, 199, 203, 219, 220, 241, 243, 246, 247, 250, 251, 254, 264, 284, 286,

288, 291, 295, 318, 320, 322, 323, 329, 333, 337, 345, 413, 431, 451, 458,

460, 468, 469, 478, 506, 511, 526, 533, 559, 568, 571, 578, 580, 581, 582,

590, 602, 612.

INDEX 263

D

definition · 13, 17, 23, 34, 35, 39, 44, 45, 51, 59, 63, 76, 79, 89, 90, 94, 96, 100,

105, 111, 115, 125, 131, 139, 141, 144, 155, 157, 181, 185, 204, 205, 244,

249, 253

footnotes: 120, 148, 177, 203, 208, 264, 324, 461-463, 470, 471, 473, 557,

576.

E

experience · 13, 17, 18, 19, 35, 40, 45, 52, 54, 61, 62, 88, 89, 90, 94, 102, 122,

133, 138, 143, 144, 145, 150, 159, 176, 183, 188, 191, 192, 193, 194, 196,

198, 209, 222, 226, 235, 242, 244, 245, 248, 250, 255

footnotes: 124, 127, 152, 154, 157, 165, 220, 225, 236, 237, 243, 250, 251,

254, 255, 268, 281, 284, 288, 290, 291, 294, 306, 307, 312, 318, 322, 325,

330, 336, 337, 343, 345, 346, 368, 398, 399, 412, 413, 420, 421, 432, 435,

451, 452, 455, 460, 474, 478, 484, 491, 514, 575, 585, 590, 591, 597, 602,

606, 607, 612, 614.

explication · 17, 34, 37, 40, 61, 70, 205, 213

footnotes: 470.

I

inference · 19, 37, 40, 47, 48, 49, 52, 53, 62, 64, 100, 101, 163, 165, 166, 185,

187, 189, 193, 205, 218, 233

footnotes: 109, 152, 312, 313, 320, 325, 326, 329, 330, 333, 337, 343, 346,

451, 478, 496.

intuition · 15, 17, 18, 20, 34, 40, 43, 53, 59, 60, 67, 81, 90, 94, 99, 100, 101, 103,

104, 105, 109, 115, 117, 118, 122, 124, 125, 128, 130, 131, 134, 135, 137,

141, 142, 143, 144, 145, 150, 151, 153, 168, 175, 183, 203, 204, 209, 211,

213, 218, 243, 244, 245, 246, 249, 254

footnotes: 109, 111, 112, 113, 114, 115, 128, 141, 146, 147, 152, 156, 157,

159, 160, 161, 162, 163, 164, 165, 166, 168, 169, 171, 189, 191, 202, 203,

204, 206, 207, 208, 211, 214, 237, 238, 240, 241, 244, 245, 246, 247, 249,

250, 251, 254, 258, 259, 260, 261, 264, 280, 281, 284, 289, 290, 293, 294,

307, 322, 323, 340, 375, 384, 393, 398, 399, 432, 458, 460, 464, 468, 469,

474, 478, 479, 480, 506, 559, 580, 586, 591, 595, 602, 607, 612.

INDEX264

J

judgment · 13, 15, 16, 27, 28, 36, 37, 48, 53, 57, 61, 73, 75, 82, 90, 91, 96, 99,

100, 101, 103, 122, 126, 129, 130, 132, 133, 135, 140, 143, 144, 150, 153,

154, 156, 158, 160, 187, 188, 189, 210, 217, 238, 251

footnotes: 97, 101, 106, 116, 123, 127, 129, 130, 131, 135, 137, 138, 139, 140,

141, 144, 145, 146, 147, 148, 150, 165, 166, 169, 202, 203, 214, 225, 227,

230, 234, 239, 244, 245, 247, 250, 255, 266, 268, 284, 292, 296, 306, 312,

313, 319, 325, 329, 333, 335, 340, 398, 399, 429, 451, 474, 478, 480, 506,

553, 559, 590, 591, 592, 593, 595, 602, 603, 607, 614.

P

principle · 16, 36, 41, 46, 47, 55, 61, 62, 69, 94, 114, 137, 143, 144, 145, 148,

151, 155, 159, 163, 183, 192, 194, 195, 203, 213, 225, 228, 231, 245, 249

footnotes: 127, 128, 141, 144, 152, 157, 163, 164, 171, 214, 219, 264, 265,

266, 280, 284, 285, 286, 288, 290, 306, 312, 320, 330, 333, 399, 420, 427,

431, 432, 435, 451, 455, 460, 474, 478, 491, 501, 511, 523, 525, 526, 527,

533, 537, 538, 542, 578, 597.

R

reason · 19, 27, 37, 42, 47, 49, 53, 55, 62, 63, 64, 65, 69, 73, 94, 98, 99, 100,

101, 118, 128, 157, 159, 163, 167, 169, 170, 174, 175, 176, 177, 183, 186,

190, 193, 194, 195, 196, 197, 198, 201, 213, 214, 215, 216, 217, 218, 221,

222, 224, 226, 229, 232, 233, 235, 239, 240, 243, 245, 246, 250, 251

footnotes: 106, 123, 127, 128, 152, 160, 162, 169, 182, 208, 211, 219, 247,

261, 285, 286, 287, 312, 313, 318, 320, 322, 324, 325, 329, 330, 333, 338,

345, 346, 348, 365, 372, 389, 397, 399, 400, 403, 415, 431, 432, 434, 435,

451, 460, 473, 474, 478, 482, 483, 485, 491, 501, 506, 511, 516, 520, 523,

526, 529, 531, 532, 540, 545, 546, 549, 559, 564, 565, 567, 568, 578, 582,

590, 595, 597, 606, 607.

S

sensation · 18, 45, 56, 104, 111, 144, 145, 148, 229

footnotes: 113, 156, 157, 160, 161, 162, 165, 170, 283, 284, 322, 368, 607,

609, 610.

INDEX 265

sensibility · 47, 48, 56, 59, 60, 103, 104, 105, 109, 137, 151, 156, 157, 214, 246,

251

footnotes: 155, 159, 162, 203, 237, 245, 249, 259, 292, 294, 298, 306, 397,

424, 597, 606.

synthetic · 13, 15, 19, 59, 61, 70, 96, 98, 100, 101, 102, 118, 122, 126, 128, 130,

134, 139, 143, 144, 150, 159, 168, 169, 172, 187, 190, 192, 203, 210, 218,

221, 224, 233, 234, 246, 248, 249, 254

footnotes: 106, 123, 127, 137, 139, 140, 141, 142, 143, 144, 145, 147, 148,

150, 152, 157, 165, 202, 203, 238, 244, 254, 255, 266, 268, 291, 330, 340,

375, 399, 413, 451, 478, 511, 591, 592, 614.

U

understanding · 13, 16, 17, 49, 57, 59, 60, 61, 90, 100, 101, 103, 104, 110, 121,

127, 128, 129, 130, 133, 137, 140, 141, 144, 152, 153, 154, 156, 157, 186,

191, 196, 241, 245, 246, 251

footnotes: 115, 123, 127, 152, 155, 162, 163, 165, 166, 171, 204, 214, 219,

225, 233, 234, 241, 247, 249, 254, 256, 258, 260, 281, 284, 286, 293, 298,

306, 308, 312, 313, 318, 320, 322, 325, 329, 330, 333, 365, 368, 371, 372,

389, 391, 399, 411, 412, 435, 451, 478, 506, 511, 590, 591, 597, 602, 606,

610, 613.

New Anti-Kant by František Přihonský

Documents